CCTV Surveillance Analog and Digital Video Practices and

CCTV Surveillance Analog and Digital Video Practices and
CCTV Surveillance
This page intentionally left blank
CCTV Surveillance
Analog and Digital Video
Practices and Technology
Second Edition
by Herman Kruegle
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD
PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Butterworth–Heinemann is an imprint of Elsevier
Senior Acquisitions Editor:
Assistant Editor:
Marketing Manager:
Project Manager:
Cover Designer:
Compositor:
Cover Printer:
Text Printer/Binder:
Mark Listewnik
Kelly Weaver
Christian Nolin
Jeff Freeland
Eric DeCicco
Integra Software Services Pvt. Ltd.
Phoenix Color Corp.
The Maple-Vail Book Manufacturing Group
Elsevier Butterworth–Heinemann
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
Linacre House, Jordan Hill, Oxford OX2 8DP, UK
Copyright © 2007, Elsevier, Inc. All rights reserved.
Exceptions: Copyright for many of the photos is not held by the publisher. Please see the
Photo Credits section, on the next page, for copyright information on these photos.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in
any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333,
e-mail: [email protected] You may also complete your request on-line
via the Elsevier homepage (http://elsevier.com), by selecting “Support & Contact”
then “Copyright and Permission” and then “Obtaining Permissions.”
Recognizing the importance of preserving what has been written, Elsevier prints its books on
acid-free paper whenever possible.
Library of Congress Cataloging-in-Publication Data
Kruegle, Herman.
CCTV surveillance : analog and digital video practices and technology / Herman Kruegle—2nd ed.
p. cm.
ISBN-13: 978-0-7506-7768-4 (casebound : alk. paper)
ISBN-10: 0-7506-7768-6 (casebound : alk. paper) 1. Closed-circuit television—Design and construction.
2. Television in security systems. I. Title.
TK6680.K78 2005
621.389’28—dc22
2005022280
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
ISBN-13: 978-0-7506-7768-4
ISBN-10: 0-7506-7768-6
For information on all Butterworth–Heinemann publications
visit our Web site at www.books.elsevier.com
Printed in the United States of America
06 07 08 09 10 11 10 9 8 7 6 5 4 3 2 1
Working together to grow
libraries in developing countries
www.elsevier.com | www.bookaid.org | www.sabre.org
Photo Credits
The publisher and author would like to thank the listed manufacturers for the photographs used in the figures.
Accele Electronics
Allan Broadband
American Dynamics
Avida
Axis Communications
CBC America
Canon USA
Casio USA
Cohu, Inc.
Controp USA
COP-USA
Dell Star
D-Link
Digispec
FM Systems
Global Solar
Gossen
Greenlee
Gyrozoom
Hitachi
Honeywell Security
ICU
IFS/GE Security
Ikegami Electronics (U.S.A. Inc.)
Integral Tech
Intellicom
International Fiber Systems
Ipix
Instrumentation Tech Systems
Keithley
Leader Instruments
Lowering Systems
Mace
Mannix
Marshall
Mitsubishi
8-9A, 8-9B
25-14A
12-1, 17-1E
2-7C, 2-7E, 2-7G, 2-7H, 2-16A, 2-16B, 2-17A, 2-17B, 2-17C, 2-17D, 2-17E,
2-17F, 4-18A, 4-27C, 4-27D, 4-27E, 4-30, 4-33A, 4-33B, 4-36, 4-37, 4-38, 4-40,
15-2A, 15-2C, 15-8A, 15-8C, 15-10B, 15-12, 15-15A, 15-15B, 16-7, 18-5A,
18-6A, 18-6B, 18-7, 18-10, 18-11A, 18-11B, 18-14A, 18-14B, 18-20A, 18-23D,
18-24, 19-22A, 19-22B, 21-2A, 21-2B, 21-4A, 21-4B, 21-4C, 22-4A, 22-4C, 22-5,
22-10B, 22-10C, 22-23A, 22-23B, 22-25, 22-26, 22-27
5-14B, 7-28A, 7-34A, 7-34B, 7-35A, 7-35B
15-9A
4-14A
7-36A
2-10A, 2-10F
17-24
18-19B
6-35A
7-36B
13-8A
25-13B
23-11A, 23-11C
25-15A
25-21A, 25-21B
4-14B
2-26D, 17-22A
9-12C, 15-2D, 15-7D, 15-10D, 15-13, 22-10B
13-8C, 13-8D
6-28, 6-30
2-10C, 4-38, 8-5A
7-36C
25-22A, 25-22B
6-28, 6-30
2-15B
16-6A, 16-6B
25-14B
25-1A, 25-2A, 25-2B, 25-2C, 25-6A, 25-10, 25-11
14-8C, 14-8D
15-10C
25-15B
8-16A
2-28A, 10-1
NVT
Omniscope
Panasonic Security Systems
Parkut
Pelco
Pentax
Radiant
Rainbow
RF-Links
Remote Video Surveillance
Sanyo Security Products
Sagebrush Technology
Selectronic
Semco
SOHOware
Sony Electronics
Smarter Systems
Tektronix
Thorlabs
Trango
Uni-Solar Ovonic
Vicon
Videolarm
Watec
Winsted
6-9A, 6-9B
2-15A
Cover image (bottom), 2-10B, 2-26A, 2-26C, 2-27B, 2-27C, 5-14A, 8-9D, 14-4B,
14-5B, 14-6A, 15-2B, 17-2, 18-20B, 20-4A, 20-4B, 20-5B
22-26, 22-27
14-5C, 15-7C, 15-14B, 15-17, 17-1A, 17-1C, 17-11B
2-7A, 2-14, 4-12A
13-8B
4-12B, 4-19, 4-22A, 4-22B
18-25B
9-12B
Cover image (middle right), 2-27A, 5-14C, 8-5B, 8-9C, 9-12A, 14-1A, 15-6C,
15-6D, 15-10A, 17-1D
17-14
8-10A
6-38C
7-10
4-26, 4-31, 5-22, 7-28B, 14-1B, 14-4A, 17-22B
23-13
25-1B, 25-1C, 25-6B, 25-13A
25-17
6-35B, 6-35C
23-11B
2-26B, 2-30A, 2-30C, 14-3, 14-5A, 14-5D, 14-6B, 15-1A, 15-1B, 15-5, 15-6A,
15-6B, 15-9B, 15-11, 15-14A, 15-14C, 15-19B, 17-1B, 17-10A, 17-11A
2-29E, 14-7A, 14-7B, 14-7C, 14-8A, 14-8B, 15-7A, 15-8B, 15-8D, 15-9C, 15-14D,
15-19A, 17-13, 17-15A, 17-15B, 22-4B
18-19A, 18-19C
20-2A, 20-2B, 20-3A, 20-3B
For Carol
This page intentionally left blank
Contents
xi
xiii
xv
Foreword
Preface
Acknowledgments
Part 1
Chapter 1
Chapter 2
Video’s Critical Role in the Security Plan
Video Technology Overview
Part 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
Chapter 11
Chapter 12
Chapter 13
Chapter 14
Chapter 15
Chapter 16
Chapter 17
Chapter 18
Chapter 19
Chapter 20
Chapter 21
Chapter 22
Chapter 23
Chapter 24
Chapter 25
Chapter 26
Chapter 27
Chapter 28
Natural and Artificial Lighting
Lenses and Optics
Cameras—Analog, Digital, and Internet
Analog Video, Voice, and Control Signal Transmission
Digital Transmission—Video, Communications, Control
Analog Monitors and Digital Displays
Analog, Digital Video Recorders
Hard Copy Video Printers
Video Switchers
Quads and Multiplexers
Video Motion Detectors
Dome Cameras
Integrated Cameras, Camera Housings, and Accessories
Electronic Video Image Splitting, Reversal, and Annotation
Camera Pan/Tilt Mechanisms
Covert Video Surveillance
Low-Light-Level Cameras, Thermal Infrared Imagers
Control Room/Console Design
Rapid Deployment Video Systems
Applications and Solutions—Sample Scenarios
System Power Sources
Video-Security Systems Integration
Video System Test Equipment
Video Check List
Education, Standards, Certification
New Video Technology
Glossary
Bibliography
Index
1
13
47
71
109
145
199
251
275
305
321
341
353
373
387
405
415
445
469
497
507
513
553
577
583
601
605
609
615
639
643
ix
This page intentionally left blank
Foreword
A few years ago I had the privilege of addressing a Congressional Subcommittee on Technology and Procurement
Policy, chaired by Congressman Tom Davis. In addition
to examining GSA’s efforts to secure federal buildings,
the Subcommittee was interested in hearing and learning about new physical security technology. When I leaf
through the pages of this book, I again realize the enormity of the task undertaken by the Subcommittee, the
necessity for doing so, and the importance of this type of
information to not only security professionals, but now to
IT professionals as well.
Closed circuit television (CCTV) and other related video
security and surveillance technology has advanced further and faster in the period from 2001 to 2005 than in
any prior comparable time period. IP cameras, mapping,
servers, platforms, LANs, WANs, and VPNs, wireless, digital migration, algorithms, etc. are all converging along
with other related security system technologies such as
access control, life safety, intrusion alarms, etc. with the
intent to configure fully integrated systems. This is the
new direction for the security industry as digital technology has become pervasive across all product lines, opening
the door to more software-oriented control platforms on
the enterprise level.
So who is the better person to chronicle, explain, and
put these terms and technology into perspective than
Herman Kruegle, one of the industry’s foremost experts
on video surveillance and related technologies. I have
had the privilege of knowing and working with Herman
for many years. He is a consummate professional who
has the innate ability to explain the technical aspects of
this emerging technology in a manner we can all understand and put into practice. Herman’s first book, CCTV
Surveillance – Video Practices and Technology, is considered,
by most of us in the industry, to be the bible of CCTV, and
I fully expect this revised edition will rise to even greater
popularity.
In the pages following, readers will find concise and
intelligent descriptions of the analog and digital video
practices and technology we have all grown up with. But
more important, Herman has included, in this revised edition, his explanation of the newest audio/video information technology (AV/IT) developments, products utilizing
the technology and applications for same. Security professionals, system integrators, architects and engineers, IT
managers, or end users who are looking for a resource
to help them navigate this complex field of IP Video
Security will not be disappointed. The material is well
researched and thoughtfully laid out to help insure the
reader’s understanding and to hopefully allow them to go
on to designing, installing, and using digital video surveillance to its fullest capacity.
Frank Abram
xi
This page intentionally left blank
Preface
Following the same philosophy contained in the first edition, the second edition is written for and contains information valuable to the end-user as well as the technical
practitioner. Each chapter begins with an overview and
then presents equipment available with their characteristics, features, and application.
The first edition of CCTV Surveillance in 1995 asked
the question “why write a CCTV surveillance book?”. At
that time, analog CCTV had progressed from a vacuum
tube to a solid state technology that provided reliable,
longlife small cameras produced at prices affordable for
most security applications.
A decade later, significant advances have been made in
camera sensors, computers, and digital transmission technology to warrant a complete review of CCTV’s role in
the security industry. The migration from legacy analog
components to digital technology and the emergence of
the Internet have accelerated the utilization of Internet
protocol (IP) video and remote monitoring in security.
The internet has permitted the widespread interconnection of other technologies including intrusion and fire and
intrusion alarm systems, access control, and other communications and control.
The ease of interconnection afforded by digital transmission of video and other pertinent security data anywhere in a facility, local environment or globally, engenders a new meaning to video transmission and remote
viewing.
The explosion of high-capacity magnetic disk, solid
state, and optical data storage memories has permitted
the generation of new products including digital video
recorders (DVR) and data compression algorithms to compress and store video images and replace the time-honored
magnetic video cassette recorder (VCR).
In this second edition of CCTV Surveillance, I have
attempted to add these new technologies to the “nonchanging” basic technologies covered in the first edition.
Physics does not change—only the technology and products do.
This new revised edition of CCTV Surveillance includes
the new digital video technology and contains eight new
chapters:
Chapter 7
Chapter
Chapter
Chapter
Chapter
Chapter
Chapter
Chapter
10
12
14
20
21
24
25
Digital Transmission, Video,
Communications and Control
Hard Copy Video Printers
Quads and Multiplexers
Dome Cameras
Control Room/Console Design
Rapid Deployment Video Systems
Video-Security Systems Integration
Video System Test Equipment
Chapter 7—Wired and wireless digital transmission represents possibly the most significant technology advancement in the video security industry. It makes use of the
Internet and intranets for remote video, data, and audio
communication over existing hard wire communication
links. Chapter 7 includes an analysis of digital wireless
video transmission using the family of 802.11x protocol
spread spectrum technology (SST). Prior to 1995–98 the
Internet was not available for commercial use and remote
video monitoring and control was accomplished primarily over existing telephone lines or expensive satellite
links with limited functionality. Ease of installation, camera addressing, and identification using IP cameras has
opened a new vista in video transmission and remote monitoring.
Chapter 10—This chapter describes the new technological advances made in hard-copy printers that improve the
quality and reduce the cost of monochrome and color
video printouts. The advances in ink-jet and laser printer
technologies using inexpensive, large solid state memories
and high resolution linear CCD imagers have been driven
by the consumer and business markets, and have given the
security industry access to low-cost, color, hard copy prints
rivaling photographic resolution and quality.
Chapter 12—While available in 1995, multiplexers have
taken on new importance because of the significant
xiii
xiv
Preface
increase in the number of cameras used in a typical security installation and their ability to be integrated into DVRs
that were not available five years ago.
Chapter 14—Dome cameras are now everywhere in
security systems. In 1995 they were used primarily in
selected locations: casinos, department stores, supermarkets, malls, and in outdoor parking lot applications. The
public at large has accepted their presence almost everywhere. Domes are easy to install and can be small and
aesthetic. Dome cameras are adjustable in pointing direction (manual or motorized, pan and tilt), and many have
motorized zoom lenses to change the camera field of view
(FOV). The use of small dome cameras has exploded
because of significant cost reduction and sophistication of
pointing and zooming capabilities. Fast pan/tilt camera
modules with remote control via analog or digital communications over two-wire or wireless communication links
are reasons for their popularity.
Chapter 20—Consoles and Control Rooms have become
more complex and require more design attention for their
successful implementation. This chapter analyzes the console and security control room with regard to lighting,
monitor locations, operator control placement, and the
other human factors required for guard efficiency and
comfort.
Chapter 21—There has always been a requirement for
a transportable Rapid Deployment Security (RDS) systems
having video and alarm intrusion equipment for protecting personnel and assets. The Post-911 era with real terror
threats has initiated the need for RDS equipment to protect military, government, business, and other personnel
on travel. The majority of these systems consist of alarm
intrusion and analog or digital video viewing system. These
RDS systems are carried from one location to another and
deployed quickly to set up an alarm perimeter and realtime video monitoring and recording. Analog or digital
transmission allows local or remote monitoring. After use,
the RDS equipment is disassembled and stored in its carrying case, ready for another deployment. The much smaller
size of the video and alarm equipment has accelerated its
use and acceptance.
Chapter 22—The Video Applications chapter has been
updated and expanded to include digital video applications including the combination of legacy analog and IP
cameras. One video monitoring application uses on-site
local networks and a second application uses the Internet and IP cameras, signal routers, and servers for remote
site video monitoring. Security applications require complete integration of communication, video, alarm, access
control, and fire to provide monitoring by the local security force, and corporate executives at a local or remote
site(s). The integration of these security functions provides the safety and security necessary to protect personnel
and assets at any facility.
Chapter 25—Installation and maintenance of video
equipment requires the use of video and computer test
equipment. Prior to the widespread use of digital technology in security systems, a limited range of test equipment
was used. Now with the many computer interfaces and
Internet protocols and connection to the Internet, more
sophisticated test equipment and some knowledge of software and computer programming is necessary. Parameters
to be tested and monitored include: (a) video signal level
and quality; (b) control data signals for pan, tilt, zoom,
focus; and (c) digital signal protocols for multiplexers, IP
cameras, signal routers and servers, DVRs, etc.
Acknowledgments
Over the years I have had opportunities to speak with
many individuals who provided technical insight in video
technology and electro-optics. I particularly appreciate the
discussions with Stanley Dolin and Lee Gallagher, on the
subjects of optics, the physics of lighting, lenses, and optical sensors. I found very helpful the technical discussions
on cameras with Frank Abram, Sanyo Security Products,
and Victor Houk. I thank Dr. Gerald Herskowitz, Stevens
Institute of Technology for contributing to the fiber-optic
section in Chapter 6 and reviewing other sections on
video transmission. I thank Robert Wimmer and Fredrick
Nilsson for their excellent technical articles in security
journals, company publications, as well as technical seminars on many aspects of video security. Thanks to Charlie Pierce for his interest in my book over the years and
enthusiasm and excellence in presenting stimulating educational video seminars. Eric Kruegle, Avida Inc., contributed his expertise on various aspects of digital video. In
particular I appreciate his help in wired and wireless video
transmission, compression, and encryption in Chapter 7.
Eric was also instrumental in keeping my computer alive,
and I thank him for rescuing me late at night from missing
files and software surprises.
I acknowledge the initial encouragement of Kevin Kopp
and editorial advice of Greg Franklin at Butterworth (now
Elsevier) during the formative stages of the first edition of
CCTV Surveillance in 1995. I thank all staff at Elsevier for
bringing out this second edition successfully: Pam Chester
for her assistance in the formulation of this edition, Mark
Listewnik for his constant encouragement, professional
suggestions, and diligence in bringing this large project
to a successful conclusion, Jeff Freeland for providing the
meticulous final editing and effort in completing this large
endeavor.
I gratefully acknowledge the dedication, patience, and
skill of my wife, Carol, in assisting in the preparation of
this book.
I would like to thank the manufacturers for the use of
the many photographs that illustrate the components used
in video security applications. Each of them contribute to
the education of the security professional and assist the
consultant, systems integrator, and end user in designing
and implementing the best security system possible.
xv
This page intentionally left blank
PART I
Chapter 1
Video’s Critical Role in the Security Plan
CONTENTS
1.1 Protection of Assets
1.1.1 Overview
1.1.2 Background
1.2 The Role of Video in Asset Protection
1.2.1 Video as Part of the Emergency and
Disaster Plan
1.2.1.1 Protecting Life and Minimizing
Injury
1.2.1.2 Reducing Exposure of Physical Assets
and Optimizing Loss Control
1.2.1.3 Restoring Normal Operations
Quickly
1.2.1.4 Documenting an Emergency
1.2.1.5 Emergency Shutdown and
Restoration
1.2.1.6 Testing the Plan
1.2.1.7 Standby Power and Communications
1.2.2 Security Investigations
1.2.3 Safety
1.2.4 The Role of the Guard
1.2.5 Employee Training and Education
1.3 Synergy through Integration
1.3.1 Integrated Functions
1.3.2 System Hardware
1.4 Video’s Role and Its Applications
1.4.1 Video System Solutions
1.4.2 Overt vs. Covert Video
1.4.3 Security Surveillance Applications
1.4.4 Safety Applications
1.4.5 Video Access Control
1.5 The Bottom Line
1.1 PROTECTION OF ASSETS
The protection of personnel and assets is a management function. Three key factors governing the planning
of an assets protection program are: (1) an adequate
plan designed to prevent losses from occurring, (2) adequate countermeasures to limit unpreventable losses, and
(3) support of the protection plan by top management.
1.1.1 Overview
Most situations today require a complete safety/security
plan. The plan should contain requirements for intrusion
detection, video assessment, fire detection, access control,
and full two-way communication. Critical functions and
locations must be monitored using wired and wireless
backup communications.
The most significant driving force behind the explosion
in the use of closed-circuit television (CCTV) has been the
worldwide increase in theft and terrorism and the commensurate concern and need to protect personnel and
assets. The terrorist attack on September 11, 2001, brought
about a quantum jump and a complete reevaluation of the
personnel and asset security requirements to safe-guard a
facility. To meet this new threat, video security has taken
on the lead role in protecting personnel and assets. Today
every state-of-the-art security system must include video as
a key component to provide the “remote eyes” for security,
fire, and safety.
The fateful day of September 11, 2001, has dramatized
the importance of reliable communications and remote
visualization of images via remote video cameras. Many
lives were saved (and lost) as a consequence of the voice,
video, alarm, and fire equipment in place and in use at the
time of the fateful attack on the World Trade Center in
New York. The availability of operational wired and wireless
two-way communication between command and control
headquarters and responders (police, fire, emergency)
played a crucial role in life and death. The availability
(or absence) at command posts of real-time video images
1
2
CCTV Surveillance
at crucial locations in the Twin Towers during the attack
and evacuation contributed to the action taken by command personnel during the tragedy. The use (or absence)
of wireless transmission from the remote video cameras in
the Twin Towers clearly had an impact on the number of
survivors and casualties.
During the 1990s, video components (cameras,
recorders, monitors, etc.) technology matured from the
legacy analog to a digital imaging technology and became
compatible with computers and now forms an essential
part of the security solution. In the late 1990s, digital cameras were introduced into the consumer market,
thereby significantly reducing price and as a result found
widespread use in the security industry. Simultaneously,
powerful microprocessors, large hard disk computer memory storage, and random access memory (RAM) became
available from the personal computer/laptop industry,
thereby providing the computing power necessary to control, view, record, and play back digital CCTV cameras in
the security system.
The home run came with the availability and explosive
acceptance and use of the Internet (and intranet) as a
new means of long distance two-way communication of
voice, data, and most importantly video. For over a decade
the long distance transmission of video was limited to slow
telephone transmission of video images—snap-shots (slowscan video). The use of dedicated high speed (expensive) land lines or expensive satellite communications was
limited to government and large-clientele users. Now the
Internet provides near-live (near real-time) video transmission communications over an inexpensive, easily accessible
worldwide transmission network.
The application and integration of video into safety and
security systems has come of age as a reliable, cost-effective
means for assessing and responding to terrorist attacks
and other life-threatening situations. Video is an effective
means for deterring crimes and protecting assets and for
apprehending and prosecuting offenders.
Security personnel today have the responsibility for multifaceted security and safety systems in which video often
plays the key role. With today’s increasing labor costs and
the need for each security officer to provide more functionality, video more than ever before is earning its place
as a cost-effective means for improving security and safety
while reducing security budgets.
Loss of assets and time due to theft is a growing cancer on our society that eats away at the profits of every
organization or business, be it government, retail, service,
or manufacturing. The size of the organization makes no
difference to the thief. The larger the organization, the
more the theft occurs and the greater the opportunity for
losses. The more valuable the product, the greater the
temptation for a thief to steal it. A properly designed and
applied video system can be an extremely profitable investment for an institution to cut losses. The prime objective
of the video system should not be the apprehension of
thieves but rather the deterrence of crime through security. A successful thief needs privacy—a video system can
deny that privacy.
As a security by-product, video has emerged as an effective training tool for managers and security personnel.
Every installation/establishment should have a security
plan in place prior to an incident. Video-based training
is easy to implement using the abundance of inexpensive camcorders and playback equipment available and
the commercial video production training video services
available. The use of training videos results in standardized procedures and improved employee efficiency and
productivity.
The public at large has accepted the use of video systems
in most public facilities. Video is being applied to reduce
asset losses and increase corporate profits and bottom
line. Many case histories show that after the installation of
video, shoplifting and employee thefts drop sharply. The
number of thefts cannot be counted exactly but shrinkage
can be measured. It has been shown that video is an effective psychological deterrent to crime and an effective tool
for criminal prosecution.
Theft is not only the unauthorized removal of valuable
property but also the removal of information, such as computer software, CDs, magnetic tape and disks, optical disks,
microfilm, and hard copy. Video surveillance systems provide a means for successfully deterring such thievery and/or
detecting or apprehending offenders. The use of video prevents the destruction of property, vandalizing buildings,
defacing elevator interiors, painting graffiti on art objects
and facilities, stealing computers, and demolishing furniture or other valuable equipment. Video offers the greatest
potential benefit when integrated with other sensing systems and used to view remote areas. Video provides the
“eyes” for many security devices and functions such as:
(1) fire sensors: smoke detector alarms, (2) watching for
presence (or absence) of personnel in an area, (3) evacuation of personnel—determining route for evacuation,
access (emergency or intruder) to determine response,
respond, and monitor response. When combined with fire
and smoke detectors, CCTV cameras in inaccessible areas
can be used to give advance warning of a fire.
Video is the critical link in the overall security of a
facility but organizations must develop a complete security
plan rather than adopt piecemeal protection measures.
To optimize use of video technology, the practitioner and
end user must understand all of its aspects—from light
sources to video monitors and recorders. The capabilities
and limitations of video during daytime and nighttime
operation must also be understood.
1.1.2 Background
Throughout history, humans have valued their own life
and the lives of their loved ones above all else. Next
Video’s Critical Role in the Security Plan
in value has been their property. Over the centuries
many techniques have been developed to protect property against invaders or aggressors threatening to take or
destroy it.
In the past as in the present, manufacturing, industrial,
and government organizations have hired “watchmen” to
protect their facilities. These private security personnel
wearing uniforms and using equipment much like the
police do are hired to prevent crime and bodily harm,
and deter or prevent theft on the premises. The very early
guard companies were Pinkerton’s and Burns. Contract
protection organizations were hired to safeguard their
employees and assets in emergency and personal threat
situations.
A significant increase in guard use came with the start of
World War II. Many guards were employed to secure industrial work sites manufacturing military equipment and
doing classified work, and to guard government facilities.
Private corporations obtained such protection through
contract agencies to guard classified facilities and work.
In the early 1960s, as electronic technology advanced,
alarm systems and video were introduced. Radio Corporation of America (RCA), Motorola, and General Electric
were the pioneering companies that began manufacturing
vacuum-tube television cameras for the security industry.
The use of video cameras during the 1960s and 1970s grew
rapidly because of increased reliability, lower cost, and
technological improvements in the tube-type camera technology. In the 1980s growth continued at a more modest
level with further improvements in functions and availability of other accessories for video security systems.
The most significant advance in video technology during the 1980s was the invention and introduction of the
solid-state video camera. By the early 1990s the solid-state
camera using the charged coupled device (CCD) image
sensor was the choice for new security installations and
was rapidly replacing the tube cameras. In the past, the
camera—in particular, the vidicon tube sensor—was the
critical component in the video system. The camera determined the overall performance and quality of visual intelligence obtainable from the security system. The vidicon
tube was the weakest link in the system and was subject to degradation with age and usage. The complexity
and variability of the image tube and its analog electrical nature made it less reliable than the other solid-state
components. Performance varied considerably between
different camera models and camera manufacturers, and
as a function of temperature and age. By contrast, the
solid-state CCD sensor and newer metal oxide semiconductor (MOS) and complimentary MOS (CMOS) sensor
cameras have long life and are stable over all operating
conditions. Another factor in the explosive use of video
in security systems has been the rapid improvement in
equipment capability at affordable prices. This has been
the result of the widespread use of solid-state camcorders
3
by consumers (lower manufacturing costs), and the availability of low-cost video cassette recorders (VCRs), digital
video recorders (DVRs), and personal computer (PC)based equipment.
The 1990s saw the integration of computer technology
with video security technology. All components were solid
state. Digital video technology needed large-scale digital
memories to manipulate and store video images and the
computer industry had them. To achieve satisfactory video
image transmission and storage, the video signal had to be
“compressed” to transmit it over the existing narrowband
phone line networks. The video-computer industry already
had compression for broadcast, industrial, and government requirements. The video industry needed a fast and
low-cost means to transmit the video images to remote
locations and the US government’s Defense Advanced
Research Projects Agency (DARPA) had already developed the Internet, the predecessor of the World Wide
Web (WWW). The Internet (and intranet) communications channels and the WWW now provide this extraordinary worldwide ability to transmit and receive video and
audio, and communicate and control data anywhere.
1.2 THE ROLE OF VIDEO IN ASSET PROTECTION
Video provides multiple functions in the overall security
plan. It provides the function of asset protection by monitoring location of assets and activity in their location. It is
used to detect unwanted entry into a facility beginning at
a perimeter location and following an unauthorized person throughout a facility. Figure 1-1 shows a typical single
site video system using either legacy analog or digital, or
a combination of both technologies.
In a perimeter protection role, video is used with
intrusion-detection alarm devices as well as video motion
detection to alert the guard at the security console that
an intrusion has occurred. If an intrusion occurs, multiple CCTV cameras located throughout the facility follow
the intruder so that there is a proper response by guard
personnel or designated employees. Management must
determine whether specific guard reaction is required and
what the response will be.
Video monitoring allows the guard to be more effective, but it also improves security by permitting the camera
scene to be transmitted to other control centers or personnel. The video image can be documented with a VCR,
DVR, and/or printed out on a hard copy video printer.
The video system for the multiple site application is best
implemented using a combination of analog/digital or an
all-digital solution (Figure 1-2).
Local site installations already using analog video cameras, monitors, etc. can be retained and integrated with
new digital Internet Protocal (IP) cameras, local area networks (LANs), intranets, and the Internet to facilitate
remote site video monitoring. The digital transmission
4
CCTV Surveillance
PERIMETER
PARKING LOT
SURVEILLANCE
SECURITY ROOM
CCTV MONITORS/RECORDERS
AUDIO COMMUNICATIONS
COMMAND AND CONTROL
LOADING DOCK
SURVEILLANCE
INTRUDER
PATH
LOBBY
SURVEILLANCE
G
FENCE LINE
FACILITY
ENTRANCE
FIGURE 1-1
PERIMETER
PARKING LOT
SURVEILLANCE
Single site video security system
network provides two-way communications of audio and
controls and excellent video image transmission to remote
sites. The digital signals can be encrypted to prevent
eavesdropping by unauthorized outside personnel. Using
a digital signal backbone allows adding additional cameras to the network or changing their configuration in the
system.
In the relatively short history of CCTV and video there
have been great innovations in the permanent recording of video images. These new technologies have been
brought about by the consumer demand for video camcorders, the television broadcast industry, and government
requirements for military and aerospace hardware and
software. One result of these requirements was the development of the VCR and DVR. The ability to record video
images provided the video security industry with a new
dimension, i.e. going beyond real-time camera surveillance. The availability of VCR and DVR technology resulting from the consumer market has made possible the
excellent time-lapse VCRs and large storage PC-based DVR
systems. These technologies provide permanent documentation of the video images in analog (magnetic tape) and
digital (solid state and hard disk drive) storage media.
The use of time-lapse recorders, computer hard disks and
video printers give management the tools to present hard
evidence for criminal prosecution. This ability to provide
a permanent record of evidence is of prime importance
to personnel responsible for providing security.
Prior to the mid-1990s the CCTV security industry primarily used monochrome solid-state cameras. In the 1990s
the widespread use of color camcorders in the video consumer market accelerated the availability of these reliable,
stable, long-life cameras for the security industry. While
monochrome cameras are still specified in low light level
(LLL) and nighttime security applications, color is now
the norm in most security applications. The increased
sensitivity and resolution of color cameras and the significant decrease in cost of color cameras have resulted in
their widespread use. Many monochrome cameras being
used for LLL applications are being augmented with active
infrared (IR) illuminators. Also coming into use is a new
generation of passive monochrome thermal IR imaging
cameras that detect the differences in temperature of objects
in the scene, compared to the scene background. These
cameras operate in total darkness. There has also been an
explosion in the use of covert video surveillance through
the use of small, inexpensive color cameras.
The development of smaller solid-state cameras has
resulted in a decrease in the size of ancillary video equipment. Camera lenses, dome cameras, housings, pan/tilt
Video’s Critical Role in the Security Plan
5
SITE 2
SITE 1
ANALOG
CAMERA(S)
ANALOG
CAMERA(S)
SERVER
CAMERAS
SERVER
BNC
RJ45
BNC
KEYBOARD
RJ45
ROUTER
KEYBOARD
DIGITAL IP
CAMERA(S)
*
DIGITAL IP
CAMERA(S)
ROUTER
INTERNET
*
DOMES
INTRANET
LOCAL AREA NETWORK (LAN)
WIDE AREA NETWORK (WAN)
WIRELESS (WiFi)
*
* COMPRESSED DIGITAL VIDEO
(MJPEG, MPEG-2, MPEG-4).
NETWORK**
VIDEO
RECORDER
** SUFFICIENT STORAGE TO SUPPORT ALL
SITES WITH SECURITY AUTHENTICATION.
*
MONITORING STATION
ANALOG
CAMERA(S)
RAID LEVEL 5 CONTROLLER FOR
EXPANDED STORAGE CAPACITY.
SERVER
ROUTER
RJ45
BNC
DIGITAL IP
CAMERA(S)
KEYBOARD
ALARM INPUT/
OUTPUT DEVICES
FIGURE 1-2
Multiple site system using analog/digital video
mechanisms, and brackets are smaller in size and weight
resulting in lower costs and providing more aesthetic
installations. The small cameras and lenses satisfy covert
video applications and are easy to conceal.
The potential importance of color in surveillance applications can be illustrated very clearly: turn off the color
on a television monitor to make it a monochrome scene.
It is obvious how much information is lost when the colors in the scene change to shades of gray. Objects that
were easily identified in the color scene become difficult
to identify in the monochrome scene. It is much easier to
pick out a person with a red shirt in the color image than
in a monochrome image.
The security industry has long recognized the value of
color to enhance personnel and article identification in
video surveillance and access control. One reason why
we can identify subjects more easily in color is that we
are used to seeing color, both in the real world and on
our TV at home. When we see a monochrome scene we
have to make an additional effort to recognize certain
information (besides the actual missing colors) thereby
decreasing the intelligence available. Color provides more
accurate identification of personnel and objects and leads
to a higher degree of apprehension and conviction of
criminals.
1.2.1 Video as Part of the Emergency
and Disaster Plan
Every organization regardless of size should have an emergency and disaster control plan that includes video as a
critical component. Depending on the organization an
anti-terrorist plan should take highest priority. Part of the
plan should be a procedure for succession of personnel
in the event one or more members of top management
are unavailable when disaster strikes. In large organizations the plan should include the designation of alternate
headquarters if possible, a safe document-storage facility,
and remote (off-site if possible) video operations capability. The plan must provide for medical aid and assure the
welfare of all employees in the organization. Using video
as a source of information, there should be a method to
alert employees in the event of a dangerous condition
and a plan to provide for quick police and emergency
response. There should be an emergency shutdown plan
6
CCTV Surveillance
and restoration procedures with designated employees acting as leaders. There should be CCTV cameras stationed
along evacuation routes and instructions for practice tests.
The evacuation plan should be prepared in advance and
tested.
A logical and effective disaster control plan should do
the following:
• Define emergencies and disasters that could occur as
they relate to the particular organization.
• Establish an organization and specific tasks with personnel designated to carry out the plan immediately before,
during, and immediately following a disaster.
• Establish a method for utilizing the organization’s
resources, in particular video, to analyze the disaster
situation and bring to bear all available resources.
• Recognize a plan to change from normal operations
into and out of the disaster emergency mode as soon as
possible.
Video plays a very important role in any emergency,
disaster and anti-terrorist plan:
• Video helps protect human life by enabling security or
safety officials to see remote locations and view first hand
what is happening, where it is happening, what is most
critical, and what areas must be attended to first.
• Aids in minimizing personal injury by permitting
“remote eyes” to get to those people who require immediate attention, or to send personnel to the area being
hit hardest to remove them from the area, or to bring
in equipment to protect them.
• Video reduces the exposure of physical assets to oncoming disaster, such as fire or flood, and prevents or at
least assesses document removal (of assets) by intruders
or any unauthorized personnel.
• Video documents the equipment and assets that were in
place prior to the disaster, recording them on VCR, DVR
or storage on an enterprise network to be compared
to the remaining assets after the disaster has occurred.
It also documents personnel and their activities before,
during, and after an incident.
• Probably more so than any other part of a security system, video will aid management and the security force
in minimizing any disaster or emergency. It is useful in
restoring an organization to normal operation by determining that no additional emergencies are in progress
and that procedures and traffic flow are normal in those
restored areas.
1.2.1.1 Protecting Life and Minimizing Injury
Through the intelligence gathered from the video system, security and disaster control personnel should move
all personnel to places of safety and shelter. Personnel
assigned to disaster control and remaining in a threatened
area should be protected by using video to monitor their
safety, and the access and egress at these locations. By such
monitoring, advance notice is available to provide a means
of support and assistance for those persons if injured, and
personnel that must be rescued or relieved.
1.2.1.2 Reducing Exposure of Physical Assets
and Optimizing Loss Control
Assets should be stored or secured properly before an emergency so that they will be less vulnerable to theft or loss. Video
is an important tool for continually monitoring safe areas
during and after a disaster to ensure that the material is not
removed. In an emergency or disaster, the well-documented
plan will call for specific personnel to locate highly valued
assets, secure them, and evacuate personnel.
1.2.1.3 Restoring Normal Operations Quickly
After an emergency situation has been brought under
control, security personnel can monitor and maintain the
security of assets and help determine that employees are
safe and have returned to their normal work routine.
1.2.1.4 Documenting an Emergency
For purposes of: (1) future planning, (2) liability and
insurance, and (3) evaluation by management and security personnel, video coverage of critical areas and operations during an emergency is an excellent tool and can
reduce financial losses significantly. Video recordings of
assets lost or stolen or personnel injured or killed can support a company’s claim that it was not negligent and that
it initiated a prudent emergency and disaster plan prior to
the event. Although video can provide crucial documentation of an event, it should be supplemented with highresolution photographs of specific instances or events.
If perimeter fences or walls were destroyed or damaged in a disaster, video can help prevent and document
intrusion or looting by employees, spectators, or other
outsiders.
1.2.1.5 Emergency Shutdown and Restoration
In the overall disaster plan, shutting down equipment
such as machinery, utilities, processes, and so on, must be
considered. If furnaces, gas generators, electrical power
equipment, boilers, high-pressure air or oil systems, chemical equipment, or rapidly rotating machinery could cause
damage if left unattended they should be shut down as
soon as possible. Again, video surveillance can be crucial
to determine if the equipment has been shut down properly, if personnel must enter the area to do so, or if it must
be shut down by other means.
Video’s Critical Role in the Security Plan
1.2.1.6 Testing the Plan
While a good emergency plan is essential, it should not
be tested for the first time in an actual disaster situation.
Deficiencies are always discovered during testing. Also, a
test serves to train the personnel who will carry out the
plan if necessary. Video can help evaluate the plan to
identify shortcomings and show personnel what they did
right and wrong. Through such peer review a practical
and efficient plan can be put in place to minimize losses
to the organization.
1.2.1.7 Standby Power and Communications
During any emergency or disaster, primary power and
communications between locations will probably be disrupted. Therefore, a standby power-generation system
should be provided for emergency monitoring and
response. This standby power comprised of a backup gaspowered generator or an uninterruptible power supply
with DC batteries to extend backup operation time will
keep emergency lighting, communications, and strategic
video equipment online as needed. Most installations use
a power sensing device that monitors the normal supply
of power at various locations. When the device senses that
power has been lost, the various backup equipments automatically switch to the emergency power source.
A prudent security plan anticipating an emergency will
include a means to power vital, audio, video, and other sensor equipment to ensure its operation during the event.
Since emergency video and audio communications must
be maintained over remote distances, alternative communication pathways should be supplied in the form of either
auxiliary hard-wired cable (copper wire or fiber optics) or a
wireless (RF, microwave, infrared) transmission system. It is
usually practical to provide a backup path to only the critical
cameras, not all of them. The standby generator supplying power to the video, safety, and emergency equipment
must be sized properly. For equipment that normally operates on 120 volt AC, inverters are used to convert the low
voltage from the backup DC batteries (typically 12 or 24
volts DC) to the required 120 volts AC (or 230 volts AC).
1.2.2 Security Investigations
Security investigators have used video very successfully with
respect to safeguarding company assets and preventing
theft, negligence, outside intrusion, and so on. By using
small, low-cost, covert CCTV (hidden camera and lens), it
is easy to positively identify a person or to document an
event without being noticed. Better video image quality,
smaller lenses and cameras, wireless video transmission,
and easier installation and removal of such equipment
have led to this high success. Many lenses and cameras that
can be hidden in rooms, hallways, or stationary objects are
7
available today. Equipment to provide such surveillance is
available for indoor or outdoor locations in bright sunlight
or in no light (IR-illuminated or thermal cameras).
1.2.3 Safety
Closed circuit television equipment is installed not always
for security reasons alone but also for safety purposes as
well. Security personnel can be alerted to unsafe practices
or accidents that require immediate attention. An attentive
guard can use CCTV cameras distributed throughout a
facility in stairwells, loading docks, around machinery, etc.
to observe and immediately document any safety violations
or incidents.
1.2.4 The Role of the Guard
Security guards are employed to protect plant assets and
personnel. Security and corporate management are aware
that guards are only one element of an organization’s
complete security plan. As such, the cost to implement the
guard force and its ability to protect assets and personnel
are analyzed in relation to the costs and roles of other
technological security solutions. In this respect video has
much to contribute: increased security for relatively low
capital investment and low operating cost, as compared
with a guard. Guards using video can increase the security
coverage or protection of a facility. Alternatively, installing
new CCTV equipment enables guards to monitor remote
sites, allowing guard count and security costs to be reduced
significantly.
1.2.5 Employee Training and Education
Video can be used as a powerful training tool. It is used
widely in education and the training of security personnel
because it can demonstrate lessons and examples vividly
to the trainee. In this post-9/11 era, security personnel
should receive professional training by all means including
real video footage. Video is an important tool for the security trainer. Example procedures of all types can be shown
conveniently in a short time period, and with instructions
given during the presentation. Videotaped real-life situations (not simulations or performances) can demonstrate
the consequences of mis-applied procedures and the benefits of proper planning and execution by trained and
knowledgeable personnel.
Every organization can supplement live training with
either professional training videos or actual scenes from
their own video system, demonstrating good and poor
practices as well as proper guard reaction in real cases
of intrusion, unacceptable employee behavior, and so on.
Such internal video systems can also be used in training
8
CCTV Surveillance
exercises: trainees may take part in videotaped simulations,
which are later critiqued by their supervisor. Trainees can
then observe their own actions to find ways to improve
and become more effective. Finally, such internal video
systems are very important tools during rehearsals or tests
of an emergency or disaster plan. After the run-through,
all team members can monitor their own reactions, and
managers or other professionals can critique them.
must specify as a minimum: (a) where and when unusual
behavior should be detected, (b) what the response should
be, and (c) how it should be reported and recorded. If
the intruder has violated a barrier or fence the intrusiondetection system should be able to determine that a
person—not an animal, bird, insect, leaf, or other object—
passed through the barrier. Video provides the most positive means for establishing this information. This breech
in security must then be communicated by some means to
security personnel so that a reaction force has sufficient
information to permit an appropriate response.
In another scenario, if material is being removed by
an unauthorized person in an interior location, a video
surveillance system activated by a video motion detector
(VMD) alarm should alert a guard and transmit the video
information to security personnel for appropriate action.
In both cases a guard force would be dispatched and the
event recorded on a VCR, DVR or network storage and/or
printed as hard copy for guard response, documentation,
and prosecution.
In summary, it is the combination of sensors, communication channels, monitoring displays, documentation
equipment and a guard force that provides the synergy to
maximize the security function. The integration of video,
intrusion-detection alarms, access control, and security
guards increases the overall security asset protection and
employee safety at a facility.
1.3 SYNERGY THROUGH INTEGRATION
Video equipment is most effective when integrated with
other security hardware and procedures to form a coherent security system. When video is combined with the
other security sensors the total security system is more than
the individual subsystems. Synergy obtains when video
assessment is combined with intrusion and motion alarm
sensors, electronic access control, fire alarms, communications, and security guard personnel (Figure 1-3).
1.3.1 Integrated Functions
Functionally the integrated security system is designed as
a coordinated combination of equipment, personnel, and
procedures that: (a) uses each component in a way that
enhances the use of every other component and (b) optimally achieves the system’s stated objective.
In designing a security system, each element’s potential contribution to loss prevention, asset protection, or
personnel safety must be considered. The security plan
1.3.2 System Hardware
Since a complete video security system may be assembled
from components manufactured by different companies,
INTEGRATED
SECURITY SYSTEM
ACCESS
CONTROL
• ELECTRONIC
• VIDEO
• BIOMETRIC
ALARMS:
• FIRE
• SAFETY
INTRUSION
DETECTION
VIDEO
SURVEILLANCE
COMMUNICATIONS
INTEGRATED SECURITY SYSTEM SYNERGY:
• MAXIMIZE ASSET AND PERSONNEL PROTECTION
• PROVIDE DISASTER CONTROL
• OPTIMIZE RECOVERY PLAN
FIGURE 1-3
Integrated security system
SECURITY
PERSONNEL
Video’s Critical Role in the Security Plan
all equipment must be compatible. The video equipment
should be specified by one consulting or architecture/engineering firm, and the system and service should
be purchased, installed, and maintained through a single
system integrator, dealer/installer, or general contractor.
If a major supplier provides a turnkey system, including
all equipment, training, and maintenance, the responsibility of system operation resides with one vendor, which
is easier to control. Buying from one source also permits management to go back to one installer or general
contractor if there are any problems instead of having
to point fingers or negotiate for service among several
vendors.
Choosing a single supplier obviously requires thorough
analysis to determine that the supplier: (1) will provide a
system that meets the requirements of the facility, (2) will
be available for maintenance when required, and (3) will
still be in business in 5 or 10 years. There are many companies that can supply complete video systems including
cameras and housings, lenses, pan/tilt mechanisms, multiplexers, time-lapse VCRs or DVRs, analog and digital
networks, and other security equipment required for an
integrated video system. If the end user chooses components from various manufacturers, care must be taken by
the system designer and installer to be aware of the differences and interface the equipment properly.
If the security plan calls for a simple system with potential for later expansion the equipment should be modular
and ready to accept new technology as it becomes available. Many larger manufacturers of security equipment
anticipate this integration and expansion requirement and
design their products accordingly.
Service is a key ingredient for successful system operation. If one component fails, repair or replacement
must be done quickly, so that the system is not shut
down. Near-continuous operation is accomplished by the
direct replacement method, immediate maintenance by
an in-house service organization, or quick-response service calls from the installer/contractor. Service consideration should be addressed during the planning and initial
design stages, as they affect choice of manufacturer and
service provider. Most vendors use the replacement technique to maintain and service equipment. If part of the
system fails, the vendor replaces the defective equipment
and sends it to the factory for repair. This service policy
decreases security system downtime.
The key to a successful security plan is to choose the
right equipment and service company, one that is customer oriented and knowledgeable about reliable, technologically superior products that satisfy the customer needs.
1.4 VIDEO’S ROLE AND ITS APPLICATIONS
In its broadest sense, the purpose of CCTV in any security plan is to provide remote eyes for a security operator:
9
to create live-action displays from a distance. The video
system should have recording means—either a VCR or
a DVR, or other storage media—to maintain permanent
records for training or evidence. Following are some applications for which video provides an effective solution:
• When overt visual observation of a scene or activity is
required from a remote location.
• An area to be observed contains hazardous material or
some action that may kill or injure personnel. Such areas
may have toxic chemicals, biological or radioactive material, substances with high potential for fire or explosion,
or items that may emit X-ray radiation or other nuclear
radiation.
• Visual observation of a scene must be covert. It is much
easier to hide a small camera and lens in a target location than to station a person in the area.
• There is little activity to watch in an area, as in an
intrusion-detection location or a storage room, but significant events must be recorded in the area when they
occur. Integration of video with alarm sensors and a
time-lapse/real-time VCR or DVR provides an extremely
powerful solution.
• Many locations must be observed simultaneously by one
person from a central security location.
• Tracing a person or vehicle from an entrance into a
facility to a final destination. The security force can predict where the person or vehicle can be interdicted.
• Often a guard or security officer must only review a
scene for activity periodically. The use of video eliminates the need for a guard to make rounds to remote
locations, which is wasteful of the guard’s time.
• When a crime has been committed, capturing the scene
using the video camera and recorder to have a permanent record and hard copy printout of the activity and
event. The proliferation of high-quality printed images
from VCR/DVR equipment has clearly made the case
for using video for creating permanent records.
1.4.1 Video System Solutions
The most effective way to determine that a theft has
occurred, when, where, and by whom, is to use video
for detection and recording. The particular event can
be identified, stored, and later reproduced for display or
hard copy. Personnel can be identified on monochrome
or color CCTV monitors. Most security installations use
color CCTV cameras that provide sufficient information
to document the activity and event or identify personnel
or articles. The color camera permits easier identification
of personnel and objects.
If there is an emergency or disaster and security personnel must see if personnel are in a particular area, video can
provide an instantaneous assessment of personnel location
and availability.
10
CCTV Surveillance
In many cases during normal operations, security personnel can help ensure the safety of personnel in a facility,
determine that employees or visitors have not entered the
facility, or confirm that personnel have exited the facility. Such functions are used for example where dangerous
jobs are performed or hazardous material is handled.
The synergistic combination of audio and video information from a remote site provides for effective security. Several camera manufacturers and installers combine
video and audio (one-way or duplex) using an external microphone or one installed directly in the camera.
The video and audio signals are transmitted over the
same coaxial, unshielded-twisted-pair (UTP), or fiber-optic
cable, to the security monitoring location where the scene
is viewed live and/or recorded. When there is activity in
the camera area the video and audio signals are switched
to the monitor and the guard sees and hears the activity
in the scene and initiates a response.
1.4.2 Overt vs. Covert Video
Most video installations use both overt and covert (hidden) CCTV cameras, with more cameras overt than covert.
Overt installations are designed to deter crime and provide general surveillance of remote areas such as parking
lots, perimeter fence lines, warehouses, entrance lobbies,
hallways, or production areas. When CCTV cameras and
lenses are exposed, all managers, employees, and visitors
realize that the premises are under constant video surveillance. When the need arises, covert installations are used
to detect and observe clandestine activity. While overt
video equipment is often large and not meant to be concealed, covert equipment is usually small and designed
to be hidden in objects in the environment or behind a
ceiling or wall. Overt cameras are usually installed permanently whereas covert cameras are usually designed to
be installed quickly, left in place for a few hours, days,
or weeks, and then removed. Since minimizing installation time is desirable when installing covert cameras, video
signal transmission often is wireless rather than wired.
snow), wind, dirt, dust, sand, salt, and smoke. The outdoor systems use natural daylight and artificial lighting
at night supplied either by parking lights or by a colocated infrared (IR) source. Some cameras can automatically switch from color operation during daylight, to
monochrome when the lighting decreases below some
specified level for nighttime operation.
Most video security applications use fixed, permanently
installed video equipment. These systems are installed for
months and years and left in place until they are superseded by new equipment or they are no longer required.
There are many cases, however, where there is a requirement for a rapid deployment of video equipment to be
used for a short period of time: days, weeks, or sometimes
months, and then removed to be used again in another
application. Chapter 21 describes some of these transportable rapid deployment video systems.
1.4.4 Safety Applications
In public, government, industrial, and other facilities, a
safety, security, and personnel protection plan must guard
personnel from harm caused by accident, human error,
sabotage, or terrorism. Security forces are expected to
monitor the conditions and activities at all locations in the
facility through the use of CCTV cameras.
In a hospital room or hallway the video cameras may
serve a dual function: monitoring patients while also determining the status and location of employees, visitors, and
others. A guard can watch entrance and exit doors, hallways, operating rooms, drug dispensaries, and other vital
areas.
Safety personnel can use video for evacuation and to
determine if all personnel have left the area and are safe.
Security personnel can use video for remote traffic monitoring and control and to ascertain high-traffic locations
and how best to control them. Video plays a critical role
in public safety, as a tool for monitoring vehicular traffic
on highways and city streets, in truck and bus depots, at
public rail and subway facilities, airports, power plants, just
to name a few.
1.4.3 Security Surveillance Applications
Many video applications fall broadly into two types,
indoor and outdoor. This division sets a natural boundary
between equipment types: those suitable for controlled
indoor environments and those suitable for harsher outdoor environments. The two primary parameters are environmental factors and lighting factors. The indoor system
requires artificial lighting that may or may not be augmented by daylight. The indoor system is subject to
only mild indoor temperature and humidity variations,
dirt, dust, and smoke. The outdoor system must withstand extreme temperatures, precipitation (fog, rain, and
1.4.5 Video Access Control
As security requirements become more complex and
demanding, video access control and electronic access
control equipments should work synergistically with each
other. For medium- to low-level access control security requirements, electronic card-reading systems are
adequate after a person has first been identified at some
exterior perimeter location. For higher security, personal
biometric descriptors (iris scanning, fingerprint, etc.)
and/or video identification are necessary.
Video’s Critical Role in the Security Plan
Video surveillance is often used with electronic or video
access control equipment. Video access control uses video
to identify a person requesting access at a remote location, on foot or in a vehicle. A guard can compare the
live image and the photo ID carried by the person on a
video monitor and then either allow or deny entry. For
the highest level of access control security the guard uses
a system to compare the live image of the person to an
image of the person retrieved from a video image database
or one stored in a smart card. The two images are displayed side by side on a split-screen monitor along with
other pertinent information. The video access control system can be combined with an electronic access control
system to increase security and provide a means to track
all attempted entries.
There are several biometric video access control systems
which can positively identify a person enrolled in the system using iris, facial, or retina identification.
1.5 THE BOTTOM LINE
The synergy of a CCTV security system implies the following functional scenario:
• An intrusion alarm sensor or VMD will detect an unauthorized intrusion or entry or attempt to remove equipment from an area.
• A video camera located somewhere in the alarm area
is viewing the area at the location or may be pointed
manually or automatically (from the guard site) to view
the alarm area.
• The information from the alarm sensor and/or camera
is transmitted immediately to the security console, monitored by personnel, and/or recorded for permanent
documentation.
• The security operator receiving the alarm information
has a plan to dispatch personnel to the location or to
take some other appropriate action.
• After dispatching a security person to the alarm area
the guard resumes normal security duties to view the
response, give additional instruction, and monitor any
future event.
• After a reasonable amount of time the person dispatched should neutralize the intrusion or other event.
The security guard resumes monitoring that situation to
bring it to a successful conclusion and continues monitoring the facility.
The use of video plays a crucial role in the overall security system plan. During an intrusion, disaster or theft,
the video system provides information to the guard, who
must make some identification of the perpetrator, assess
the problem, and respond appropriately. An installation
containing suitable and sufficient alarm sensors and video
11
cameras permits the guard to follow the progress of the
event and assist the response team in countering the
attack.
The use of video and the VMD capability to track an
intruder is most effective. With an intrusion alarm and
visual video information, all the elements are in place for
a timely, reliable transfer of information to the security
officer. For maximum effectiveness, all parts of the security
system must work together synergistically. If an intrusion
alarm fails, the command post may not see the intruder
with sufficient advance notice. If the video fails, the guard
cannot identify the perpetrator or evaluate the extent of
the security breech even though he may know that an
intrusion has occurred. It is important that the security
officer be alert and that proper audio and visual cues are
provided to alert the guard when an alarm has occurred. If
inadequate alarm annunciation is provided and the guard
misses or misinterprets the alarm and video input, the data
from either or both are not acted upon and the system
fails.
In an emergency such as a terrorist attack, fire, flood,
malfunctioning machinery, burst utility pipeline, etc. the
operation of video, safety sensors, and human response
at the console are all required. Video is an inexpensive
investment for preventing accidents and minimizing damage when an accident occurs. Since the reaction time to
a terrorist attack, fire or other disaster is critical, having
various cameras at the critical locations before personnel
arrive is very important. Closed circuit television cameras
act as real-time eyes at the emergency location, permitting security and safety personnel to send the appropriate
reaction force with adequate equipment to provide optimum response. In the case of a fire, while a sprinkler may
activate or a fire sensor may produce an alarm, a CCTV
camera can quickly ascertain whether the event is a false
alarm, a minor alarm, or a major event. The automatic
sprinkler and fire alarm system might alert the guard to
the event but the video “eyes” viewing the actual scene
prior to the emergency team’s dispatch often save lives
and reduce asset losses.
In the case of a security violation, if a sensor detects
an intrusion the guard monitoring the video cameras can
determine if the intrusion requires the dispatch of personnel or some other response. In the event of a major,
well-planned attack on a facility by a terrorist organization or other intrusion, a diversionary tactic such as a false
alarm can quickly be discovered through the use of video
thereby preventing an inappropriate response.
To justify expenditures on security and safety equipment an organization must expect a positive return on
investment. The value of assets protected must be greater
than the amount spent on security, and the security system must adequately protect personnel and visitors. An
effective security system reduces theft, saves money, and
saves lives.
This page intentionally left blank
Chapter 2
Video Technology Overview
CONTENTS
2.1 Overview
2.2 The Video System
2.2.1 The Role of Light and
Reflection
2.2.2 The Lens Function
2.2.3 The Camera Function
2.2.4 The Transmission Function
2.2.5 The Monitor Function
2.2.6 The Recording Function
2.3 Scene Illumination
2.3.1 Natural Light
2.3.2 Artificial Light
2.4 Scene Characteristics
2.4.1 Target Size
2.4.2 Reflectivity
2.4.3 Effects of Motion
2.4.4 Scene Temperature
2.5 Lenses
2.5.1 Fixed-Focal-Length Lens
2.5.2 Zoom Lens
2.5.3 Vari-Focal Lens
2.5.4 Panoramic—360 Lens
2.5.5 Covert Pinhole Lens
2.5.6 Special Lenses
2.6 Cameras
2.6.1 The Scanning Process
2.6.1.1 Raster Scanning
2.6.1.2 Digital and Progressive Scan
2.6.2 Solid-State Cameras
2.6.2.1 Analog
2.6.2.2 Digital
2.6.2.3 Internet
2.6.3 Low-Light-Level Intensified
Camera
2.6.4 Thermal Imaging Camera
2.6.5 Panoramic 360 Camera
2.7 Transmission
2.7.1 Hard-Wired
2.7.1.1 Coaxial Cable
2.7.1.2 Unshielded Twisted Pair
2.7.1.3 LAN, WAN, Intranet and Internet
2.7.2 Wireless
2.7.3 Fiber Optics
2.8 Switchers
2.8.1 Standard
2.8.2 Microprocessor-Controlled
2.9 Quads and Multiplexers
2.10 Monitors
2.10.1 Monochrome
2.10.2 Color
2.10.3 CRT, LCD, Plasma Displays
2.10.4 Audio/Video
2.11 Recorders
2.11.1 Video Cassette Recorder (VCR)
2.11.2 Digital Video Recorder (DVR)
2.11.3 Optical Disk
2.12 Hard-copy Video Printers
2.13 Ancillary Equipment
2.13.1 Camera Housings
2.13.1.1 Standard-rectangular
2.13.1.2 Dome
2.13.1.3 Specialty
2.13.1.4 Plug and Play
2.13.2 Pan/Tilt Mounts
2.13.3 Video Motion Detector (VMD)
2.13.4 Screen Splitter
2.13.5 Camera Video Annotation
2.13.5.1 Camera ID
2.13.5.2 Time and Date
2.13.6 Image Reversal
2.14 Summary
13
14
CCTV Surveillance
2.1 OVERVIEW
The second half of the 1990s has witnessed a quantum
jump in video security technology. This technology has
manifest with a new generation of video components,
i.e. digital cameras, multiplexers, DVRs, etc. A second significant activity has been the integration of security systems
with computer-based LANs, wide area networks (WANs),
wireless networks (WiFi), intranets and, Internet and the
World Wide Web (WWW) communications systems.
Although today’s video security system hardware is based
on new technology which takes advantage of the great
advances in microprocessor computing power, solid-state
and magnetic memory, digital processing, and wired and
wireless video signal transmission (analog, digital over the
Internet, etc.), the basic video system still requires the
lens, camera, transmission medium (wired cable, wireless),
monitor, recorder, etc. This chapter describes current
video security system components and is an introduction
to their operation.
The primary function of any video security or safety
system is to provide remote eyes for the security force
located at a central control console or remote site. The
video system includes the illumination source, the scene
to be viewed, the camera lens, the camera, and the means
of transmission to the remote monitoring and recording
CAMERA SITE
equipment. Other equipment often necessary to complete
the system include video switchers, multiplexers, VMDs,
housings, scene combiners and splitters, and character
generators.
This chapter describes the technology used to: (1) capture the visual image, (2) convert it to a video signal,
(3) transmit the signal to a receiver at a remote location,
(4) display the image on a video monitor, and (5) record
and print it for permanent record. Figure 2-1 shows the
simplest video application requiring only one video camera and monitor.
The printer and video recorder are optional. The camera may be used to monitor employees, visitors, or people
entering or leaving a building. The camera could be
located in the lobby ceiling and pointed at the reception
area, the front door, or an internal access door. The monitor might be located hundreds or thousands of feet away,
in another building or another city or country with the
security personnel viewing that same lobby, front door, or
reception area. The video camera/monitor system effectively extends the eyes, reaching from observer location to
the observed location. The basic one-camera system shown
in Figure 2-1 includes the following hardware components.
• Lens. Light from the illumination source reflects off
the scene. The lens collects the light from the scene
CONSOLE MONITORING SITE
MONITOR
(CRT/LCD)
VIDEO
CAMERA
TRANSMISSION MEANS
COAX
CABLE
COAXIAL
UTP (UNSHILDED TWISTED PAIR)
OPTICAL
LENS
VIDEO
PRINTER
SCENE
ILLUMINATION
SOURCE
(NATURAL,
ARTIFICIAL)
FIGURE 2-1
Single camera video system
VIDEO RECORDER
(DVR/VCR)
Video Technology Overview
•
•
•
•
•
image, using thermal, inkjet, laser, or other printing
technology.
and forms an image of the scene on the light-sensitive
camera sensor.
Camera. The camera sensor converts the visible scene
formed by the lens into an electrical signal suitable
for transmission to the remote monitor, recorder, and
printer.
Transmission Link. The transmission media carries
the electrical video signal from the camera to the
remote monitor. Hard-wired media choices include:
(a) coaxial, (b) two-wire unshielded twisted-pair (UTP),
(c) fiber-optic cable, (d) LAN, (e) WAN, (f) intranet,
and (g) Internet network. Wireless choices include:
(a) radio frequency (RF), (b) microwave, or (c) optical
infrared (IR). Signals can be analog or digital.
Monitor. The video monitor or computer screens display (CRT, LCD or plasma) the camera image by converting the electrical video signal back into a visible
image on the monitor screen.
Recorder. The camera scene is permanently recorded
by a real-time or TL VCR onto a magnetic tape cassette
or by a DVR using a magnetic disk hard drive.
Hard-copy Printer. The video printer produces a hardcopy paper printout of any live or recorded video
The first four components are required to make a simple video system work. The recorder and/or printer is
required if a permanent record is required.
Figure 2-2 shows a block diagram of a multi-camera
analog video security system using these components plus
additional hardware and options to expand the capability
of the single-camera system to multiple cameras, monitors,
recorders, etc. providing a more complex video security
system.
Additional ancillary supporting equipment for more
complex systems includes: camera switchers, quads, multiplexers, environmental camera housings, camera pan/tilt
mechanisms, image combiners and splitters, and scene
annotators.
• Camera Switcher, Quad, Multiplexer. When a CCTV
security system has multiple cameras, an electronic
switcher, quad, or multiplexer is used to select different cameras automatically or manually to display the
images on a single or multiple monitors, as individual
or multiple scenes. The quad can digitally combine four
CAMERA SITE
LENS
CONSOLE MONITORING SITE
• QUAD
• MULTIPLEXER
• SWITCHER
TRANSMISSION MEANS
• COAXIAL
• UTP
CAMERA 1
• OPTICAL
QUAD
1
2
3
MONITOR
(CRT/LCD)
4
1 SEQUENCE
2
MONITOR
(CRT/LCD)
3
o
o
VIDEO PRINTER
o
N
FIGURE 2-2
Comprehensive video security system
15
VIDEO RECORDER
(DVR/VCR)
16
CCTV Surveillance
cameras. The multiplexer can digitally combine 4, 9, 16,
and even 32 separate cameras.
• Housings. The many varieties of camera/lens housings
fall into three categories: indoor, outdoor and integral
camera/housing assemblies. Indoor housings protect
the camera and lens from tampering and are usually
constructed from lightweight materials. Outdoor housings protect the camera and lens from the environment:
from precipitation, extremes of heat and cold, dust, dirt,
and vandalism.
• Dome Housing. The dome camera housing uses a
hemispherical clear or tinted plastic dome enclosing
a fixed camera or a camera with pan/tilt and zoom
lens capability.
• Plug and Play Camera/Housing Combination. To simplify surveillance camera installations many manufacturers are now packaging the camera-lens-housing as
a complete assembly. These plug-and-play cameras are
ready to mount in a wall or ceiling and to connect the
power in and the video out.
• Pan/Tilt Mechanism. When a camera must view a large
area, a pan and tilt mount is used to rotate it horizontally (panning) and to tilt it, providing a large angular
coverage.
• Splitter/Combiner/Inserter. An optical or electronic
image combiner or splitter is used to display more
than one camera scene on a single monitor.
• Annotator. A time and date generator annotates the
video scene with chronological information. A camera
identifier puts a camera number (or name—FRONT
DOOR, etc.) on the monitor screen to identify the
scene displayed by the camera.
The digital video surveillance system includes most of the
devices in the analog video system. The primary differences manifest in using digital electronics and digital processing within the video devices. Digital video components
use digital signal processing (DSP), digital video signal
compression, digital transmission, recording and viewing.
Figure 2-3 illustrates these devices and signal paths and the
overall system block diagram for the digital video system.
SITE 2
SITE 1
ANALOG
CAMERA(S)
ANALOG
CAMERA(S)
CAMERAS
SERVER
SERVER
ROUTER
RJ45
BNC
KEYBOARD
BNC RJ45
*
KEYBOARD
DIGITAL IP
CAMERA(S)
DIGITAL IP
CAMERA(S)
ROUTER
INTERNET
*
INTRANET
LOCAL AREA NETWORK (LAN)
DOMES
WIDE AREA NETWORK (WAN)
WIRELESS (WiFi)
*
* COMPRESSED DIGITAL VIDEO
(MJPEG, MPEG-2, MPEG-4).
NETWORK **
VIDEO
RECORDER
** SUFFICIENT STORAGE TO SUPPORT ALL SITES
WITH SECURITY AUTHENTICATION.
*
MONITORING STATION
RAID LEVEL 5 CONTROLLER FOR
EXPANDED STORAGE CAPACITY.
ANALOG
CAMERA(S)
ROUTER
RJ45
MODEM
BNC
DIGITAL IP
CAMERA(S)
MODEM
KEYBOARD
POTS
DSL
OTHER
ALTERNATE LAND LINE
SITE TO SITE CONNECTION
FIGURE 2-3
Networked digital video system block diagram
SERVER
ALARM INPUT/
OUTPUT DEVICES
Video Technology Overview
2.2 THE VIDEO SYSTEM
Figure 2-4 shows the essentials of the CCTV camera
environment: illumination source, camera, lens, and the
camera–lens combined field of view (FOV), that is the
scene the camera–lens combination sees.
2.2.1 The Role of Light and Reflection
A scene or target area to be viewed is illuminated by natural or artificial light sources. Natural sources include the
sun, the moon (reflected sunlight), and starlight. Artificial
sources include incandescent, sodium, metal arc, mercury, fluorescent, infrared, and other man-made lights.
Chapter 3 describes all of these light sources in detail.
The camera lens receives the light reflected from the
scene. Depending on the scene to be viewed the amount
of light reflected from objects in the scene can vary from
5 or 10% to 80 or 90% of the light incident on the
scene. Typical values of reflected light for normal scenes
such as foliage, automobiles, personnel, and streets fall in
the range from about 25–65%. Snow-covered scenes may
reach 90%.
The amount of light received by the lens is a function
of the brightness of the light source, the reflectivity of the
scene, and the transmission characteristics of the intervening atmosphere. In outdoor applications there is usually a
considerable optical path from the source to the scene and
back to the camera; therefore the transmission through
the atmosphere must be considered. When atmospheric
conditions are clear, there is generally little or no attenuation of the reflected light from the scene. However,
when there is precipitation (rain, snow, or sleet, or when
fog intervenes) or in dusty, smoky, or sand-blown environments, this attenuation might be substantial and must
be considered. Likewise in hot climates thermal effects
(heat waves) and humidity can cause severe attenuation
and/or distortion of the scene. Complete attenuation of
the reflected light from the scene (zero visibility) can
occur, in which case no scene image is formed.
Since most solid-state cameras operate in the visible and
near-infrared wavelength region the general rule of thumb
with respect to visibility is that if the human eye cannot
see the scene neither can the camera. Under this situation, no amount of increased lighting will help; however,
if the visible light can be filtered out of the scene and only
the IR portion used, scene visibility might be increased
NATURAL OR ARTIFICIAL
ILLUMINATION SOURCE
SCENE VIEWED BY
CAMERA/LENS
REFLECTED LIGHT FROM SENSOR
LENS
CAMERA
LENS FIELD
OF VIEW (FOV)
SENSOR:
CCD
CMOS
INTENSIFIER
THERMAL IR
FIGURE 2-4
17
Video camera, scene, and source illumination
VIDEO OUT
POWER IN
18
CCTV Surveillance
SOLID STATE
CCD, CMOS
SENSOR
SENSOR GEOMETRY
d = SENSOR
v = 3 UNITS HIGH
DIAGONAL
v = VERTICAL
v
d
h = 4 UNITS WIDE
h
h = HORIZONTAL
CCTV
CAMERA
CAMERA SENSOR
FOV
LENS
D
SCENE
D = DISTANCE FROM SCENE
TO LENS
H
V
HORIZONTAL
WIDTH (H)
FIGURE 2-5
VERTICAL HEIGHT (V)
Video scene and sensor geometry
somewhat. This problem can often be overcome by using a
thermal infrared (IR) imaging camera that works outside
of the visible wavelength range. These thermal IR cameras
produce a monochrome display with reduced image quality and are much more expensive than the charge coupled
device (CCD) or complimentary metal oxide semiconductor (CMOS) cameras (see Section 2.6.4). Figure 2-5 illustrates the relationship between the viewed scene and the
scene image on the camera sensor.
The lens located on the camera forms an image of the
scene and focuses it onto the sensor. Almost all video
systems used in security systems have a 4-by-3 aspect ratio
(4 units wide by 3 units high) for both the image sensor
and the field of view. The width parameter is designated
as h, and H, and the vertical as v, and V. Some cameras
have a 16 units wide by 9 units high definition television
(HDTV) format.
2.2.2 The Lens Function
The camera lens is analogous to the lens of the human
eye (Figure 2-6) and collects the reflected radiation from
the scene much like the lens of your eye or a film camera.
The function of the lens is to collect reflected light from
the scene and focus it into an image onto the CCTV camera sensor. A fraction of the light reaching the scene from
the natural or artificial illumination source is reflected
toward the camera and intercepted and collected by the
camera lens. As a general rule, the larger the lens diameter, the more light will be gathered, the brighter the image
on the sensor, and the better the final image on the monitor. This is why larger-aperture (diameter) lenses, having
a higher optical throughput, are better (and more expensive) than smaller-diameter lenses that collect less light.
Under good lighting conditions—bright indoor lighting, outdoors under sunlight—the large-aperture lenses
are not required and there is sufficient light to form
a bright image on the sensor by using small-diameter
lenses.
Most video applications use a fixed-focal-length (FFL)
lens. The FFL lens like the human eye lens covers a constant angular field of view (FOV). The FFL lens images
a scene with constant fixed magnification. A large variety
of CCTV camera lenses are available with different focal
lengths (FLs) that provide different FOVs. Wide-angle,
medium-angle, and narrow-angle (telephoto) lenses produce different magnifications and FOVs. Zoom and varifocal lenses can be adjusted to have variable FLs and FOVs.
Most CCTV lenses have an iris diaphragm (as does
the human eye) to adjust the open area of the lens and
change the amount of light passing through it and reaching the sensor. Depending on the application, manual or
automatic-iris lenses are used. In an automatic-iris CCTV
lens, as in a human eye lens, the iris closes automatically
when the illumination is too high and opens automatically
Video Technology Overview
19
EYE OR CAMERA SENSOR SCENE
SCENE
CAMERA SENSOR
FIELD OF VIEW
LENS
IRIS
EYE RETINA
CAMERA SENSOR
EYE FIELD OF VIEW
AT SCENE
17 mm
EYE MAGNIFICATION = 1
EYE LENS FOCAL LENGTH = 17 mm (0.67")
FIGURE 2-6
Comparing the human eye to the video camera lens
when it is too low, thereby maintaining the optimum illumination on the sensor at all times. Figure 2-7 shows representative samples of CCTV lenses, including FFL, varifocal, zoom, pinhole, and a large catadioptric lens for long
range outdoor use (which combines both mirror and glass
optical elements). Chapter 4 describes CCTV lens characteristics in detail.
2.2.3 The Camera Function
The lens focuses the scene onto the camera image sensor which acts like the retina of the eye or the film in a
photographic camera. The video camera sensor and electronics convert the visible image into an equivalent electrical signal suitable for transmission to a remote monitor.
Figure 2-8 is a block diagram of a typical analog CCTV
camera.
The camera converts the optical image produced by
the lens into a time-varying electric signal that changes
(modulates) in accordance with the light-intensity distribution throughout the scene. Other camera electronic
circuits produce synchronizing pulses so that the timevarying video signal can later be displayed on a monitor or
recorder, or printed out as hard copy on a video printer.
While cameras may differ in size and shape depending on
specific type and capability, the scanning process used by
most cameras is essentially the same. Almost all cameras
must scan the scene, point by point, as a function of time.
(An exception is the image intensifier.) Solid-state CCD
or CMOS color and monochrome cameras are used in
most applications. In scenes with low illumination, sensitive CCD cameras with infrared (IR) illuminators are used.
In scenes with very low illumination and where no active
illumination is permitted (i.e. covert) low-light-level (LLL)
intensified CCD (ICCD) cameras are used. These cameras
are complex and expensive (Chapter 19).
Figure 2-9 shows a block diagram of a the analog camera
with (a) digital signal processing (DSP) and (b) the all
digital internet protocol (IP) video camera.
In the early 1990s the non-broadcast, tube-type color
cameras available for security applications lacked longterm stability, sensitivity, and high resolution. Color cameras did not find much use in security applications until
solid-state color CCTV cameras became available through
the development of solid-state color sensor technology
and widespread use of consumer color CCD cameras used
in camcorders. Color cameras have now become standard in security systems and most CCTV security cameras
in use today are color. Figure 2-10 shows representative
CCTV cameras including monochrome and color solidstate CCD and CMOS cameras, a small single board camera, and a miniature remote head camera. Chapters 5, 14,
15 and 19 describe standard and LLL security CCTV cameras in detail.
2.2.4 The Transmission Function
Once the camera has generated an electrical video signal
representing the scene image, the signal is transmitted to
a remote security monitoring site via some transmission
20
CCTV Surveillance
(A) MOTORIZED ZOOM
(B) CATADIOPTRIC LONG FFL
(D) WIDE FOV FFL
(F) NARROW FOV (TELEPHOTO) FFL
FIGURE 2-7
(C) FLEXIBLE FIBER OPTIC
(E) RIGID FIBER OPTIC
(G) MINI-LENS
(H) STRAIGHT AND RIGHT-ANGLE
PINHOLE LENSES
Representative video lenses
means: coaxial cable, two-wire twisted-pair, LAN, WAN,
intranet, Internet, fiber optic, or wireless techniques. The
choice of transmission medium depends on factors such
as distance, environment, and facility layout.
If the distance between the camera and the monitor is
short (10–500 feet), coaxial cable, UTP, and fiber optic
or wireless is used. For longer distances (500 to several
thousand feet) or where there are electrical disturbances,
fiber-optic cable and UTP are preferred. For very long
distances and in harsh environments (frequent lightning
storms) or between separated buildings where no electrical grounding between buildings is in place, fiber optics is
the choice. In applications where the camera and monitor
are separated by roadways or where there is no right-of-
way, wireless systems using RF, microwave or optical transmission is used. For transmission over many miles or from
city to city the only choice is the digital or Internet IP
camera using compression techniques and transmitting
over the Internet and WWW. Images from these Internet
systems are not real-time but sometimes come close to
real-time. Chapters 6 and 7 describe all of these video
transmission media.
2.2.5 The Monitor Function
At the monitoring site a cathode ray tube (CRT), LCD
or plasma monitor converts the video signal back into a
Video Technology Overview
21
IN
HORIZONTAL
AND VERTICAL
SCANNING
TIMING AND
SYNCHRONIZING
SENSOR:
CCD, CMOS,
IR, ETC.
VIDEO
AMPLIFIER
HORIZONTAL
AND VERTICAL
SYNC OUT
(OPTIONAL)
OUT
LENS
DIGITAL
SIGNAL
PROCESSING
(DSP)
VIDEO
OUTPUT
ANALOG/
DIGITAL
VIDEO OUT
75 ohm
OPTICAL IMAGE
FOCUSED ONTO
IMAGE SENSOR
FIGURE 2-8
Analog CCTV camera block diagram
COLUMN/ROW
PIXEL
SCANNING
ANALOG
IN
TIMING AND
SYNCHRONIZING
OUT
HORIZONTAL
AND VERTICAL
SYNC-OPTIONAL
LENS
SENSOR:
CCD, CMOS
INTENSIFIER,
INFRARED
DIGITAL
SIGNAL
PROCESSING
(DSP)
VIDEO
AMPLIFIER
ANALOG
VIDEO
OUTPUT
VIDEO OUT
75 ohm
SDRAM/FLASH
MEMORY
DIGITAL
LENS, P/T
DRIVERS
AF, IRIS, ZOOM,
P/T, SHUTTER
LENS
INTERFACE
LOGIC
TIMING AND
SYNCHRONIZING
COLUMN/ROW
PIXEL
SCANNING
VIDEO
PROCESSOR
VIDEO SIGNAL
COMPRESSION
MJPEG, MPEG-4
(DSP)
WIRED ETHERNET PORT
INTERNET
INTRANET
LAN/WAN
WIRELESS PORT
802.11 a/b/g
SENSOR:
CCD, CMOS
INTENSIFIER,
INFRARED
NTSC/PAL
PORT
ALARM TRIGGERS
ALARM OUTPUTS
FIGURE 2-9
DIGITAL VIDEO
RECORDER
Analog camera with DSP and all digital camera block diagram
LCD DISPLAY
22
CCTV Surveillance
(A) INTENSIFIED CCD CAMERA
(ICCD)
(D) MINIATURE CAMERA
FIGURE 2-10
(B) 1/3" FORMAT CS MOUNT
COLOR CAMERA
(C) 1/2" FORMAT CS MOUNT
MONOCHROME CAMERA
(E) REMOTE HEAD CAMERA
(F) THERMAL
Representative video cameras
visual image on the monitor face via electronic circuitry
similar but inverse to that in the camera. The final scene
is produced by a scanning electron beam in the CRT in
the video monitor. This beam activates the phosphor on
the cathode-ray tube, thereby producing a representation
of the original image onto the faceplate of the monitor.
Alternatively the video image is displayed point by point
on an LCD or plasma screen. Chapter 8 describes monitor
and display technology and hardware. A permanent record
of the monitor video image is made using a VCR tape or
DVR hard disk magnetic recorder and a permanent hard
copy is printed with a video printer.
2.2.6 The Recording Function
For decades the VCR has been used to record
monochrome and color video images. The real-time and
TL VCR magnetic tape systems have been a reliable and
efficient means for recording security scenes.
Beginning in the mid-1990s the DVR was developed
using a computer hard disk drive and digital electronics to provide video image recording. The availability of
large memory disks (hundreds of megabytes) made these
machines available for long duration security recording.
Significant advantages of the DVR over the VCR are the
high reliability of the disk as compared with the cassette
tape, its ability to perform high speed searches (retrieval
of images) anywhere on the disk, absence of image deterioration after many copies are made.
2.3 SCENE ILLUMINATION
A scene is illuminated by either natural or artificial illumination. Monochrome cameras can operate with any type
of light source. Color cameras need light that contains all
the colors in the visible spectrum and light with a reasonable balance of all the colors to produce a satisfactory
color image.
2.3.1 Natural Light
During daytime the amount of illumination and spectral
distribution of light (color) reaching a scene depends on
the time of day and atmospheric conditions. The color
spectrum of the light reaching the scene is important if
color CCTV is being used. Direct sunlight produces the
highest-contrast scene, allowing maximum identification
of objects. On a cloudy or overcast day, less light is received
by the objects in the scene resulting in less contrast. To
produce an optimum camera picture under the wide variation in light levels (daytime to nighttime), an automaticiris camera system is required. Table 2-1 shows the light
levels for outdoor illumination under bright sun, partial
clouds, and overcast day down to overcast night.
Scene illumination is measured in foot candles (Fc)
and can vary over a range of 10,000 to 1 (or more). This
exceeds the dynamic operating range of most camera sensors for producing a good-quality video image. After the
sun has gone below the horizon and if the moon is overhead, reflected sunlight from the moon illuminates the
Video Technology Overview
23
ILLUMINATION
COMMENTS
CONDITION
(lux)
(FtCd)
10,000
107,500
1,000
10,750
OVERCAST DAY
100
1,075
VERY DARK DAY
10
DIRECT SUNLIGHT
FULL DAYLIGHT
107.5
1
TWILIGHT
DAYLIGHT
RANGE
10.75
1.075
DEEP TWILIGHT
.1
FULL MOON
.01
.1075
QUARTER MOON
.001
.01075
STARLIGHT
.0001
.001075
OVERCAST NIGHT
.00001
.0001075
LOW
LIGHT
LEVEL
RANGE
NOTE: 1 lux = .093 FtCd
Table 2-1
Light Levels under Daytime and Nighttime Conditions
scene and may be detected by a sensitive monochrome
camera. Detection of information in a scene under this
condition requires a very sensitive camera since there is
very little light reflected into the camera lens from the
scene. As an extreme, when the moon is not overhead
or is obscured by cloud cover, the only light received is
ambient light from: (1) local man-made lighting sources,
(2) night-glow caused by distant ground lighting reflecting
off particulate (pollution), clouds, and aerosols in the
lower atmosphere, and (3) direct light caused by starlight.
This is the most severe lighting condition and requires
either: (1) ICCD, (2) monochrome camera with IR LED
illumination, or (3) thermal IR camera. Table 2-2 summarizes the light levels occurring under daylight and these
LLL conditions and the operating ranges of typical cameras. The equivalent metric measure of light level (lux)
compared with the foot candle (Fc) is given. One Fc is
equivalent to approximately 9.3 lux.
2.3.2 Artificial Light
Artificial illumination is often used to augment outdoor
lighting to obtain adequate video surveillance at night.
The light sources used are: tungsten, tungsten-halogen,
metal-arc, mercury, sodium, xenon, IR lamps, and light
emitting diode (LED) IR arrays. Figure 2-11 illustrates several examples of these lamps.
The type of lighting chosen depends on architectural
requirements and the specific application. Often a particular lighting design is used for safety reasons so that personnel at the scene can see better, as well as for improving
the video picture. Tungsten and tungsten halogen lamps
have by far the most balanced color and are best for color
cameras. The most efficient visual outdoor light types are
the low- and high-pressure sodium-vapor lamps to which
the human eye is most sensitive. These lamps, however,
do not produce all colors (missing blue and green) and
therefore are not good light sources for color cameras.
Metal-arc lamps have excellent color rendition. Mercury
arc lamps provide good security illumination but are missing the color red and therefore are not as good as the
metal-arc lamps at producing excellent-quality color video
images. Long-arc xenon lamps having excellent color rendition are often used in outdoor sports arenas and large
parking areas.
Light emitting diode IR illumination arrays either
mounted in monochrome video cameras or located near
the camera are used to illuminate scenes when sufficient
lighting is not available. Since they only emit energy in
the IR spectrum they can only be used with monochrome
cameras. They are used at short ranges (10–25 feet)
with wide angle lenses (50–75 FOV) or at medium
long ranges (25–200 feet) with medium to narrow FOV
lenses (5–20 ).
Artificial indoor illumination is similar to outdoor
illumination, with fluorescent lighting used extensively
in addition to the high-pressure sodium, metal-arc
and mercury lamps. Since indoor lighting has a relatively constant light level, automatic-iris lenses are often
unnecessary. However, if the CCTV camera views a scene
near an outside window or a door where additional
light comes in during the day, or if the indoor lighting
changes between daytime and nighttime operation, then
an automatic-iris lens or electronically shuttered camera is
required. The illumination level from most indoor lighting is significantly lower by 100–1000 times than that of
sunlight. Chapter 3 describes outdoor natural and artificial
lighting and indoor man-made lighting systems available
for video surveillance use.
24
CCTV Surveillance
CAMERA REQUIREMENT PER LIGHTING CONDITIONS
ILLUMINATION
CONDITION
ILLUMINATION
(FtCd)
(lux)
OVERCAST NIGHT
.00001
.0001075
STARLIGHT
.0001
.001075
QUARTER MOON
.001
.01075
FULL MOON
.01
.1075
DEEP TWILIGHT
.1
1.075
TWILIGHT
1
10.075
VERY DARK DAY
10
107.5
OVERCAST DAY
100
1,075
1,000
10,750
10,000
107,500
FULL DAYLIGHT
DIRECT SUNLIGHT
VIDICON *
CCD
CMOS
ICCD
ISIT *
OPERATING RANGE OF
TYPICAL CAMERAS
* FOR REFERENCE ONLY
Table 2-2
Camera Capability under Natural Lighting Conditions
(A) TUNGSTEN HALOGEN
(B) FLUORESCENT
• STRAIGHT
• U
(C) HIGH PRESSURE SODIUM
(D) TUNGSTEN PAR
• SPOT
• FLOOD
(E) XENON LONG ARC
(F) HIGH INTENSITY DISCHARGE
METALARC
NOTE: PAR = PARABOLIC ALUMINIZED REFLECTOR
FIGURE 2-11
Representative artificial light sources
Video Technology Overview
2.4 SCENE CHARACTERISTICS
The quality of the video image depends on various scene
characteristics that include: (1) the scene lighting level,
(2) the sharpness and contrast of objects relative to the
scene background, (3) whether objects are in a simple,
uncluttered background or in a complicated scene, and
(4) whether objects are stationary or in motion. These
scene factors will determine whether the system will be
able to detect, determine orientation, recognize, or identify objects and personnel. As will be seen later the
scene illumination—via sunlight, moonlight, or artificial
sources—and the actual scene contrast play important
roles in the type of lens and camera necessary to produce
a quality image on the monitor.
2.4.1 Target Size
In addition to the scene’s illumination level and the
object’s contrast with respect to the scene background, the
object’s apparent size—that is, its angular FOV as seen by
the camera—influences a person’s ability to detect it. (Try
to find a football referee with a striped shirt in a field of
zebras.)
The requirements of a video system are a function of the
application. These include: (1) detection of the object or
movement in the scene; (2) determination of the object’s
orientation; (3) recognition of the type of object in the
scene, that is, adult or child, car or truck; or (4) identification of the object (Who is the person? Exactly what
kind of truck is it?). Making these distinctions depends on
the system’s resolution, contrast, and signal-to-noise ratio
(S/N). In a typical scene the average observer can detect
a target about one-tenth of a degree in angle. This can be
related to a standard video picture that has 525 horizontal lines (NTSC) and about 350 TV line vertical and 500
TV line horizontal resolution. Figure 2-12 and Table 2-3
summarize the number of lines required to detect, orient, recognize, or identify an object in a television picture. The number of TV lines required will increase for
conditions of poor lighting, highly complex backgrounds,
reduced contrast, or fast movement of the camera or
target.
2.4.2 Reflectivity
The reflectivity of different materials varies greatly depending on its composition and surface texture. Table 2-4 gives
DETECTION
1 TV LINE
ORIENTATION
2 TV LINES
RECOGNITION
5 TV LINES
IDENTIFICATION
7 TV LINES
NOTE: 1 TV LINE (BRIGHT AND DARK LINE) = 1 LINE PAIR
FIGURE 2-12
Object size vs. intelligence obtained
25
26
CCTV Surveillance
INTELLIGENCE
MINIMUM *
TV LINES
1 ± 0.25
DETECTION
ORIENTATION
1.4 ± 0.35
RECOGNITION
4 ± 0.8
6.4 ± 1.5
IDENTIFICATION
* ONE TV LINE CORRESPONDS TO A LIGHT AND
DARK LINE (ONE TV LINE PAIR)
Table 2-3 TV Lines vs. Intelligence
Obtained
some examples of materials and objects viewed by video
cameras and their respective reflectivities.
Since the camera responds to the amount of light
reflected from the scene it is important to recognize that
objects have a large range of reflectivities. The objects with
the highest reflectivities produce the brightest images. To
detect one object located within the area of another the
objects must differ in reflectivity, color, or texture. Therefore, if a red box is in front of a green wall and both
have the same reflectivity and texture, the box will not
be seen on a monochrome video system. In this case, the
total reflectivity in the visible spectrum is the same for the
green wall and the red box. This is where the color camera
shows its advantage over the monochrome camera.
The case of a color scene is more complex. While the
reflectivity of the red box and the green wall may be the same
as averaged over the entire visible spectrum from blue to red,
the color camera can distinguish between green and red.
It is easier to identify a scene characteristic by a difference in color in a color scene than it is to identify it by
a difference in gray scale (intensity) in a monochrome
scene. For this reason the target size required to make an
identification in a color scene is generally less than it is to
make the same identification in a monochrome scene.
2.4.3 Effects of Motion
A moving object in a video image is easier to detect, but
more difficult to recognize than a stationary one provided that the camera can respond to it. Low light level
cameras produce sharp images for stationary scenes but
smeared images for moving targets. This is caused by a
phenomenon called “lag” or “smear.” Solid-state sensors
(CCD, CMOS, and ICCD) do not exhibit smear or lag
at normal light levels and can therefore produce sharp
images of both stationary and moving scenes. Some image
intensifiers exhibit smear when the scene moves fast or
when there is a bright light in the FOV of the lens.
When the target in the scene moves very fast the inherent camera scan rate (30 frames per second) causes a
blurred image of this moving target in the camera. This is
analogous to the blurred image in a still photograph when
the shutter speed is too slow for the action. There is no
cure for this as long as the standard NTSC (National Television System Committee) television scan rate (30 frames
per second) is used. However, CCTV snapshots can be
MATERIAL
REFLECTIVITY (%) *
SNOW
ASPHALT
85–95
5
PLASTER (WHITE)
90
SAND
40–60
TREES
20
GRASS
40
CLOTHES
15–30
CONCRETE-NEW
40
CONCRETE-OLD
CLEAR WINDOWS
25
70
HUMAN FACE
15–25
WOOD
10–20
PAINTED WALL (WHITE)
75–90
RED BRICK
25–35
PARKING LOT AND AUTOMOBILES
40
ALUMINUM BUILDING (DIFFUSE)
65–70
* VISIBLE SPECTRUM: 400–700 NANOMETERS
Table 2-4
Reflectivity of Common Materials
Video Technology Overview
taken without any blurring using fast-shuttered CCD cameras. For special applications in which fast-moving targets
must be imaged and tracked, higher scan rate cameras are
available.
27
provide low resolution and low identification capabilities.
Narrow-FOV or telephoto lenses have high magnification,
with high resolution and high identification capabilities.
2.5.2 Zoom Lens
2.4.4 Scene Temperature
Scene temperature has no effect on the video image in a
CCD, CMOS, or ICCD sensor. These sensors do not respond
to temperature changes or temperature differences in the
scene. On the other hand, IR thermal imaging cameras
do respond to temperature differences and changes in
temperature in the scene. Thermal imagers do not respond
to visible light or the very near-IR radiation like that produced by IR LEDs. The sensitivity of IR thermal imagers is
defined as the smallest change in temperature in the scene
that can be detected by the thermal camera.
The zoom lens is more versatile and complex than the FFL
lens. Its FL is variable from wide-angle to narrow-angle
(telephoto) FOV (Figure 2-14).
The overall camera/lens FOV depends on the lens FL
and the camera sensor size as shown in Figure 2-14. Zoom
lenses consist of multiple lens groups that are moved
within the lens barrel by means of an external zooming
ring (manual or motorized), thereby changing the lens
FL and angular FOV without having to switch lenses or
refocusing. Zoom focal length ratios can range from 6 to
1 up to 50 to 1. Zoom lenses are usually large and used
on pan/tilt mounts viewing over large areas and distances
(25–500 feet).
2.5 LENSES
A lens collects reflected light from the scene and focuses
it onto the camera image sensor. This is analogous to the
lens of the human eye focusing a scene onto the retina
at the back of the eye (Figure 2-6). As in the human eye,
the camera lens inverts the scene image on the image
sensor, but the eye and the camera electronics compensate
(invert the image) to perceive an upright scene. The retina
of the human eye differs from any CCTV lens in that it
focuses a sharp image only in the central 10% of its total
160 FOV. All vision outside the central focused scene is
out of focus. This central imaging part of the human eye
can be characterized as a medium FL lens: 16–25 mm. In
principle, Figure 2-6 represents the function of any lens
in a video system.
Many different lens types are used for video surveillance
and safety applications. They range from the simplest
FFL manual-iris lenses to the more complex variable-focallength (vari-focal) and zoom lenses, with an automatic iris
being an option for all types.
In addition, pinhole lenses are available for covert applications, split-image lenses for viewing multiple scenes on
one camera, right-angle lenses for viewing a scene perpendicular to the camera axis, and rigid or flexible fiber-optic
lenses for viewing through thick walls, under doors, etc.
2.5.1 Fixed-Focal-Length Lens
Figure 2-13 illustrates three fixed focal length (FFL) or
fixed FOV lenses with narrow (telephoto), medium, and
wide FOVs and the corresponding FOV obtained when
used with a 1/3-inch camera sensor format.
Wide-FOV (short FL) lenses permit viewing a very large
scene (wide angle) with low magnification and therefore
2.5.3 Vari-Focal Lens
The vari-focal lens is a variable focal length lens used in
applications where a FFL lens would be used. In general
they are smaller and cost much less than zoom lenses.
Like the zoom lens, the vari-focal lens is used because its
focal length (angular FOV) can be changed manually or
automatically, using a motor, by rotating the barrel on the
lens. This feature makes it convenient to adjust the FOV
to a precise angle when installed on the camera. Typical
vari-focal lenses have focal lengths of 3–8 mm, 5–12 mm,
8–50 mm. With just these three lenses focal lengths of from
3 to 50 mm (91–5 horizontal FOV) can be covered on
a 1/3-inch format sensor. Unlike zoom lenses, vari-focal
lenses must be refocused each time the FL and the FOV
are changed. They are not suitable for zoom or pan/tilt
applications.
2.5.4 Panoramic—360 Lens
There has always been a need to see “all around,” i.e.
an entire room or other location, seeing 360 with one
panoramic camera and lens. In the past, 360 FOV camera
viewing systems have only been achieved by using multiple
cameras and lenses and combining the scenes on a splitscreen monitor.
Panoramic lenses have been available for many years but
have only recently been combined with digital electronics and sophisticated mathematical transformations to take
advantage of their capabilities. Figure 2-15 shows two lenses
having a 360 horizontal FOV and a 90 vertical FOV.
The panoramic lens collects light from the 360
panoramic scene and focuses it onto the camera sensor as
28
CCTV Surveillance
TELEPHOTO (NARROW ANGLE)
2.5 –15°
2.5 TO 15°
FL = 150 TO 25 mm
*
NORMAL
15 TO 45°
FL = 25 TO 8 mm
WIDE ANGLE
45 TO 85°
FL = 8 TO 2.1 mm
*
*
15–45°
45–85°
* SENSOR FORMAT:1/2"
FIGURE 2-13
Representative FFL lenses and their fields of view (FOV)
CAMERA
SENSOR
10.5–105 mm FL
ZOOM LENS
WIDE
ANGLE
10.5 mm FL
SENSOR
FORMAT
HORIZONTAL
FOV (DEGREES)
WIDE
10.5 mm
NARROW
105 mm
1/4"
18.6
2.0
1/3"
24.8
2.6
1/2"
33.0
3.5
2/3"
45.5
4.8
FIGURE 2-14
Zoom video lens horizontal field of view (FOV)
NARROW
ANGLE
105 mm
Video Technology Overview
(A)
FIGURE 2-15
29
(B)
Panoramic 360 lens
a donut-shaped image. The electronics and mathematical
algorithm converts this donut-shaped panoramic image
into the rectangular (horizontal and vertical) format for
normal monitor viewing (Section 2.6.5).
has been labeled a pinhole lens. Figure 2-16 shows examples of straight and right-angle pinhole lenses used with C
or CS mount cameras. The very small mini-pinhole lenses
are used on the low-cost, small board cameras.
2.5.5 Covert Pinhole Lens
2.5.6 Special Lenses
This special security lens is used when the lens and
CCTV camera must be hidden. The front lens element or
aperture is small (from 1/16 to 5/16 of an inch in diameter). While this is not the size of a pinhead it nevertheless
(A) PINHOLE LENSES
FIGURE 2-16
Pinhole and mini-pinhole lenses
Some special lenses useful in security applications
include split-image, right-angle, relay, and fiber optic
(Figure 2-17).
(B) MINI-LENSES
30
CCTV Surveillance
(A) DUAL SPLIT IMAGE LENS
(B) TRI SPLIT IMAGE LENS
(C) RIGHT ANGLE LENS
(D) RIGID FIBER OPTICS
(E) RELAY LENS
(F) FLEXIBLE FIBER OPTICS
FIGURE 2-17
Special video lenses
The dual-split and tri-split lenses use only one camera
to produce multiple scenes. These are useful for viewing
the same scene with different magnifications or different
scenes with the same or different magnifications. Using
only one camera can reduce cost and increases reliability. These lenses are useful when two or three views are
required and only one camera was installed.
The right-angle lens permits a camera using a wideangle lens installed to view a scene that is perpendicular to
the camera’s optical axis. There are no restrictions on the
focal lengths so they can be used in wide- or narrow-angle
applications.
The flexible and rigid coherent fiber-optic lenses are
used to mount a camera several inches to several feet away
from the front lens as might be required to view from the
opposite side of a wall or in a hazardous environment. The
function of the fiber-optic bundle is to transfer the focused
visual image from one location to another. This may be
useful for: (1) protecting the camera, and (2) locating the
lens in one environment (outdoors) and the camera in
another (indoors).
2.6 CAMERAS
The camera lens focuses the visual scene image onto
the camera sensor area point-by-point and the camera
electronics transforms the visible image into an electrical
signal. The camera video signal (containing all picture
information) is made up of frequencies from 30 cycles per
second, or 30 hertz (Hz), to 4.2 million cycles per second,
or 4.2 megahertz (MHz). The video signal is transmitted
via a cable (or wireless) to the monitor display.
Almost all security cameras in use today are color or
monochrome CCD with the rapid emergence of CMOS
types. These cameras are available as low-cost single
printed circuit board (PCB) cameras with small lenses
already built in, with or without a housing used for covert
and overt surveillance applications. More expensive cameras in a housing are larger and more rugged and have
a C or CS mechanical mount for accepting any type of
lens. These cameras have higher resolution and light sensitivity and other electrical input/output features suitable
for multiple camera CCTV systems. The CCD and CMOS
Video Technology Overview
cameras with LED IR illumination arrays can extend the
use of these cameras to nighttime use. For LLL applications, the ICCD and IR cameras provide the highest
sensitivity and detection capability.
Significant advancements in camera technology have
been made in the last few years particularly in the use of
digital signal processing (DSP) in the camera, and development of the IP camera. All security cameras manufactured between the 1950s and 1980s were the vacuum tube
type, either vidicon, silicon, or LLL types using silicon
intensified target (SIT) and intensified SIT (ISIT). In the
1980s the CCD and CMOS solid-state video image sensors
were developed and remain the mainstay in the security
industry. Increased consumer demand for video recorders
using CCD sensors in camcorders and the CMOS sensor in
digital still frame cameras caused a technology explosion
and made these small, high resolution, high sensitivity,
monochrome and color solid-state cameras available for
security systems.
The security industry now has at its disposal both analog
and digital surveillance cameras. Up until the mid-1990s
analog cameras dominated, with only rare use of DSP electronics, and the digital Internet camera was only being
31
introduced to the security market. Advances in solid-state
circuitry, the demand from the consumer market and the
availability of the Internet were responsible for the rapid
use of digital cameras for security applications.
2.6.1 The Scanning Process
Two methods used in the camera and monitor video scanning process are raster scanning and progressive scanning.
In the past, analog video systems have all used the raster
scanning technique, however, newer digital systems are
now using progressive scanning. All cameras use some
form of scanning to generate the video picture. A block
diagram of the CCTV camera and a brief description of
the analog raster scanning process and video signal are
shown in Figures 2-8, 2-9, 2-18, and 2-19.
The camera sensor converts the optical image from
the lens into an electrical signal. The camera electronics
process the video signal and generate a composite video
signal containing the picture information (luminance and
color) and horizontal and vertical synchronizing pulses.
Signals are transmitted in what is called a frame of picture
SENSOR
TYPEWRITER ANALOGY
SCENE
PAGE 1
FIELD 1
SCANNING RASTER PATTERN
LENS
LINE 1
LINE 3
LINE 5
2
1
1 = 1ST FIELD
LINE 262 V2
(1/60 sec)
2 = 2ND FIELD
PAGE 2
FIELD 2
(1/60 sec)
LINE 2
LINE 4
LINE 6
1 + 2 = FRAME
(1/30 sec)
2
1
LINE 525
1 SCAN LINE
VIDEO
SIGNAL
(VOLTS)
1
1/30 sec
62.5 MICROSECONDS
(1 FRAME = 525 SCAN LINES)
WHITE LEVEL
PICTURE INFORMATION
BLACK LEVEL
–0.4
1/60 sec
(1 FIELD = 262 1/2 SCAN LINES)
FIGURE 2-18
Analog video scanning process and video display signal
SYNC
PULSES
TIME
32
CCTV Surveillance
video, made up of two fields of information. Each field is
transmitted in 1/60 of a second and the entire frame in
1/30 of a second, for a repetition rate of 30 frames per
second (fps). In the United States, this format is the Electronic Industries Association (EIA) standard called the
NTSC (National Television System Committee) system.
The European standard uses 625 horizontal lines with a
field taking 1/50 of a second and a frame 1/25 of a second
and a repetition rate of 25 fps.
2.6.1.1 Raster Scanning
In the NTSC system the first picture field is created by scanning 2621/2 horizontal lines. The second field of the frame
contains the second 2621/2 lines, which are synchronized
so that they fall between the gaps of the first field lines
thus producing one completely interlaced picture frame
containing 525 lines. The scan lines of the second field fall
exactly halfway between the lines of the first field resulting
in a 2-to-1 interlace system. As shown in Figure 2-18 the first
field starts at the upper-left corner (of the camera sensor
or the CRT monitor) and progresses down the sensor (or
screen), line by line, until it ends at the bottom center of
the scan.
Likewise the second field starts at the top center of
the screen and ends at the lower-right corner. Each time
one line in the field traverses from the left side of the
scan to the right it corresponds to one horizontal line as
shown in the video waveform at the bottom of Figure 2-18.
The video waveform consists of negative synchronization
pulses and positive picture information. The horizontal
and vertical synchronization pulses are used by the video
monitor (and VCR, DVR, or video printer) to synchronize
the video picture and paint an exact replica in time and
intensity of the camera scanning function onto the monitor face. Black picture information is indicated on the
waveform at the bottom (approximately 0 volts) and the
white picture information at the top (1 volt). The amplitude of a standard NTSC signal is 1.4 volts peak to peak.
In the 525-line system the picture information consists of
approximately 512 lines. The lines with no picture information are necessary for vertical blanking, which is the
time when the camera electronics or the beam in the monitor CRT moves from the bottom to the top to start a new
field.
Random-interlace cameras do not provide complete synchronization between the first and the second fields. The
horizontal and the vertical scan frequencies are not locked
together and therefore fields do not interlace exactly.
This condition, however, results in an acceptable picture,
and the asynchronous condition is difficult to detect. The
2-to-1 interlace system has an advantage when multiple
cameras are used with multiple monitors and/or recorders
in that they prevent jump or jitter when switching from
one camera to the next.
The scanning process for solid-state cameras is different. The solid-state sensor consists of an array of very
small picture elements (pixels) that are read out serially
(sequentially) by the camera electronics to produce the
same NTSC format—525 TV lines in 1/30 of a second
(30 fps)—as shown in Figure 2-19.
The use of digital cameras and digital monitors has
changed the way the camera and monitor signals are processed, transmitted, and displayed. The final presentation
on the monitor looks similar to the analog method but
instead of seeing 525 horizontal lines (NTSC system),
individual pixels are seen in a row and column format.
In the digital system the camera scene is divided into
rows and columns of individual pixels (small points in
the scene) each representing the light intensity and color
for each point in the scene. The digitized scene signal is
transmitted to the digital display be it LCD, plasma, or
other, and reproduced on the monitor screen pixel-bypixel providing a faithful representation of the original
scene.
2.6.1.2 Digital and Progressive Scan
The digital scanning is accomplished in either the 2-to-1
interlace mode as in the analog system, or in a progressive mode. In the progressive mode each line is scanned
in linear sequence: line 1, then line 2, line 3, etc. Solidstate camera sensors and monitor displays can be manufactured with a variety of horizontal and vertical pixels
formats. The standard aspect ratio is 4:3 as in the analog
system, the wide-screen 16:9, and others are used. Likewise there are many different combinations of the number of pixels in the sensor and display available. Some
standard formats for color CCD cameras are 512 h × 492 v
for 330 TV line resolution and 768 h × 494 v for 480 TV
line resolution, and for color LCD monitors is 1280 h ×
1024 v.
2.6.2 Solid-State Cameras
Video security cameras have gone through rapid technological change during the last half of the 1980s to the
present. For decades the vidicon tube camera was the only
security camera available. In the 1980s the more sensitive
and rugged silicon-diode tube camera was the best available. In the late 1980s the invention and development of
the digital CCD and later the CMOS cameras replaced
the tube camera. This technology coincided with rapid
advancement in DSP in cameras, the IP camera, and use
of digital transmission of the video signal over local and
wide area networks and the Internet.
The two generic solid-state cameras accounting for
most security applications are the CCD and the CMOS.
Video Technology Overview
33
PROGRESSIVE SCAN DISPLAY
LINE 1
2
PROGRESSIVE SCAN CAMERA
3
LINE 1
2
3
525
525
512 h × 492 v
768 h × 494 v
330 TVLines
480 TVLines
TYPICAL STANDARD FORMATS
VGA
XGA
SXGA
640 h × 480 v
1024 h × 768 v
1280 h × 1024 v
VIDEO
SIGNAL
LINE 1
FIGURE 2-19
LINE 2
LINE 525
LINE 3
Digital and progressive scanning process and video display signal
The first generation of solid-state cameras available from
most manufacturers had 2/3-inch (sensor diagonal) and
1/2-inch sensor formats. As the technology improved,
smaller formats evolved. Most solid-state cameras in use
today are available in three image sensor formats: 1/2-,
1/3-, and 1/4-inch. The 1/2-inch format produces higher
resolution and sensitivity at a higher cost. The 1/2-inch
and smaller formats permitted the use of smaller, less
expensive lenses as compared with the larger formats.
Many manufacturers now produce 1/3-inch and 1/4-inch
format cameras with excellent resolution and light sensitivity. Solid-state sensor cameras are superior to their predecessors because of their: (1) precise, repeatable pixel
geometry, (2) low power requirements, (3) small size, (4)
excellent color rendition and stability, and (5) ruggedness
and long life expectancy. At present, solid-state cameras
have settled into three main categories: (1) analog, (2)
digital, and (3) Internet.
2.6.2.1 Analog
Analog cameras have been with the industry since CCTV
has been used in security. Their electronics are straightforward and the technology is still used in many applications.
2.6.2.2 Digital
Since the second half of 1990s there has been an increased
use of DSP in cameras. It significantly improves the performance of the camera by: (1) automatically adjusting to
large light level changes (eliminating the automatic-iris),
(2) integrating the VMD into the camera, and (3) automatically switching the camera from color operation to
higher sensitivity monochrome operation, as well as other
features and enhancements.
2.6.2.3 Internet
The most recent camera technology advancement is manifest in the IP camera. This camera is configured with
electronics that connects to the Internet, WWW network
through an Internet service provider (ISP). Each camera
is provided with a registered Internet address and can
transmit the video image anywhere on the network. This
is really remote video monitoring at its best! The camera
site is viewed from anywhere by entering the camera Internet address (ID number) and proper password. Password
security is used so that only authorized users can enter the
website and view the camera image. Two-way communication is used so that the user can control camera parameters
and direct the camera operation (pan, tilt, zoom, etc.)
from the monitoring site.
34
CCTV Surveillance
2.6.3 Low-Light-Level Intensified Camera
When a security application requires viewing during nighttime conditions where the available light is moonlight,
starlight, or other residual reflected light, and the surveillance must be covert (no active illumination like IR LEDs),
LLL intensified CCD cameras are used. The ICCD cameras have sensitivities between 100 and 1000 times higher
than the best solid-state cameras. The increased sensitivity
is obtained through the use of a light amplifier mounted
in between the lens and the CCD sensor. LLL cameras
cost between 10 and 20 times more than CCD cameras.
Chapter 19 describes the characteristics of these cameras.
2.6.4 Thermal Imaging Camera
An alternative to the ICCD camera is the thermal IR
camera. Visual cameras see only visible light energy
from the blue end of the visible spectrum to the
red end (approximately 400–700 nanometers). Some
monochrome cameras see beyond the visible region into
the near-IR region of the spectrum up to 1000 nanometers
(nm). This IR energy, however, is not thermal IR energy.
Thermal IR cameras using thermal sensors respond to
thermal energy in the 3–5 micrometer (m) and 8–14 m
range. The IR sensors respond to the changes in heat (thermal) energy emitted by the targets in the scene. Thermal
imaging cameras can operate in complete darkness. They
require no visible or IR illumination whatever. They are
truly passive nighttime monochrome imaging sensors.
They can detect humans and any other warm objects (animals, vehicle engines, ships, aircraft, warm/hot spots in
buildings) or other objects against a scene background.
2.6.5 Panoramic 360 Camera
Powerful mathematical techniques combined with the
unique 360 panoramic lens (see Section 2.5.4) have made
possible a 360 panoramic camera. In operation the lens
collects and focuses the 360 horizontal by up to 90 vertical scene (one-half of a sphere, a hemisphere) onto the
camera sensor. The image takes the form of a “donut” on
the sensor (Figure 2-20).
The camera/lens is located at the origin (0). The scene
is represented by the surface of the hemisphere. As shown,
a small part (slice) of the scene area (A,B,C,D) is “mapped”
onto the sensor as a,b,c,d. In this way the full scene is
mapped onto the sensor. Direct presentation of the donutring video image onto the monitor does not result in a
useful picture to work with. That is where the use of a
RAW DONUT IMAGE
FROM CAMERA SENSOR
360 HORIZONTAL, 90 VERTICAL
360° PANORAMIC CAMERA
360°
HORIZ. FOV
A
IMAGE TRANSFORMATION:
DONUT TO RECTANGULAR
CD
0
B
TYP.
SLICE
90°
VERT. FOV
B
A
D C
DISPLAY CONFIGURATION DRIVER
RECTANGULAR:
HALF SPLIT
A
B
D
C
0°
180°
LENS SEES FULL HEMISPHERE:
360° × 180°
RECTANGULAR:
4-WAY SPLIT
MIXED:
RECTANGULAR
AND DONUT
DONUT: RAW
SENSOR IMAGE
180°
CD
360°
B
FOUR TYPICAL DISPLAY FORMATS
FIGURE 2-20
Panoramic 360 camera
A
Video Technology Overview
powerful mathematical algorithm comes in. Digital processing in the computer using the algorithm transforms
the donut-shaped image into the normal format seen on
a monitor, i.e. horizontal and vertical.
All of the 0 to 360 horizontal by 90 vertical images cannot be presented on a monitor in a useful way – there is just
too much picture “squeezed” into the small screen area.
This condition is solved by computer software by looking
at only a section of the entire scene at any particular time.
The main attributes of the panoramic system are:
(1) captures a full 360 FOV, (2) can digitally pan/tilt to
anywhere in the scene and digitally zoom any scene area,
(3) has no moving parts (no motors, etc. that can wear
out), and (4) multiple operators can view any part of the
scene in real-time or at a later time.
The panoramic camera requires a high resolution camera since so much scene information is contained in the
image. Camera technology has progressed so that these
digital cameras are available and can present a good image
of a zoomed-in portion of the panoramic scene.
35
the monitor may be from tens of feet to many miles or perhaps completely around the globe. The transmission path
may be inside buildings, outside buildings, above ground,
under ground, through the atmosphere, or in almost any
environment imaginable. For this reason the transmission
means must be carefully assessed and an optimum choice
of hardware made to satisfactorily transmit the video signal
from the camera to the monitoring site. There are many
ways to transmit the video signal from the camera to the
monitoring site. Figure 2-21 shows some examples of transmission cables.
The signal can be analog or digital. The signal can be
transmitted via electrical conductors using coaxial cable or
UTP, by fiber optic, by LAN or WAN, intranet or Internet.
Particular attention should be paid to transmission
means when transmitting color video signals since the
color signal is significantly more complex and susceptible
to distortion than monochrome. Chapters 6 and 7 describe
and analyze the characteristics, advantages, and disadvantages of all of the transmission means and the hardware
available to transmit the video signal.
2.7 TRANSMISSION
2.7.1 Hard-Wired
By definition, the camera must be remotely located from
the monitor and therefore the video signal must be transmitted by some means from one location to another. In
security applications, the distance between the camera and
There are several hard-wired means for transmitting a
video signal, including coaxial cable, UTP, LAN, WAN,
intranet, Internet, and fiber-optic cable. Fiber-optic cable
COAXIAL
CABLE
COPPER CONDUCTOR
PROTECTIVE OUTER
JACKET
COPPER SHEATH
INSULATED
TWO WIRE
TX
REC
UNSHIELDED
TWISTED
PAIR (UTP)
TX
REC
TX = TRANSMITTER
REC = RECEIVER
FIBER-OPTIC
CABLE
GLASS FIBER
DIELECTRIC
INSULATOR
FIGURE 2-21
STRENGTHENING
MEMBER
Hard wired copper and fiber-optic transmission means
PROTECTIVE OUTER
JACKET
36
CCTV Surveillance
is used for long distances and when there is interfering
electrical noise. Local area networks and Internet connections are digital transmission techniques used in larger
security systems and where the signal must be transmitted
over existing computer networks or over long distances.
2.7.1.1 Coaxial Cable
The most common video signal transmission method is the
coaxial cable. This cable has been used since the inception
of CCTV and continues to be used to this day. The cable is
inexpensive, easy to terminate at the camera and monitor
ends, and transmits a faithful video signal with little or
no distortion or loss. This cable has a 75 ohm electrical
impedance which matches the impedance of the camera
and monitor insuring a distortion-free video image. This
coaxial cable has a copper electrical shield and center
conductor works well over distances up to 1000 feet.
2.7.1.2 Unshielded Twisted Pair
In the 1990s unshielded twisted pair (UTP) video transmission came into vogue. The technique uses a transmitter at the camera and a receiver at the monitor with
two twisted copper wires connecting them. Several reasons for its increased popularity are: (1) can be used over
longer distances than coaxial cable, (2) uses inexpensive
wire, (3) many locations already have two-wire twistedpair installed, (4) low-cost transmitter and receiver, and
(5) higher electrical noise immunity as compared to coaxial cable. The UTP using a sophisticated electronic transmitter and receiver can transmit the video signal up to
2000–3000 feet.
2.7.1.3 LAN, WAN, Intranet and Internet
The evolution of the LAN, WAN, intranet and Internet revolutionized the transmission of video signals in
a new form—digital—which significantly expanded the
scope and effectiveness of video for security systems. The
widespread use of business computers and consequent use
of these networks provided an existing digital network protocol and communications suitable for video transmission.
The Internet and WWW attained widespread use in the
late 1990s and truly revolutionized digital video transmission. This global computer network provided the digital
backbone path to transmit digital video, audio, and command signals from anywhere on the globe.
The video signal transmission techniques described so
far provide a means for real-time transmission of a video
signal, requiring a full 4.2 MHz bandwidth to reproduce
real-time motion. When these techniques cannot be used
for real-time video, alternative digital techniques are used.
In these systems, a non-real-time video transmission takes
place, so that some scene action is lost. Depending on the
action in the scene, the resolution, from near real-time
(15 fps.) to slow-scan (a few frames/sec) of the video image
are transmitted. The digitized and compressed video signal
is transmitted over a LAN or Internet network and decompressed and reconstructed at the receiver/monitoring site.
2.7.2 Wireless
In legacy analog video surveillance systems, it is often more
economical or beneficial to transmit the real-time video
signal without cable—wireless—from the camera to the
monitor using a radio frequency (RF) or IR atmospheric
link. In digital video systems using digital transmission, the
use of wireless networks (WiFi) permits routing the video
and control signals to any remote location. In both the
analog and the digital systems some form of video scrambling or encryption is often used to remove the possibility
of eavesdropping by unauthorized personnel outside the
system. Three important applications for wireless transmission are: (1) covert and portable rapid deployment
video installations, (2) building-to-building transmission
over a roadway, and (3) parking lot light poles to building.
The Federal Communications Commission (FCC) restricts
some wireless transmitting devices using microwave frequencies or RF to government and law enforcement use
but has given approval for many RF and microwave transmitters for general security use. These FCC approved
devices operate above the normal television frequency
bands at approximately 920 MHz, 2.4 GHz, and 5.8 GHz.
The atmospheric IR link is used when a high security link
is required. This link does not require an FCC approval
and transmits a video image over a narrow beam of visible light or near-IR energy. The beam is very difficult to
intercept (tap). Figure 2-22 illustrates some of the wireless
transmission techniques available today.
2.7.3 Fiber Optics
Fiber-optic transmission technology has advanced significantly in the last 5–10 years and represents a highly
reliable, secure means of transmission. Fiber-optic transmission holds several significant advantages over other
hard-wired systems: (1) very long transmission paths up
to many miles without any significant degradation in the
video signal with monochrome or color, (2) immunity to
external electrical disturbances from weather or electrical
equipment, (3) very wide bandwidth, permitting one or
more video, control, and audio signals to be multiplexed
on a single fiber, and (4) resistance to tapping (eavesdropping) and therefore a very secure transmission means.
While the installation and termination of fiber-optic
cable requires a more skilled technician, it is well within
the capability of qualified security installers. Many hardwired installations requiring the optimum color and resolution rendition use fiber-optic cable.
Video Technology Overview
INFRARED TRANSMISSION
VIDEO
MONITOR
VIDEO
FROM
CAMERA
LENS
POWER
INPUT
37
LENS
POWER
INPUT
MICROWAVE TRANSMISSION
VIDEO TO
MONITOR
VIDEO
FROM
CAMERA
POWER
INPUT
POWER
INPUT
RF TRANSMISSION
*DIPOLE
*YAGGI ANTENNA
FOR DIRECTIONAL
LONGER RANGE
TRANSMISSION
VIDEO
TO
MONITOR
VIDEO
FROM
CAMERA
POWER
INPUT
POWER
INPUT
FIGURE 2-22
RF, microwave and IR video transmission links
2.8 SWITCHERS
2.8.2 Microprocessor-Controlled
The video switcher accepts video signals from many different video cameras and connects them to one or more
monitors or recorders. Using manual or automatic activation or an alarming signal input, the switcher selects
one or more of the cameras and directs its video signal
to a specified monitor, recorder, or some other device or
location.
When the security system requires many cameras in various
locations with multiple monitors and other alarm input
functions, a microprocessor-controlled switcher and keyboard is used to manage these additional requirements
(Figure 2-24).
In large security systems the switcher is microprocessor controlled and can switch hundreds of cameras to
dozens of monitors, recorders, or video printers via an
RS-232 or other communication control link. Numerous
manufacturers make comprehensive keyboard-operated,
computer-controlled consoles that integrate the functions
of the switcher, pan/tilt pointing, automatic scanning,
automatic preset pointing for pan/tilt systems, and many
other functions. The power of the software-programmable
console resides in its flexibility, expandability, and ability to accommodate a large variety of applications and
changes in facility design. In place of a dedicated hardware
system built for each specific application this computercontrolled system can be configured via software for the
application. Chapter 11 describes types of switchers and
their functions and applications.
2.8.1 Standard
There are four basic switcher types: manual, sequential,
homing, and alarming. Figure 2-23 shows how these are
connected into the video security system.
The manual switcher connects one camera at a time to
the monitor, recorder, or printer. The sequential switcher
automatically switches the cameras in sequence to the
output device. The operator can override the automatic
sequence with the homing sequential switcher. The alarming switcher connects the alarmed camera to the output
device automatically, when an alarm is received.
38
CCTV Surveillance
INPUT
OUTPUT
C
INPUT
C
C
MANUAL
SWITCHER
C
M
C
DVR/VCR
C
C
C
.
.
.
.
.
.
(0–32)
(0–32)
C
C
C
HOMING
SWITCHER
C
M
C
DVR/VCR
C
C
SEQUENTIAL
SWITCHER
M
ALARMING
SWITCHER
M
DVR/VCR
DVR/VCR
C
.
.
.
(0–32)
SYMBOLS:
C
= CAMERA
M
= MONITOR
DVR/VCR
FIGURE 2-23
OUTPUT
ALARM 1
ALARM 2
ALARM 3
ALARM 4
= DIGITAL VIDEO RECORDER
VIDEO CASSETTE RECORDER
Basic video switcher types
2.9 QUADS AND MULTIPLEXERS
A quad or a multiplexer is used when multiple camera scenes need to be displayed on one video monitor.
It is interposed between the cameras and the monitor,
accepts multiple camera inputs, memorizes the scenes
from each camera, compresses them, and then displays
multiple scenes on a single video monitor. Equipment is
available to provide 2, 4, 9, 16, and up to 32 separate video
scenes on one single monitor. Figure 2-25 shows a block
diagram of quad and multiplexer systems.
The most popular presentation is the quad screen showing four pictures. This presentation significantly improves
camera viewing ability in multi-camera systems, decreases
security guard fatigue, and requires three fewer monitors
in a four-camera system. There is a loss of resolution when
more than one scene is presented on the monitor with
resolution decreasing as the number of scenes increases.
One-quarter of the resolution of a full screen is obtained
on a quad display (half in horizontal and half in vertical). Quads and multiplexers have front panel controls so
that: (1) a full screen image of a camera can be selected,
(2) multiple cameras can be displayed (quad, 9, etc.), or
(3) the full screen images of all cameras can be sequentially switched with dwell times for each camera, set by
the operator. Chapter 12 describes video quads and multiplexers in detail.
2.10 MONITORS
Video monitors can be divided into several categories: (1) monochrome, (2) color, (3) CRT, (4) LCD,
(5) plasma, and (6) computer display. Contrary to a popular misconception, larger video monitors do not necessarily have better picture resolution or the ability to increase
the amount of intelligence available in the picture. All
US NTSC security monitors have 525 horizontal lines—
regardless of their size or whether they are monochrome
or color; therefore the vertical resolution is about the
same regardless of the CRT monitor size. The horizontal
resolution is determined by the system bandwidth. With
the NTSC limitation the best picture quality is obtained
by choosing a monitor having resolution equal to or better than the camera or transmission link bandwidth. With
the use of a higher resolution computer monitor and corresponding higher resolution camera and commensurate
bandwidth to match, higher resolution video images are
Video Technology Overview
39
ALARM 1
ALARM 2
ALARM 3
ALARM 4
INPUT
OUTPUT
(0–2048)
M
INPUT
C1
M
C2
M
MATRIX
M
SWITCHER
(0–128)
C3
C4
(0–2048)
DVR/VCR
RS232, RS485
COMMUNICATIONS
PORT
VP
(0–32)
SYMBOLS:
KEYBOARD
C
= CAMERA
M
= MONITOR
=
DVR/VCR
= VIDEO PRINTER
VP
FIGURE 2-24
Microprocessor controlled switcher and keyboard
QUAD SYSTEM
3
2
MULTIPLEXER SYSTEM
SCENE 3
1
SCENE 2
4
1
DIGITAL VIDEO RECORDER
VIDEO CASSETTE RECORDER
SCENE 4
SCENE 1
1
2
3
4
1
2
3
4
4
1
1
2
3
4
2
3
SCENE 3
16
SCENE 4
SCENE 16
2
3
4
16
SELECT
SEQUENCE
1
3
SCENE 2
SELECT
QUAD
MENU
2
SCENE 1
4
MONITOR DISPLAYS QUAD PICTURES OR ANY
INDIVIDUAL SCENES IN THREE MODES:
QUAD – FOUR COMPRESSED PICTURES
SELECT – ONE FULL PICTURE
SEQUENCE THROUGH 4 SCENES
MENU
1
2
3
4
QUAD
ANY SCENE
FULL SCREEN
1
1
9
NINE
1
2
3
4
SEQUENCE
16
SIXTEEN
1
1
9
MONITOR DISPLAYS 1, 4, 9, 16 PICTURES
OR ANY INDIVIDUAL SCENE – FULL SCREEN
FIGURE 2-25
Quad and multiplexer block diagrams
16
40
CCTV Surveillance
(A) TRIPLE 5"
(B) DUAL 9"
(C) LCD
FIGURE 2-26
(D) PLASMA
Standard 5- and 9-inch single/multiple CRT, LCD and plasma monitors
obtained. Chapter 8 gives more detailed characteristics
of monochrome and color monitors used in the security industry. Figure 2-26 shows representative examples of
video monitors.
2.10.1 Monochrome
Until the late 1990s the most popular monitor used
in CCTV systems was the monochrome CRT monitor.
It is still used and is available in sizes ranging from
a 1-inch-diagonal viewfinder to a large 27-inch-diagonal
CRT. By far the most popular monochrome monitor size
is the 9-inch-diagonal that optimizes video viewing for a
person seated about 3 feet away. A second reason for
its popularity is that two of these monitors fit into the
standard EIA 19-inch-wide rack-mount panel. Figure 2-26b
shows two 9-inch monitors in a dual rack-mounted version. A triple rack-mount version of a 5-inch-diagonal
monitor is used when space is at a premium. The triple
rack-mounted monitor is popular, since three fit conveniently into the 19-inch EIA rack. The optimum viewing
distance for the triple 5-inch-diagonal monitor is about
1.5 feet.
2.10.2 Color
Color monitors are now in widespread use and range in
size from 3 to 27 inch diagonal and have required viewing
distances and capabilities similar to those of monochrome
monitors. Since color monitors require three differentcolored dots to produce one pixel of information on
the monitor, they have lower horizontal resolution than
monochrome monitors. Popular color monitor sizes are
13, 15, and 17 inch diagonal.
2.10.3 CRT, LCD, Plasma Displays
The video security picture is displayed on three basic types
of monitor screens: (1) cathode ray tube (CRT), (2) liquid
crystal display (LCD), and most recently (3) the plasma
display (Figure 2-26d). The analog CRT has seen excellent
service from the inception of video and continues as a
strong contender providing a low-cost, reliable security
monitor. The digital LCD monitor is growing in popularity because of its smaller size (smaller depth), 2–3 inches
vs. 12–20 inches for the CRT. The LCD is an all solid-state
display accepts the VGA computer signal. Most small (3–10
inch diagonal) and many large (10–17 inch diagonal)
Video Technology Overview
LCD monitors also accept an analog video input. The
most recent monitor entry into the security market is the
digital plasma display. This premium display excels in
resolution and brightness and viewing angle and produces
the highest quality image in the industry. It is also the most
expensive. Screen sizes range from 20 to 42 inches diagonal. Overall depths are small and range in size from 3 to 4
inches. They are available in 4:3 and HDTV 16 9 format.
41
a second. The VCR cassette tape is transportable and the
DVR and optical disk systems are available with or without removable disks. This means that the video images
(digital data) can be transported to remote locations or
stored in a vault for safekeeping. The removable DVR
and optical disks are about the same size as Victor Home
System (VHS) cassettes. Chapter 9 describes analog and
digital video recording equipment in detail. The digital
DVR technology has all but replaced the analog VCR.
2.10.4 Audio/Video
Many monitors have built-in audio channel with speakers,
to produce audio and video simultaneously.
2.11 RECORDERS
The video camera, transmission means, and monitor provide the remote eyes for the security guard but as soon
as the action or event is over the image disappears from
the monitor screen forever. When a permanent record
of the live video scene is required a VCR, DVR, network
recorder, or optical disk recorder is used (Figure 2-27).
The video image can be recorded in real-time, near
real-time, or TL. The VCRs record the video signal on a
magnetic tape cassette with a maximum real-time recording time of 6 hours and near real-time of 24 hours. When
extended periods of recording are required (longer than
the 6 hour real-time cassette), a TL recorder is used.
In the TL process the video picture is not recorded continuously (real-time), but rather “snap-shots” are recorded.
These snap shots are spread apart in time by a fraction
of a second or even seconds so that the total elapsed
time for the recording can extend for hundreds of hours.
Some present TL systems record over an elapsed time of
1280 hours.
The DVR records the video image on a computer magnetic HD(hard drive) and the optical disk storage on an
optical disk media. The DVR and optical disk systems
have a significant advantage over the VCR with respect
to retrieval time of a particular video frame. VCRs take
many minutes to fast-forward or fast-rewind the magnetic
tape to locate a particular frame on the tape. Retrieval
times on DVRs and optical disks are typically a fraction of
(A) SINGLE CHANNEL DVR
FIGURE 2-27
2.11.1 Video Cassette Recorder (VCR)
Magnetic storage media have been used universally to
record the video image. The VCR uses the standard VHS
cassette format. The 8 mm Sony format is used in portable
surveillance equipment because of its smaller size. Super
VHS and Hi-8 formats are used to obtain higher resolution. VCRs can be subdivided into two classes: realtime and TL. The TL recorder has significantly different
mechanical and electrical features permitting it to take
snapshots of a scene at predetermined (user-selectable)
intervals. It can also record in real-time when activated by
an alarm or other input command. Real-time recorders
can record up to 6 hours in monochrome or color.
Time-lapse VCRs are available for recording time-lapse
sequences up to 720 hours.
2.11.2 Digital Video Recorder (DVR)
The DVR has emerged as the new generation of magnetic
recorder of choice. A magnetic HD like those used in a
microcomputer can store many thousands of images and
many hours of video in digital form. The rapid implementation and success of the DVR has resulted from the
availability of inexpensive digital magnetic memory storage devices and the advancements made in digital signal
compression techniques. Present DVRs are available in single channel, 4 and 16 channels and may be cascaded to
provide many more channels.
A significant feature of the DVR is the ability to access
(retrieve) a particular frame or recorded time period anywhere on the disk in a fraction of a second. The digital
(B) 16 CHANNEL DVR
DVR and NVR video disk storage equipment
(C) 32 CHANNEL NVR
42
CCTV Surveillance
technology also allows making many generations (copies)
of the stored video images without any errors or degradation of the image.
printout can then be given to another guard to take action.
For courtroom uses, time, date, and any other information can be annotated on the printed image. Chapter 10
describes hard-copy video printer systems in detail.
2.11.3 Optical Disk
2.13 ANCILLARY EQUIPMENT
When very large volumes of video images need to be
recorded, an optical disk system is used. Optical disks have
a much larger video image database capacity than magnetic disks given the same physical space they occupy.
These disks can record hundreds of times longer than
their magnetic counterparts.
2.12 HARD-COPY VIDEO PRINTERS
A hard-copy printout of a video image is often required
as evidence in court, as a tool for apprehending a vandal
or thief, or as a duplicate record of some document or
person. The printout is produced by a hard-copy video
printer, a thermal printer that “burns” the video image
onto coated paper or an ink-jet or laser printer. The thermal technique used by many hard-copy printer manufacturers produces excellent-quality images in monochrome
or color. Figure 2-28 shows a monochrome thermal printer
and a sample of the hard-copy image quality it produces.
In operation, the image displayed on the monitor or
played back from the recorder is immediately memorized
by the printer and printed out in less than 10 seconds.
This is particularly useful if an intrusion or unauthorized
act has occurred and been observed by a security guard.
An automatic alarm or a security guard can initiate printing the image of the alarm area or of the suspect and the
Most video security systems require additional accessories and equipment, including: (1) camera housings,
(2) camera pan/tilt mechanisms and mounts, (3) camera
identifiers, (4) VMDs, (5) image splitters/inserters, and
(6) image combiners. These are described in more detail
in Chapters 13, 15, 16, and 17. The two accessories most
often used with the basic camera, monitor and transmission link, described previously are camera housings and
pan/tilt mounts. Outdoor housings are used to protect
the camera and lens from vandalism and the environment. Indoor housings are used primarily to prevent vandalism and for aesthetic reasons. The motorized pan/tilt
mechanisms rotate and point the system camera and lens
via commands from a remote control console.
2.13.1 Camera Housings
Indoor and outdoor camera housings protect cameras and
lenses from dirt, dust, harmful chemicals, the environment, and vandalism. The most common housings are
rectangular metal or plastic products, formed from high
impact indoor or outdoor plastic, painted steel, or stainless
steel (Figure 2-29). Other shapes and types include cylindrical (tube), corner-mount, ceiling- mount, and dome
housings.
(A) PRINTER
FIGURE 2-28
Thermal monochrome video printer and hard copy
(B) HARDCOPY
Video Technology Overview
FIGURE 2-29 Standard indoor/outdoor video housings: (a) corner, (b) elevator corner, (c) ceiling, (d) outdoor
environmental rectangular, (e) dome, (f) plug and play
43
44
CCTV Surveillance
2.13.1.1 Standard-rectangular
The rectangular type housing is the most popular. It protects the camera from the environment and provides a
window for the lens to view the scene. The housings are
available for indoor or outdoor use with a weatherproof
and tamper resistant design. Options include: heaters,
fans, and window washers.
2.13.1.2 Dome
A significant part of video surveillance is accomplished
using cameras housed in the dome housing configuration.
The dome camera housing can range from a simple fixed
monochrome or color camera in a hemispherical dome
to a “speed-dome” housing having a high resolution color
camera with remote controlled pan/tilt/zoom/focus.
Other options include presets and image stabilization.
The dome-type housing consists of a plastic hemispherical dome on the bottom half. The housing can be clear,
tinted, or treated with a partially transmitting optical coating that allows the camera to see in any direction. In a
freestanding application (e.g. on a pole, pedestal, or overhang), the top half of the housing consists of a protective
cover and a means for attaching the dome to the structure. When the dome housing is mounted in a ceiling, a
simpler housing cover is provided and mounted above the
ceiling level to support the dome.
2.13.1.3 Specialty
There are many other specialty housings for mounting in
or on elevators, ceilings, walls, tunnels, pedestals, hallways,
etc. These special types include: explosion proof, bullet
(A) TOP-MOUNTED
FIGURE 2-30
proof and extreme environmental construction for artic
and desert use.
2.13.1.4 Plug and Play
In an effort to reduce installation time for video surveillance cameras, manufacturers have combined the camera,
lens, and housing in one assembly ready to be mounted on
a ceiling, wall or pole and plugged into the power source
and video transmission cable. These assemblies are available in the form of domes, corner mounts, ceiling mounts,
etc. making for easy installation in indoor or outdoor
applications. Chapter 15 describes these camera housing
assemblies and their specific applications in detail.
2.13.2 Pan/Tilt Mounts
To extend the angle of coverage of a CCTV lens/camera
system a motorized pan/tilt mechanism is often used.
Figure 2-30 shows three generic outdoor pan/tilt types:
top-mounted, side-mounted, and dome camera.
The pan/tilt motorized mounting platform permits the
camera and lens to rotate horizontally (pan) or vertically
(tilt) when it receives an electrical command from the central monitoring site. Thus the camera lens is not limited
by its inherent FOV and can view a much larger area of a
scene. A camera mounted on a pan/tilt platform is usually
provided with a zoom lens. The zoom lens varies the FOV
in the pointing direction of the camera/lens from a command from the central security console. The combination
of the pan/tilt and zoom lens provides the widest angular
coverage for video surveillance. There is one disadvantage
with the pan/tilt/zoom configuration compared with the
fixed camera installation. When the camera and lens are
(B) SIDE-MOUNTED
Video pan/tilt mechanisms: top-mounted, side-mounted, indoor dome
(C) INDOOR DOME
Video Technology Overview
pointing in a particular direction via the pan/tilt platform,
most of the other scene area the camera is designed to
cover is not being viewed. This dead area or dead time is
unacceptable in many security applications and therefore
a careful consideration should be given to the adequacy of
their wide-FOV pan/tilt design. Pan/tilt platforms range
from small, indoor, lightweight units that only pan, up
to large, outdoor, environmental designs carrying large
cameras, zoom lenses, and large housings. Choosing the
correct pan/tilt mechanism is important since it generally
requires more service and maintenance than any other
part of the video system. Chapter 17 describes several
generic pan/tilt designs and their features.
2.13.3 Video Motion Detector (VMD)
Another important component in a video surveillance system is a VMD that produces an alarm signal based on a
change in the video scene. The VMD can be built into
the camera or be a separate component inserted between
the camera and the monitor software in a computer. The
VMD electronics, either analog or digital, store the video
frames, compare subsequent frames to the stored frames,
and then determine whether the scene has changed. In
operation the VMD digital electronics decides whether the
change is significant and whether to call it an alarm to
alert the guard or some equipment, or declare it a false
alarm. Chapter 13 describes various VMD electronics, their
capabilities and their limitations.
2.13.4 Screen Splitter
The electronic or optical screen splitter takes a part of
several camera scenes (two, three, or more), combines
the scenes and displays them on one monitor. The splitters do not compress the image. In an optical splitter the
image combining is implemented optically at the camera lens and requires no electronics. The electronic splitter/combiner is located between the camera output and
the monitor input. Chapter 16 describes these devices
in detail.
2.13.5 Camera Video Annotation
2.13.5.1 Camera ID
When multiple cameras are used in a video system some
means must be provided to identify the camera. The system uses a camera identifier component that electronically
assigns an alphanumeric code and/or name to each camera displayed on a monitor, recorded on a recorder, or
printed on a printer. Alphanumeric and symbol character
generators are available to annotate the video signal with
the names of cameras, locations in a building, etc.
45
2.13.5.2 Time and Date
When time and date is required on the video image a
time/date generator is used to annotate the video picture. This information is mandatory for any prosecution
or courtroom procedure.
2.13.6 Image Reversal
Occasionally video surveillance systems use a single mirror
to view the scene. This mirror reverses the video image
from the normal left-to-right to a right-to-left (reversed
image). The image reversal unit corrects the reversal.
Chapter 16 describes this device.
2.14 SUMMARY
Video surveillance serves as the remote eyes for management and the security force. It provides security personnel
with advance notice of breeches in security, hostile, and
terrorist acts, and is a part of the plan to protect personnel
and assets. It is a critical subsystem for any comprehensive security plan. In this chapter an introduction to most
of the current video technology and equipment has been
described.
Lighting plays an important role in determining
whether a satisfactory video picture will be obtained with
monochrome and color cameras and LLL ICCD cameras. Thermal IR cameras are insensitive to light and only
require temperature differences between the target and
the background.
There are many types of lenses available for video systems: FFL, vari-focal, zoom, pinhole, panoramic, etc. The
vari-focal and zoom lenses extend the FOV of the FFL lens.
The panoramic 360 lens provides entire viewing of the
scene. The proper choice of lens is necessary to maximize
the intelligence obtained from the scene.
Many types of video cameras are available: color,
monochrome (with or without IR illumination), LLL
intensified, and thermal IR, analog and digital, simple and
full featured, daytime and nighttime. There are cameras
with built-in VMD to alert security guards and improve
their ability to detect and locate personnel and be alerted
to activity in the scene.
An important component of the video system is the analog or digital video signal transmission means from the
camera to the remote site, to the monitoring and recording site. Hard wire or fiber optics is best if the situation
permits. Analog works for short distances and digital for
long distances. The Internet works globally.
In multiple camera systems the quad and multiplexers
permit multi-camera displays on one monitor. Fewer monitors in the security room can improve guard performance.
46
CCTV Surveillance
The CRT monitor is still a good choice for many video
applications. The LCD is the solid-state digital replacement for the CRT. The plasma displays provides an
all solid state design that has the highest resolution,
brightness, and largest viewing angle, but at the highest
cost.
Until about the year 2000 the only practical means for
recording a permanent image of the scene was the VCR
real-time or TL recorder. Now, new and upgraded systems
replace the VCR with the DVR recorder with its increased
reliability and fast search and retrieve capabilities, to distribute the recorded video over a LAN, WAN, intranet
or Internet or wirelessly-WiFi using one of the 802.11
protocols.
Thermal, ink-jet and laser hard copy printers produce
monochrome and color prints for immediate picture dissemination and permanent records for archiving.
All types of camera/lens housings are available for
indoor and outdoor applications. Specialty cameras/ housings are available for elevators, stairwells, dome housings
for public facilities: casinos, shopping malls, extreme outdoor environments, etc.
Pan/tilt assemblies for indoor and outdoor scenarios
significantly increase the overall FOV of the camera system. Small, compact speed domes have found widespread
use in many indoor and outdoor video surveillance environments.
Plug and play surveillance cameras permit quick installation and turn-on and are available in almost every housing
configuration and camera type.
The video components summarized above are used
in most video security applications including: (1) retail
stores, (2) manufacturing plants, (3) shopping malls,
(4) offices (5) airports, (6) seaports, (7) bus and rail terminals, (8) government facilities etc. There is widespread
use of small video cameras and accessories for temporary
covert applications. The small size and ease of deployment of many video components and the flexibility in
transmission means over short and long distances has
made rapid deployment equipment for portable personnel
protection systems practical and important. Chapters 21
and 22 describe video surveillance systems designed for
some of these applications.
It is clear that the direction the video security industry is
taking is the integration of the video security function with
digital computing technology and the other parts of the
security system: access control, intrusion alarms, fire and
two-way communications. Video security is rapidly moving
from the legacy analog technology to the digital automatic
video surveillance (AVS) technology.
PART II
Chapter 3
Natural and Artificial Lighting
CONTENTS
3.1
3.2
3.3
3.4
3.5
3.6
Overview
Video Lighting Characteristics
3.2.1 Scene Illumination
3.2.1.1 Daytime/Nighttime
3.2.1.2 Indoor/Outdoor
3.2.2 Light Output
3.2.3 Spectral Output
3.2.4 Beam Angle
Natural Light
3.3.1 Sunlight
3.3.2 Moonlight and Starlight
Artificial Light
3.4.1 Tungsten Lamps
3.4.2 Tungsten-Halogen Lamps
3.4.3 High-Intensity-Discharge Lamps
3.4.4 Low-Pressure Arc Lamps
3.4.5 Compact Short-Arc Lamps
3.4.6 Infrared Lighting
3.4.6.1 Filtered Lamp Infrared Source
3.4.6.2 Infrared-Emitting Diodes
3.4.6.3 Thermal (Heat) IR Source
Lighting Design Considerations
3.5.1 Lighting Costs
3.5.1.1 Operating Costs
3.5.1.2 Lamp Life
3.5.2 Security Lighting Levels
3.5.3 High-Security Lighting
Summary
3.1 OVERVIEW
Scene lighting affects the performance of any
monochrome or color video security system. Whether the
application is indoor or outdoor, daytime or nighttime,
the amount of available light and its color (wavelength)
energy spectrum must be considered, evaluated, and compared with the sensitivity of the cameras to be used. In
bright sunlight daytime applications some cameras require
the use of an automatic-iris lens or electronic shutter. In
nighttime applications the light level and characteristics
of available and artificial light sources must be analyzed
and matched to the camera’s spectral and illumination
sensitivities to ensure a good video picture. In applications
where additional lighting can be installed the available
types of lamps—tungsten, tungsten-halogen, metal-arc,
sodium, mercury, and others—must be compared to optimize video performance. In applications where no additional lighting is permissible, the existing illumination
level, color spectrum, and beam angle must be evaluated
and matched to the video camera/lens combination.
An axiom in video security applications is: the more
light the better the picture. The quality of the monitor
picture is affected by how much light is available and
how well the sensor responds to the colors in the light
source. This is particularly true when color cameras are
used since they need more light and the correct colors of
light, than monochrome cameras. The energy from light
radiation is composed of a spectrum of colors, including
“invisible light” produced by long-wavelength IR and shortwavelength ultraviolet (UV) energy. Most monochrome
CCTV cameras respond to visible and near-IR energy but
color cameras are made to respond to visible light only.
Although many consider lighting to be only a decorator’s or an architect’s responsibility, the type and intensity
is of paramount importance in any video security system
and therefore the security professional must be knowledgeable.
This chapter analyzes the available natural and artificial
light sources and provides information to help in choosing an optimum light source or in determining whether
existing light levels are adequate.
47
48
CCTV Surveillance
3.2 VIDEO LIGHTING CHARACTERISTICS
The illumination present in the scene determines the
amount of light ultimately reaching the CCTV camera
lens. It is therefore an important factor in the quality of the
video image. The illumination can be from natural sources
such as the sun, moon, starlight or thermal (heat), or from
artificial sources such as tungsten, mercury, fluorescent,
sodium, metal-arc, LEDs or other lamps. Considerations
about the source illuminating a scene include: (1) source
spectral characteristics, (2) beam angle over which the
source radiates, (3) intensity of the source, (4) variations
in that intensity, and (5) location of the CCTV camera
relative to the source. Factors to be considered in the
scene include: (1) reflectance of objects in the scene,
(2) complexity of the scene, (3) motion in the scene, and
(4) degree of contrast in the scene.
3.2.1 Scene Illumination
In planning a video system it is necessary to know the kind
of illumination, the intensity of light falling on a surface,
and how the illumination varies as a function of distance
from the light source.
The video camera image sensor responds to reflected
light from the scene. To obtain a better understanding of
scene and camera illumination, consider Figure 3-1, which
shows the illumination source, the scene to be viewed, and
the CCTV camera and lens. The radiation from the illuminating source reaches the video camera by first reflecting
off the objects in the scene.
3.2.1.1 Daytime/Nighttime
Before any camera system is chosen the site should be surveyed to determine whether the area under surveillance
will receive direct sunlight and whether the camera will be
pointed toward the sun (to the south or the west). Whenever possible, cameras should be pointed away from the
sun to reduce glare and potential damage to the camera.
Also, when the camera views a bright background or bright
source, persons or objects near the camera may be hard
to identify since not much light illuminates them from
the direction of the camera. The light level from different
sources varies from a maximum of 10,000 FtCd for natural
bright sunlight to a minimum of 1 FtCd (from artificial
lamplight at night), giving a ratio of 10,000 to 1.
During nighttime, dawn or dusk operation, the camera
system may see moonlight and/or starlight, and reflected
NATURAL OR ARTIFICIAL
ILLUMINATION SOURCE
SOURCE PARAMETERS:
INTENSITY
SPECTRAL INTENSITY
(COLOR)
BEAM ANGLE
SCENE VIEWED BY
CAMERA / LENS
SCENE
REFLECTED LIGHT
FROM SCENE
CAMERA: SOLID STATE
OR TUBE SENSOR
LENS
FIELD OF VIEW
(FOV)
SCENE PARAMETERS:
ABSOLUTE REFLECTANCE
SPECTRAL REFLECTANCE
COMPLEXITY OF SCENE (FINE DETAIL)
MOTION IN SCENE
FIGURE 3-1
CCTV camera, scene, and source illumination
VIDEO SIGNAL
OUT
Natural and Artificial Lighting
light from artificial illumination. For nighttime operation the most widely used lamps are tungsten, tungstenhalogen, sodium, mercury, and high-intensity-discharge
(HID) metal-arc and xenon types.
than having to increase the lighting in a parking lot or
exterior perimeter in order to obtain a satisfactory picture
with a less expensive camera.
3.2.2 Light Output
3.2.1.2 Indoor/Outdoor
For indoor applications, the solid-state CCD and CMOS
cameras usually have sufficient sensitivity and dynamic
range to produce a good image and can operate with
manual-iris lenses. When video surveillance cameras view
an outdoor scene, the light source is natural or artificial,
depending on the time of day. During the daytime, operating conditions will vary, depending on whether there is
bright sun, clouds, overcast sky, or precipitation; the light’s
color or spectral energy, as well as its intensity, will vary.
The CCTV camera for outdoor applications, where the
light level and scene contrast range widely, requires automatic light-level adjustment, usually an automatic-iris lens
or an electronic shutter in the camera. Most outdoor cameras must have automatic-iris-control lenses or shuttered
CCDs to adjust over the large light-level range encountered. Very often an expensive CCTV camera may cost less
The amount of light produced by any light source is
defined by a parameter called the “candela”—related to
the light from one candle (Figure 3-2).
One FtCd of illumination is defined as the amount of
light received from a 1-candela source at a distance of 1
foot. A light meter calibrated in FtCd will measure 1 FtCd
at a distance of 1 foot from that source. As shown in
Figure 3-2, the light falling on a 1-square-foot area at a
distance of 2 feet is one-quarter FtCd. This indicates that
the light level varies inversely as the square of the distance
between the source and observer. Doubling the distance
from the source reduces the light level to one-quarter of
its original level. Note that exactly four times the area is
illuminated by the same amount of light—which explains
why each quarter of the area receives only a quarter of
the light.
LIGHTING TERMINOLOGY
REFERENCE
SOURCE
POWER TO ENERGY (LIGHT) SOURCE: POWER IN (WATTS)
1 fc
LUMINOUS INTENSITY (I )
(1 CANDLE)
1/4 fc
1
AREA = A = 1
D = DISTANCE FROM SOURCE = 2
ILLUMINATION ON SURFACE
AREA = A = 4
ILLUMINANCE (E ) AT A DISTANCE (D ) IS
PROPORTIONAL TO 1/DISTANCE SQUARED
E = I/D 2
ILLUMINANCE AT THE SCENE = LUMENS/SQUARE FOOT (FtCd)
OR: LUMENS/SQUARE METER (lux)
FIGURE 3-2
49
Illumination defined—the inverse square law
50
CCTV Surveillance
3.2.3 Spectral Output
Since different CCTV camera types respond to different
colors it is important to know what type of light source
is illuminating the surveillance area as well as what type
might have to be added to get the required video picture.
Figure 3-3 shows the spectral light–output characteristics
from standard tungsten, tungsten-halogen, and sodium
artificial sources, as well as that from natural sunlight.
Superimposed on the figure is the spectral sensitivity of
the human eye. Each source produces light at different
wavelengths or colors. To obtain the maximum utility from
any video camera it must be sensitive to the light produced
by the natural or artificial source. Sunlight, moonlight,
and tungsten lamps produce energy in a range in which
all video cameras are sensitive. Solid-state CCD sensors
are sensitive to visible and near-IR sources but many CCD
cameras have IR cut filters which reduce this IR sensitivity.
3.2.4 Beam Angle
Another characteristic important in determining the
amount of light reaching a scene is the beam angle over
which the source radiates.
One parameter used to classify light sources is their
light-beam pattern: Do they emit a wide, medium, or narrow beam of light? The requirement for this parameter
is determined by the FOV of the camera lens used and
the total scene to be viewed. It is best to match the camera lens FOV (including any pan and tilt motion) to the
light-beam radiation pattern to obtain the best uniformity of illumination over the scene, and the best picture
quality and light efficiency. Most lighting manufacturers
have the coefficient of utilization (CU) for specific fixture
luminaires. The CU expresses how much light the fixture
luminaire (lens) directs to the desired location (example: CU = 75%). Figure 3-4 shows the beam patterns of
natural and artificial light sources. The natural sources
are inherently wide while artificial sources are available
in narrow-beam (a few degrees) to wide-beam (30–90 )
patterns.
The sun and moon, as well as some artificial light sources
operating without a reflector, radiate over an entire scene.
Artificial light sources and lamps almost always use lenses
and reflectors and are designed or can sometimes be
adjusted to produce narrow- or wide-angle beams. If a
large area is to be viewed, either a single wide-beam source
or multiple sources must be located within the scene to
illuminate it fully and uniformly. If a small scene at a long
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
100
HUMAN EYE
RESPONSE
80
60
SUN
HIGH PRESSURE
SODIUM (YELLOW)
W,WI
HID
40
MERCURY (Hg)
20
FIGURE 3-3
700
800
RED
BLUE
600
ORANGE
500
GREEN
YELLOW
400
VIOLET
0
Light and IR output from common illumination sources
900
1000
1100
WAVELENGTH
(NANOMETERS
W = TUNGSTEN
WI = TUNGSTEN HALOGEN
Natural and Artificial Lighting
PAR
SPOT
SUN
PAR
FLOOD
FLUORESCENT
51
HIGH INTENSITY
DISCHARGE (HID)
40
20
NOTE: PAR LAMP (PARABOLIC ALUMINIZED REFLECTOR)
HIGH INTENSITY DISCHARGE (HID)
MERCURY
METALARC
SODIUM
FIGURE 3-4
Beam patterns from common sources
range is to be viewed, it is necessary to illuminate only that
part of the scene to be viewed, resulting in a reduction in
the total power needed from the source.
3.3 NATURAL LIGHT
There are two broad categories of light and heat sources:
natural and artificial. Natural light sources include the
sun, moon (reflected sunlight), stars, and thermal (heat).
The visible natural sources contain the colors of the visible
spectrum (blue to red) as shown in Figure 3-3. Sunlight
and moonlight contain IR radiation in addition to visible
light spectra and are classified as broadband light sources,
that is, they contain all colors and wavelengths. Far-IR radiation in the 3–5 micrometer (m) and 8–11 m spectrum
produces heat energy. Only thermal IR imaging cameras
are sensitive to this far-IR energy.
Artificial light sources can be broadband or narrowband, i.e. containing only a limited number of colors or
all of them. Monochrome video systems cannot perceive
the color distribution or spectrum of colors from different
light sources. The picture quality of monochrome cameras depends solely on the total amount of energy emitted
from the lamp that the camera is sensitive to. When the
lamp output spectrum falls within the range of the camera
sensor spectral sensitivity then the camera produces the
best picture.
For color video systems the situation is more complex
and critical. Broadband light sources containing most of
the visible colors are necessary for a color camera. To get a
good color balance the illumination source should match
the sensor sensitivity. For the camera to be able to respond
to all the colors in the visible spectrum the light source
must contain all the colors of the spectrum. Color cameras
have an automatic white-balance control that automatically
adjusts the camera electronics to produce the correct color
balance. The light source must contain the colors in order
for them to be seen on the monitor. Broadband light
sources such as the sun, tungsten or tungsten-halogen,
and xenon produce the best color pictures because they
contain all the colors in the spectrum.
If the scene in Figure 3-1 is illuminated by sunlight,
moonlight, or starlight, it will receive uniform illumination. If it is illuminated by several artificial sources, the
lighting may vary considerably over the FOV of the camera and lens. For outdoor applications the camera system must operate over the full range from direct sunlight
52
CCTV Surveillance
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
NEAR
INFRARED (IR)
SPECTRUM
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
100
SUN ENERGY
SPECTRUM
HUMAN EYE
RESPONSE
3400 K
80
3000 K
CMOS
60
CCD
CMOS
40
FIGURE 3-5
700
800
900
1000
1100
RED
600
ORANGE
500
GREEN
YELLOW
400
BLUE
0
VIOLET
20
WAVELENGTH
(NANOMETERS)
Spectral characteristics of natural sources and camera sensors
to nighttime conditions, and must have an automatic
light control means to compensate for this light-level
change. Figure 3-5 summarizes the characteristics of natural sources, i.e. the sun, moon, and starlight, and how
different camera types respond to them.
Table 3-1 summarizes the overall light-level ranges, from
direct sunlight to overcast starlight.
During the first few hours in the morning and the last
few hours in the evening, the sunlight’s spectrum is shifted
toward the orange–red region, so things look predominantly orange and red. During the midday hours, when
the sun is brightest and most intense, blues and greens are
LIGHT LEVEL
3.3.1 Sunlight
The sun is the energy source illuminating an outdoor
scene during the daylight hours. The sun emits a continuum of all wavelengths and colors to which monochrome
and color television cameras are sensitive. This continuum includes visible radiation in the blue, green, yellow,
orange, red, and also in the IR range of the spectrum. The
sun also produces long wavelength thermal IR heat energy
that is used by thermal (heat) imaging IR cameras. All
monochrome and color solid-state cameras are sensitive to
the visible spectrum, and some monochrome cameras to
the visible and near-IR spectrum. Color cameras are sensitive to all the color wavelengths in the visible spectrum
(as is the human eye), but color cameras are purposely
designed to be insensitive to near-IR wavelengths.
LIGHTING CONDITION
fc*
UNOBSTRUCTED SUN
10,000
SUN WITH LIGHT CLOUD
7,000
SUN WITH HEAVY CLOUD
2,000
SUNRISE, SUNSET
50
TWILIGHT
.4
FULL MOON
.02
QUARTER MOON
.002
OVERCAST MOON
.0007
CLEAR NIGHT SKY
.0001
AVERAGE STARLIGHT
.00007
OVERCAST NIGHT SKY
.000005
lux **
100,000
70,000
20,000
500
4
.2
.02
.007
.001
.0007
.00005
* LUMENS PER SQUARE FOOT (fc)
** LUMENS PER SQUARE METER (lux)
NOTE: 1 fc EQUALS APPROXIMATELY 10 lux
Table 3-1
Light-Level Range from Natural Sources
Natural and Artificial Lighting
brightest and reflect the most balanced white light. For
this reason a color camera must have an automatic whitebalance control that adjusts for color shift during the day,
so that the resulting video picture is color corrected.
3.3.2 Moonlight and Starlight
After the sun sets in an environment with no artificial
lighting, the scene may be illuminated by the moon, the
stars, or both. Since moonlight is the reflected light from
the sun it contains most of the colors emitted from the sun.
However, the low level of illumination reaching the earth
from the moon (or stars) prevents color cameras (and the
human eye) from providing good color rendition.
3.4 ARTIFICIAL LIGHT
The following sections describe some of the artificial light
sources in use today and how their characteristics affect
their use in video security applications. Artificial light
sources consist of the several types of lamps used in outdoor parking lots, storage facilities, fence lines, or in
indoor environments for lighting rooms, hallways, work
areas, elevators, etc. Two types of lamps are common: tungsten or tungsten-halogen lamps having solid filaments,
and gaseous or arc lamps containing low- or high-pressure
gas in an enclosed envelope. Arc lamps can be further
classified into HID, low-pressure, and high-pressure shortarc types. High-intensity-discharge-lamps are used most
extensively because of their high efficacy (efficiency in
converting electrical energy into light energy) and long
life. Low-pressure arc lamps include fluorescent and lowpressure sodium types used in many indoor and outdoor
installations. Long-arc xenon lamps are used in large outdoor sports arenas. High-pressure short-arc lamps find use
in applications that require a high-efficiency, well-directed
narrow beam to illuminate a target at long distances
(hundreds or thousands of feet). Such lamps include
xenon, metal-halide, high-pressure sodium, and mercury.
For covert security applications some lamps are fitted with
a visible-light-blocking filter so that only invisible IR radiation illuminates the scene.
Narrow-band light sources such as mercury-arc or
sodium-vapor lamps do not produce a continuous spectrum of colors, so color is rendered poorly. A mercury
lamp has little red light output and therefore red objects
appear nearly black when illuminated by a mercury arc.
Likewise a high-pressure sodium lamp contains large quantities of yellow, orange, and red light and therefore a blue
or blue–green object will look dark or gray or brown in
its light. A low-pressure sodium lamp produces only yellow light and consequently is unsuitable for color video
applications.
53
A significant advance in tungsten lamp development
came with the use of a halogen element (iodine or
bromine) in the lamp’s quartz envelope, with the lamp
operating in what is called the “tungsten-halogen cycle.”
This operation increases a lamp’s rated life significantly
even though it operates at a high temperature and light
output. Incandescent filament lamps are available with
power ratings from a fraction of a watt to 10 kilowatts.
High intensity discharge arc lamps comprise a broad
class of lamps in which the arc discharge takes place
between electrodes contained in a transparent or translucent bulb. The spectral radiation output and intensity are
determined principally by the chemical compounds and
gaseous elements that fill the bulb. The lamp is started
using a high-voltage ignition circuit with some form of
electrical ballasting used to stabilize the arc. In contrast,
tungsten lamps operate directly from the power source.
Compact short-arc lamps are only a few inches in size
but emit high-intensity, high-lumen output radiation with
a variety of spectral characteristics.
Long-arc lamps, such as fluorescent, low-pressure
sodium vapor, and xenon have output spectral characteristics determined by the gas in the arc or the tube-wall
emitting material. The fluorescent lamp has a particular
phosphor coating on the inside of the glass and bulb that
determines its spectral output. Power outputs available
from arc-discharge lamps range from a few watts up to
many tens of kilowatts.
An important aspect of artificial lighting is the consideration of the light-beam pattern from the lamp and the
camera lens FOV. A wide-beam flood lamp will illuminate
a large area with a fairly uniform intensity of light and
therefore produce a well-balanced picture. A narrow-beam
light or spotlight will illuminate a small area and consequently areas at the edge of the scene and beyond will be
darker. A scene that is illuminated non-uniformly (i.e. with
high contrast) and having “hotspots” will result in a nonuniform picture. For maximum efficiency the camera–lens
combination FOV should match the lamp beam angle.
If a lamp illuminates only a particular area of the scene
the camera–lens combination FOV should only be viewing that area illuminated by the lamp. This source beam
angle problem does not exist for areas lighted by natural illumination such as the sun, which usually uniformly
illuminates the entire scene except for shadows.
3.4.1 Tungsten Lamps
The first practical artificial lighting introduced in 1907
took the form of an incandescent filament tungsten lamp.
These lamps used a tungsten mixture formed into a filament and produced an efficacy (ratio of light out to power
in) of approximately 7 lumens per watt of visible light.
This represented a great increase over anything existing
at the time, but represents a low efficiency compared to
54
CCTV Surveillance
most other present lamp types. In 1913 ductile tungsten
wire fabricated into coiled filaments increased efficacy to
20 lumens per watt.
Today the incandescent lamp is commonplace and is
still used in most homes, businesses, factories, and public facilities. While its efficacy does not measure up to
that of the arc lamp, the tungsten and tungsten-halogen
incandescent lamps nevertheless offer a low-cost installation for many applications. Since it is an incandescent
source, it radiates all the colors in the visible spectrum
as well as the near-IR spectrum providing an excellent
light source for monochrome and color cameras. Its two
disadvantages when compared with arc lamps are: (1) relatively low efficacy, which makes it more expensive to
operate, and (2) relatively short operating life of several
thousand hours.
Incandescent filament lamp efficacy increases with
filament operating temperature; however, lamp life
expectancy decreases rapidly as lamp filament temperature increases. Maximum practical efficacy is about
35 lumens per watt in high-wattage lamps operated at
approximately 3500 K color temperature. A tungsten lamp
cannot operate at this high temperature since it will last
only a few hours. At lower temperatures, life expectancy
increases to several thousand hours, which is typical of
incandescent lamps used in general lighting.
An incandescent lamp consists of a tungsten filament
surrounded by an inert gas sealed inside a transparent or
frosted-glass envelope. The purpose of the frosted glass
is to increase the apparent size of the lamp, thereby
(A) TUNGSTEN HALOGEN
IN QUARTZ ENVELOPE
FIGURE 3-6
decreasing its peak intensity and reducing glare and
hotspots in the illuminated scene.
Incandescent lamp filaments are usually coiled to
increase their efficiency. The coils are sometimes coiled
again (coiled-coiled) to further increase the filament area
and increase the luminance. Filament configurations are
designed to optimize the radiation patterns for specific
applications. Sometimes long and narrow filaments are
used and mounted into cylindrical reflectors to produce
a rectangular beam pattern. Others have small filaments
so as to be incorporated into parabolic reflectors to produce a narrow collimated beam (spotlight). Others have
larger filament areas and are used to produce a wide-angle
beam (such as a floodlight). Figure 3-6 shows several lamp
configurations.
Figure 3-7 shows some standard lamp luminaires used
in industrial, residential, and security applications. The
luminaire fixtures house the tungsten, HID, and lowpressure lamps.
The tungsten-halogen lamp design is a significant
improvement over the incandescent lamp. In conventional
gas-filled, tungsten-filament incandescent lamps, tungsten
molecules evaporate from the incandescent filament, flow
to the relatively cool inner surface of the bulb wall (glass).
The tungsten adheres to the glass and forms a thin film
that gradually thickens during the life of the lamp and
causes the bulb to darken. This molecular action reduces
the lumen light output and efficacy in two ways. First,
evaporation of tungsten from the filament reduces the
filament wire’s diameter and increases its resistance, so
(B) TUNGSTEN FILAMENT
Generic tungsten, tungsten–Halogen lamp configurations
(C) TUNGSTEN HALOGEN IN PARABOLIC
ALUMINIZED REFLECTOR (PAR)
Natural and Artificial Lighting
FIGURE 3-7
(A) TUNGSTEN HALOGEN
(B) SODIUM
(C) FLUORESCENT
(D) METAL ARC MERCURY
55
Standard lamp luminaires
that light output and color temperature increase. Second,
the tungsten deposited on the bulb wall increases the
opacity (reduces transmission of light through the glass)
as it thickens. Figure 3-8 illustrates the relative amount
of energy produced by tungsten-filament and halogenquartz-tungsten lamps as compared with other arc-lamp
types, including fluorescent, metal-arc, and sodium, in the
visible and near-IR spectral range.
On an absolute basis, the energy produced by the tungsten lamp in the visible spectral region is significantly
lower than that provided by HID lamps. However, the total
amount of energy produced by the tungsten lamp over the
entire spectrum is comparable to that of the other lamps.
Figure 3-8 shows the human eye response and spectral
sensitivity of standard CCTV camera sensors.
3.4.2 Tungsten-Halogen Lamps
The discovery of the tungsten-halogen cycle significantly increased the operating life of the tungsten lamp.
Tungsten-halogen lamps, like conventional incandescent
lamps, use a tungsten filament in a gas-filled lighttransmitting envelope and emit light with a spectral
distribution similar to that of a tungsten lamp. Unlike
the standard incandescent lamp, the tungsten-halogen
lamp contains a trace vapor of one of the halogen elements (iodine or bromine) along with the usual inert fill
gas. Also, tungsten-halogen lamps operate at much higher
gas pressure and bulb temperature than non-halogen
incandescent lamps. The higher gas pressure retards the
tungsten evaporation, allowing the filament to operate
at a higher temperature, resulting in higher efficiencies
than conventional incandescent lamps. To withstand these
higher temperatures and pressures, the lamps use quartz
bulbs or high-temperature “hard” glass. The earliest version of these lamps used fused quartz bulbs and iodine
vapor and were called “quartz iodine lamps.” After it was
found that other halogens could be used, the more generic
tungsten-halogen lamp is now used.
The important result achieved with the addition of halogen was caused by the “halogen regenerative cycle,” which
maintains a nearly constant light output and color temperature throughout the life of the lamp and significantly
56
CCTV Surveillance
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
100
CMOS
80
CCD WITHOUT
IR FILTER
CCD WITH
IR FILTER
HID
60
WI
40
HIGH PRESSURE
SODIUM (YELLOW)
HG
20
FLUORESCENT
FIGURE 3-8
700
800
RED
BLUE
600
ORANGE
500
GREEN
YELLOW
400
VIOLET
0
900
1000
1100
WAVELENGTH
(NANOMETERS)
WI = TUNGSTEN HALOGEN
HID = HIGH INTENSITY DISCHARGE
Spectral characteristics of lamps and camera sensors
extends the life of the lamp. The halogen chemical cycle
permits the use of more compact bulbs compared to those
of tungsten filament lamps of comparable ratings and permits increasing either lamp life or lumen output and color
temperature to values significantly above those of conventional tungsten filament lamps.
Incandescent and xenon lamps are good illumination
sources for IR video applications when the light output
is filtered with a covert filter (one that blocks or absorbs
the transmission of visible radiation) and they transmit
only the near-IR radiation. Figure 3-9 shows a significant
portion of the emitted spectrum of the lamp radiation
falling in the near-IR region that is invisible to the human
eye but to which solid-state silicon sensor cameras are
sensitive. The reason for this is shown in Figure 3-9, which
details the spectral characteristics of these lamps.
When an IR-transmitting/visible-blocking filter is placed
in front of a tungsten-halogen lamp, only the IR energy
illuminates the scene and reflects back to the CCTV camera lens. This combination produces an image on the
video monitor from an illumination source that is invisible to the eye. This technology is commonly referred to
as “seeing in the dark” i.e. there is no visible radiation
and yet a video image is discernible. Some monochrome
solid-state CCD and CMOS sensors are responsive to this
near-IR radiation. Since the IR region has no “color,” color
cameras are designed to be insensitive to the filtered IR
energy. Approximately 90% of the energy emitted by the
tungsten-halogen lamp occurs in the IR region. However,
only a fraction of this IR light can be used by silicon sensors, since they are responsive only up to approximately
1100 nanometers (nm). The remaining IR energy above
1100 nm manifests as heat, which does not contribute to
the image. While the IR source is not visible to the human
eye it is detectable by silicon camera devices and other
night vision devices (Chapter 19).
3.4.3 High-Intensity-Discharge Lamps
An enclosed arc high-intensity-discharge (HID) lamp is in
widespread use for general lighting and security applications. There are three major types of HID lamps, each
one having a relatively small arc tube mounted inside a
heat-conserving outer jacket and filled with an inert gas to
prevent oxidation of the hot arc tube seals (Figure 3-10).
Natural and Artificial Lighting
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
100
NEAR
INFRARED (IR)
SPECTRUM
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
CCD WITH
IR FILTER
57
W,WI WITH
IR FILTER
SILICON TUBE
80
CCD WITHOUT
FILTER
CMOS
60
XENON
W,WI
40
20
FIGURE 3-9
700
800
900
1000
1100
RED
600
ORANGE
500
GREEN
YELLOW
400
BLUE
0
VIOLET
CMOS
WAVELENGTH
(NANOMETERS)
W = TUNGSTEN
WI = TUNGSTEN HALOGEN
Filtered tungsten and xenon lamps vs. camera spectral sensitivity
The principle in all vapor-arc lighting systems is the
same: (1) an inert gas is contained within the tube to
spark ignition, (2) the inert gas carries current from one
electrode to the other, (3) the current develops heat and
vaporizes the solid metal or metallic-oxide inside the tube,
and (4) light is discharged from the vaporized substance
through the surface of the discharge tube and into the
area to be lighted.
The three most popular HID lamps are: (1) mercury
in a quartz tube, (2) metal halide in a quartz tube, and
(3) high-pressure sodium in a translucent aluminum-oxide
tube. Each type differs in electrical input, light output,
shape, and size. While incandescent lamps require no auxiliary equipment and operate directly from a suitable voltage, discharge sources in HID lamps require a high-voltage
starting device and electrical ballast while in operation.
The high-voltage ignition provides the voltage necessary
to start the lamp; once the lamp is started, the ballast
operates the lamp at the rated power (wattage) or current
level. The ballast consumes power, which must be factored
into calculations of system efficiency. HID lamps, unlike
incandescent or fluorescent lamps, require several minutes to warm up before reaching full brightness. If turned
off momentarily they take several minutes before they can
be turned on again (reignited).
The primary overriding advantages of HID lamps are
high efficacy and their long life, provided they are
operated at a minimum of several hours per start. Lamp
lifetime is typically 16,000 to 24,000 hours and light efficacy ranges from 60 to 140 lumens per watt. These lamps
cannot be electrically dimmed without drastically affecting
the starting warm-up luminous efficiency, color, and life.
These lamps are the most widely used lamps for lighting industrial and commercial buildings, streets, sports
fields, etc. One disadvantage of short-arc lamps just mentioned is their significant warm-up time—usually several
minutes to ten minutes. If accidentally or intentionally
turned off, these lamps cannot be restarted until they have
cooled down sufficiently to reignite the arc. This may be
2–5 minutes and then take an additional 5 minutes to
return to full brightness. Dual-HID bulbs are now available, which include two identical HID lamp units, only one
of which operates at a time. If the first lamp is extinguished
momentarily, the cold lamp may be ignited immediately,
eliminating the waiting time to allow the first lamp to
cool down.
58
CCTV Surveillance
FIGURE 3-10
(A) MERCURY
(B) XENON
(C) METAL ARC
(D) SODIUM
High-intensity-discharge lamps
Mercury HID lamps are available in sizes from 40 to
1500 watts. Spectral output is high in the blue region
but extremely deficient in the red region. Therefore they
should be used in monochrome but not color video applications (Figure 3-11).
A second class of HID lamp is the metal-halide that
is filled with mercury-metallic iodides. These lamps are
available with power ratings from 175 to 1500 watts. The
addition of metallic salts to the mercury arc improves the
efficacy and color by adding emission lines in the red
end of the spectrum. With different metallic additives or
different phosphor coatings on the outside of the lamp,
the lamp color varies from an incandescent spectrum to
a daylight spectrum. The color spectrum from the metalhalide lamp is significantly improved over the mercury
lamp and can be used for monochrome or color video
applications.
The third class of HID lamp is the high-pressure sodium
lamp. This lamp contains a special ceramic-arc tube material that withstands the chemical attack of sodium at high
temperatures, thereby permitting high luminous efficiency
and yielding a broader spectrum, compared with lowpressure sodium arcs. However, because the gas is only
sodium, the spectral output distribution from the highpressure sodium HID lamp is yellow–orange and has
only a small amount of blue and green. For this reason the lamp is not suitable for good color video security applications. The primary and significant advantage
of the high-pressure sodium lamp over virtually all other
lamps is its high efficacy, approximately 60–140 lumens
per watt. It also enjoys a long life, approximately 24,000
hours. The sodium lamp is an extremely good choice
for monochrome surveillance applications. High-intensitydischarge lamps are filled to atmospheric pressure (when
not operating) and rise to several atmospheres when operating. This makes them significantly safer than short-arc
lamps that are under much higher pressure at all times.
The choice of lamp is often determined by architectural
criteria, but the video designer should be aware of the
color characteristics of each lamp to ensure their suitability
for monochrome or color video.
Natural and Artificial Lighting
59
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
100
80
METAL HALIDE
60
METAL HALIDE
DAYLIGHT
HIGH PRESSURE
SODIUM (YELLOW)
40
MERCURY (HG)
20
FIGURE 3-11
700
RED
600
ORANGE
BLUE
500
GREEN
YELLOW
400
VIOLET
0
800
900
1000
1100
WAVELENGTH
(NANOMETERS)
Spectral output from HID lamps
3.4.4 Low-Pressure Arc Lamps
Fluorescent and low-pressure sodium lamps are examples
of low-pressure arc lamp illumination sources. These
lamps have tubular bulb shapes and long arc lengths (several inches to several feet). A ballast is necessary for proper
operation, and a high-voltage pulse is required to ignite
the arc and start the lamp.
The most common type is the fluorescent lamp with
a relatively high efficacy of approximately 60 lumens per
watt. The large size of the arc tube (diameter as well as
length) requires that it be placed in a large luminaire
(reflector) to achieve a defined beam shape. For this reason, fluorescent lamps are used for large-area illumination
and produce a fairly uniform pattern. The fluorescent
lamp system is sensitive to the surrounding air temperature
and therefore is used indoors or in moderate temperatures. When installed outdoors in cold weather a special
low-temperature ballast must be used to ensure that the
starting pulse is high enough to start the lamp.
The fluorescent lamp combines a low-pressure mercury arc with a phosphor coating on the interior of the
bulb. The lamp arc produces UV radiation from the
low-pressure mercury arc, which is converted into visible
radiation by the phosphor coating on the inside wall of the
outside tube. A variety of phosphor coatings is available to
produce almost any color quality (Figure 3-12).
Colors range from “cool white,” which is the most popular variety, to daylight, blue white, and so on. Lamps
are available with input powers from 4 watts to approximately 200 watts. Tube lengths vary from 6 to 56 inches
(15–144 cm). Fluorescent lamps can be straight, circular,
or U-shaped. Fluorescent lamps can emit a continuous
spectrum like an incandescent lamp simulating a daylight
spectrum and suitable for color cameras.
A second class of low-pressure lamp is the sodium lamp
which emits a single yellow color (nearly monochromatic).
These lamps have ratings from 18 to 180 watts. The lowpressure sodium lamp has the highest efficacy output of
any lamp type built to date, approximately 180 lumens
per watt. While the efficacy is high, the lamp’s pure yellow light limits it to some monochrome video surveillance
applications and to roadway lighting applications. If used
with color cameras, only yellow objects will appear yellow;
all other objects will appear brown or black.
The low-pressure sodium light utilizes pure metal
sodium with an inert-gas combination of neon–argon
enclosed in a discharge tube about 28 inches long.
The pressure in the tube is actually below atmospheric
60
CCTV Surveillance
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
100
80
LOW PRESSURE
SODIUM (YELLOW)
60
FLUORESCENT—DAYLIGHT
40
FLUORESCENT—WARM
FIGURE 3-12
700
RED
600
ORANGE
500
GREEN
YELLOW
400
BLUE
0
VIOLET
20
800
900
1000
1100
WAVELENGTH
(NANOMETERS)
Light output from low pressure arc lamps
pressure, which causes the glass to collapse inward if it is
ruptured—a good safety feature.
A unique advantage of the low-pressure sodium amber
light is its better “modeling” (showing of texture and
shape) of any illuminated surface, for both the human
eye and the CCTV camera. It provides more contrast, and
since the monochrome CCTV camera responds to contrast, images under this light are clearer, according to
some reports. The yellow output from the sodium lamp
is close to the wavelength region at which the human eye
has its peak visual response (560 nanometers).
Some security personnel and the police have identified
low-pressure sodium as a uniquely advantageous off-hour
lighting system for security because the amber yellow color
clearly tells people to keep out. This yellow security lighting also sends the psychological message that the premises
are well guarded.
3.4.5 Compact Short-Arc Lamps
Enclosed short-arc lamps comprise a broad class of lamps
in which the arc discharge takes place between two closely
spaced electrodes, usually tungsten, and is contained in a
rugged transparent or frosted bulb. The spectrum radiated
by these lamps is usually determined by the elements and
chemical compounds inside.
They are called short-arc because the arc is short compared with its electrode size, spacing, and bulb size and
operates at relatively high currents and low voltages. Such
lamps are available with power ratings ranging from less
than 50 watts to more than 25 kilowatts. These lamps usually operate at less than 100 volts, although they need a
high-voltage pulse (several thousand volts) to start. Most
short-arc lamps operate on AC or DC power and require
some form of current-regulating device (ballast) to maintain a uniform output radiation.
Several factors limit the useful life of compact lamps
compared with HID lamps, especially the high current
density required which reduces electrode lifetime. Compact short-arc lamps generally have a life in the low
thousands of hours and operate at internal pressures up
to hundreds of atmospheres. Therefore they must be
operated in protected enclosures and handled with care.
The most common short-arc lamps are mercury, mercuryxenon, and xenon. Figure 3-13 shows the spectral output
of mercury-xenon lamps.
Natural and Artificial Lighting
61
RELATIVE
SPECTRAL
INTENSITY
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
100
80
XENON
MERCURY (HG)
60
40
MERCURY–XENON
20
FIGURE 3-13
700
RED
BLUE
600
ORANGE
500
GREEN
YELLOW
400
VIOLET
0
800
900
1000
1100
WAVELENGTH
(NANOMETERS)
Spectral outputs of mercury–xenon lamps
Short-arc xenon lamps are not common in security
applications because of their high cost and short lifetime.
However, they play an important role for IR sources used
in covert surveillance. The light output from the mercury
arc lamp is primarily in the blue region of the visible
spectrum and therefore only fair results are obtained with
monochrome CCD or CMOS solid-state cameras. Despite
mercury lighting’s good appearance to the human eye,
typical solid-state cameras respond poorly to it.
The mercury-xenon lamp, containing a small amount
of mercury in addition to xenon gas, offers fair color
rendition. Immediately after lamp ignition the output is
essentially the same as the spectrum of a xenon lamp.
The xenon gas produces a background continuum that
improves the color rendition. As the mercury vaporizes
over several minutes, the spectral output becomes that of
mercury vapor, with light output in the blue, green, yellow,
and orange portions of the spectrum. The xenon shortarc’s luminous efficiency ranges from 20 to 53 lumens per
watt over lamp wattage ranges of 200–7000 watts.
The color temperature of the arc is approximately
6000 K, which is almost identical to that of sunlight. The
xenon lamp output consists of specific colors as well as a
continuum and some IR radiation, and produces similar
color lighting to that of the sun (Figure 3-13). The greater
percentage of the continuum radiation at all wavelengths
closely matches the spectral radiation characteristics of
sunlight. Compared with all other short-arc lamps, the
xenon lamp is the ideal artificial light choice for accurate
color rendition. The lamp spectral output does not change
with lamp life, so color rendition is good over the useful life of the lamp. Color output is virtually independent
of operating temperature and pressure, thereby ensuring
good color rendition under adverse operating conditions.
Xenon lamps are turned on with starting voltage pulses
of 10–50 kilovolts (kV). Typical lamps reach full output
intensity within a few seconds after ignition. The luminous efficiency of the xenon lamp ranges from 15 to 50
lumens per watt over a wattage range of approximately 75
to 10,000 watts.
A characteristic unique to the compact short-arc lamp
is the small size of the radiating source, usually a fraction of a millimeter to a few millimeters in diameter. Due
to optical characteristics, one lamp in a suitable reflector
can produce a very concentrated beam of light. Parabolic
and spherical reflectors, among others, are used to provide optical control of the lamp output: the parabolic
for search- or spotlights and the spherical for floodlights.
Compact short-arc lamps are often mounted in a parabolic
reflector to produce a highly collimated beam used to
62
CCTV Surveillance
housing to illuminate the scene. The second technique
uses a non-thermal IR LED or LED array to generate IR
radiation through electronic recombination in a semiconductor device. Both techniques produce narrow or wide
beams, resulting in excellent images when the scene is
viewed with an IR-sensitive camera, such as a solid-state
CCD, CMOS, or ICCD camera.
illuminate distant objects. This configuration also produces an excellent IR spotlight when an IR transmitting
filter is mounted in front of the lamp. Even when not
used for spotlighting, the small arc size of compact shortarc lamps allows the luminaire reflector to be significantly
smaller than other lamp reflectors.
Mounting orientation can affect the performance of
short-arc lamps. Most xenon lamps are designed for vertical or horizontal operation but many mercury-xenon and
mercury lamps must be operated vertically to prevent premature burnout.
3.4.6.1 Filtered Lamp Infrared Source
Xenon and incandescent lamps can illuminate a scene
many hundreds of feet from the camera and produce sufficient IR radiation to be practical for a covert video system
(Figure 3-9). Since thermal IR sources (tungsten, xenon
lamps) consume significant amounts of power and become
hot, they may require a special heat sink or air cooling
to operate continuously. Figure 3-14 shows the configuration of several tungsten and xenon lamp IR sources that
produce IR beams and that have built-in reflectors and IR
transmitting (visual blocking) filters.
These lamp systems use thin-film dichroic optical coatings (a light-beam splitter) and absorbing filters that direct
the very-near-IR rays toward the front of the lamp and
out into the beam, while reflecting visible and long-IR
radiation to the back of the lamp, where it is absorbed
3.4.6 Infrared Lighting
A covert IR lighting system is a solution when conventional
security lighting is not appropriate, for example when the
presence of a security system (1) would attract unwanted
attention, (2) would alert intruders to a video surveillance
system, or (3) would disturb neighbors.
There are two generic techniques for producing IR
lighting. One method uses the IR energy from a thermal incandescent or xenon lamp. These IR sources are
fitted with optical filters that block the visible radiation
so that only IR radiation is transmitted from the lamp
IR
TRANSMITTING
FILTER
PARABOLIC
REFLECTOR
XENON
SHORT ARC
LAMP
SPOT OR
FLOOD LAMP
(PAR)
METAL HOUSING
WITH COOLING
FINS (HEAT SINK)
AND CONVECTION
NON-VISIBLE
IR ENERGY OUT
IR TRANSMITTING
FILTER
ELECTRICAL
POWER IN
• 117 VAC
• 24 VAC
• 12 VDC
RETAINING
RING
SWIVEL
MOUNT
REFLECTED
OR ABSORBED
VISIBLE LIGHT
ENERGY
NOTE: PAR—PARABOLIC ALUMINIZED REFLECTOR
FIGURE 3-14
Thermal IR source configurations
Natural and Artificial Lighting
HEAT
SINK
HOUSING
63
TUNGSTEN
HALOGEN LAMP
(INSIDE)
IR TRANSMITTING
VISIBLE BLOCKING
WINDOW/FILTER
FIGURE 3-15
High-efficiency thermal (IR) lamp
by the housing material. The housing acts as an efficient
heat sink that effectively dissipates the heat. The system
can operate continuously in hot environments without a
cooling fan.
An especially efficient configuration using a tungstenhalogen lamp as the radiating source and a unique filtering and cooling technique is shown in Figure 3-15.
The figure shows the functioning parts of a 500-watt
IR illuminating source using a type PAR 56 encapsulated
tungsten-halogen lamp. The PAR 56 lamp filament operates at a temperature of approximately 3000 K and has an
average rated life of 2000–4000 hours. The lamp’s optical
dichroic mirror coatings on the internal surfaces of the
reflector and front cover lens are made of multiple layers of silicon dioxide and titanium dioxide. In addition to
this interference filter, there is a “cold” mirror—a quartzsubstrate shield—between the tungsten-halogen lamp and
the coated cover lens to control direct visible-light output from the filament. The lamp optics have a visible
absorbing filter between the lamp and the front lens that
transmits less than 0.1% of all wavelengths shorter than
730 nanometers. This includes the entire visible spectrum.
The compound effect of this filtering ensures that only
IR radiation leaves the front of the lamp and that visible
and long-IR radiation (longer than is useful to the silicon
camera sensor) cannot leave the front of the lamp. The
lamp output is consequently totally invisible to the human
eye. The IR lamp system is available with different front
lenses to produce beam patterns for a wide range of applications, covering wide scene illumination to long-range
spotlighting.
Table 3-2 summarizes the types of lamp lenses available
and the horizontal and vertical beam angles they produce.
These beam angles vary from 12 for a narrow beam (spotlight) to 68 for a very wide beam (flood lamp).
3.4.6.2 Infrared-Emitting Diodes
Video security systems for covert and nighttime
illumination are using IR LEDs consisting of an array
of gallium arsenide (GaAs) semiconductor diodes. These
LEDs emit a narrow band of deep red 880 nm or IR 950 nm
radiation and no other discernible visible light. These efficient devices typically convert 50% of electrical energy
to optical IR radiation. They operate just slightly above
room temperature, dissipate little heat and therefore usually require minimum cooling. The light is generated in
the diode at the PN-junction and emits IR radiation when
electrically biased in a forward direction. The 800–900 nm
IR energy is directed toward the magnifying dome lens
built into each LED emitter and directed toward the scene.
To adequately illuminate an entire scene requires an array
of ten, hundred to several hundred diodes that are connected in series with the power source. The array is powered from a conventional 12 VDC or 117 VAC source.
The IR light output from each diode adds up to produce
enough radiation to illuminate the scene and target with
sufficient IR energy to produce a good video picture with a
solid-state CCD or CMOS camera. Figure 3-16 shows an IR
LED GaAs array that produces a high-efficiency IR beam
for covert and nighttime applications.
3.4.6.3 Thermal (Heat) IR Source
All objects emit light when sufficiently hot. Changing the
temperature of an object changes the intensity and color
of the light emitted from it. For instance, iron glows dull
red when first heated, then red-orange when it becomes
hotter and eventually white hot. In a steel mill, molten
iron appears yellow–white because it is hotter than the
red-orange of the lower-temperature iron. The tungsten
filament of an incandescent lamp is hotter yet and emits
64
CCTV Surveillance
SOURCE
TYPE
INPUT POWER
(WATTS)
(VOLTAGE)
BEAM ANGLE
(DEGREES)
MAXIMUM
RANGE (ft)
100
60 HORIZ
60 VERT
30
WIDE
FLOOD
FILTERED WI
INCANDESCENT
SPOT
FILTERED WI
INCANDESCENT
100
10 HORIZ
10 VERT
200
WIDE
FLOOD
FILTERED WI
INCANDESCENT
500
40 HORIZ
16 VERT
90
SPOT
FILTERED WI
INCANDESCENT
500
12 HORIZ
8 VERT
450
FLOOD
FILTERED
XENON ARC
400
(AC)
40
500
SPOT
FILTERED
XENON ARC
400
(AC)
12
1500
FLOOD
LED
50
(12 VDC)
30
200
FLOOD
LED
8
(12 VDC)
40
70
WI = TUNGSTEN HALOGEN
LED = LIGHT EMITTING DIODE (880 nm-DEEP RED GLOW, 950 nm-INVISIBLE IR)
WI AND XENON THERMAL LAMPS USE VISUAL BLOCKING FILTERS
Table 3-2
Beam angles for IR Lamps
nearly white light. Any object that is hot enough to glow
is said to be incandescent: hence the term for heatedfilament bulbs. A meaningful parameter for describing
color is the color temperature or apparent color temperature of
an object when heated to various temperatures.
In the laboratory a special radiating source that emits
radiation with 100% efficiency at all wavelengths when
heated is called a blackbody radiator. The blackbody radiator emits energy in the ultraviolet, visible, and infrared
spectrums following specific physical laws.
Tungsten lamps and the sun radiate energy like a blackbody because they radiate with a continuous spectrum, that
is, they emit at all wavelengths and colors. Other sources
such as mercury, fluorescent, sodium, and metal-arc lamps
do not emit a continuous spectrum but only produce
narrow bands of colors: mercury produces a green–blue
band; sodium produces a yellow–orange band. Thermal
IR cameras are used to view temperature differences in
objects in a scene (Chapter 19).
3.5 LIGHTING DESIGN CONSIDERATIONS
The design of the lighting system for video security systems requires consideration of: (1) initial installation cost,
(2) efficiency of lamp type chosen, (3) cost of operation,
(4) maintenance costs, (5) spectral intensity, and (6) beam
angle of the lamp and luminaire.
3.5.1 Lighting Costs
The cost of lighting an indoor or an outdoor area depends
on factors including: (1) initial installation, (2) maintenance, and (3) operating costs (energy usage). The initial
installation costs are lowest for incandescent lighting, followed by fluorescent lighting, and then by HID lamps. All
incandescent lamps can be connected directly to a voltage
supply. They are available for alternating current electrical
supply voltages of: 240, 120, and 24 VAC and direct current
12 VDC with no need for electrical ballasting or highvoltage starting circuits. All that is required is a suitably
designed luminaire that directs the lamp illumination into
the desired beam pattern. Some incandescent lamps are
pre-focused with built-in luminaires to produce spot or
flood beam coverage. Fluorescent lamps are installed in
diffuse light reflectors and require only an igniter and simple ballast for starting and running. HID lamps require
more complex ballast networks, which are more expensive, larger and bulkier, consume electrical power, and
add to installation and operating costs.
All lamps and lamp fixtures are designed for easy
lamp replacement. Fluorescent and HID lamps that have
Natural and Artificial Lighting
65
90–130 ft
SINGLE LED
IR SOURCE
60–80 ft
BEAM
PROFILE
(DISPERSION)
50°
RELATIVE
SPECTRAL
INTENSITY
20°
380 nm
780 nm
VISIBLE
SPECTRUM
(380–780 nm)
UV
SPECTRUM
INFRARED (IR)
SPECTRUM
880
950
100
80
60
40
FIGURE 3-16
700
800
900
1000
1100
RED
600
ORANGE
500
GREEN
YELLOW
400
BLUE
0
VIOLET
20
WAVELENGTH
(NANOMETERS)
Single LED and LED array beam output characteristics
ballast modules and high-voltage starting circuits require
additional maintenance since they will fail sometime during the lifetime of the installation. Table 3-3 compares
the common lamp types including the deep red and
IR LEDs.
3.5.1.1 Operating Costs
Energy efficiency of the illumination system must be considered in a video security system. Translated into dollars
and cents, this relates to the number of lumens or light
output per kilowatt of energy input that additional lighting might cost or that could be saved if an LLL ICCD
video camera or thermal IR camera was installed.
The amount of light available directly affects the quality
and quantity of intelligence on the video monitor. If the
lighting already exists on the premises, the security professional must determine quantitatively whether the lamp
type is suitable and the amount of lighting is sufficient.
The result of a site survey will determine whether more
lighting must be added. Computer design programs are
available to calculate the location and size of the lamps
necessary to illuminate an area with a specified number
of FtCds. If adding lighting is an option, the analysis will
compare that cost with the cost of installing more sensitive
and expensive video cameras.
If the video security system includes color cameras,
the choice of lighting becomes even more critical. All
color cameras require a higher level of lighting than their
monochrome counterparts. To produce a color image having a signal-to-noise ratio or noise-free picture as good as
a monochrome system, as much as ten times more lighting is required. To obtain faithful color reproduction of
facial tones, objects, and other articles in the scene, the
light sources chosen or already installed must produce
enough of these colors for the camera to detect and balance them. Since a large number of different generic lighting types are currently installed in industrial and public
sites, the security professional must be knowledgeable in
the spectral output of such lights.
66
CCTV Surveillance
EFFICIENC Y *
LUMENS/WATT
SPECTRAL
OUTPUT
TYPE
LIFETIME
(HOURS)
POWER RANGE
(WATTS)
WARM–UP/
RESTRIKE
(MINUTES)
INITIAL
MEAN
BLUE–GREEN
32–63
25–43
16,000–24,000
50–1,000
5–7/3–6
HIGH PRESSURE
SODIUM
YELLOW–WHITE
64–140
58–126
20,000–24,000
35–1,000
3–4/1
METAL ARC:
METAL HALIDE
MULTI-VAPOR
GREEN–YELLOW
80–115
57–92
10,000–20,000
175–1,000
2–4/10–15
WHITE
74–100
49–92
12,000–20,000
28–215
IMMEDIATE
YELLOW–WHITE
YELLOW–WHITE
17–24
15–23
750–1,000
2,000
100–1,500
IMMEDIATE
IMMEDIATE
MERCURY
FLUORESCENT
INCANDESCENT:
TUNGSTEN
TUNGSTEN HALOGEN
* REFERRED TO AS EFFICACY IN LIGHTING (LUMENS/WATT)
Table 3-3
Comparison of Lamp Characteristics
TYPE
LIFETIME
(HOURS)
INITIAL
COST
OPERATING
COST
TOTAL OWNING AND
OPERATING COST
MERCURY
16,000–24,000
HIGH
MEDIUM
MEDIUM
HIGH PRESSURE SODIUM
20,000–24,000
HIGH
LOW
LOW
METAL ARC:
METAL HALIDE
MULTI-VAPOR
10,000–20,000
HIGH
LOW
LOW
FLUORESCENT
12,000–20,000
MEDIUM
MEDIUM
MEDIUM
LOW
HIGH
HIGH
INCANDESCENT:
TUNGSTEN HALOGEN
TUNGSTEN
Table 3-4
750–1,000
2,000
Light Output vs. Lamp Type over Rated Life
Since the lamp operating costs often exceeds the initial installation and maintenance costs put together, it is
important to know the efficacy of each lamp type. To
appreciate the significant differences in operating costs for
the different lamp types, Table 3-4 compares the average
light output over the life of each lamp.
For the various models of incandescent, mercury vapor
(HID), fluorescent, and high-pressure sodium lamps, lamp
life in hours is compared with input power and operating cost, kilowatt-hours (kWh) used, based on 4000
hours of annual operation. The comparisons are made
for lamps used in different applications, including duskto-dawn lighting, wall-mounted aerial lighting, and floodlighting. In each application, there is a significant saving
in operational costs (energy costs) between the highpressure sodium and fluorescent lamps as compared with
the mercury vapor and standard incandescent lamps.
Choosing the more efficient lamp over the less efficient
one can result in savings of double or triple the operational
costs, depending on the cost of electricity in a particular
location.
3.5.1.2 Lamp Life
Lamp life plays a significant role in determining the cost
efficiency of different light sources. Actual lamp replacement costs and labor costs must be considered, as well as
the additional risk of interrupted security due to unavailable lighting. Table 3-5 summarizes the average lamp life
in hours for most lamp types in use today.
At the top of the list are the high- and low-pressure
sodium lamps and the HID mercury vapor lamp, each
providing approximately 24,000 hours of average lamp
life. Next, some fluorescent lamp types have a life of
10,000 hours. At the bottom of the list are the incandescent and quartz-halogen lamps, having rated lives of
Natural and Artificial Lighting
TYPE
LIFETIME (HOURS)
POWER IN (WATTS)
MERCURY
24,000
100
250
1,000
4,100
12,100
57,500
HIGH PRESSURE
SODIUM
24,000
50
150
1,000
4,000
16,000
1,40,000
METAL ARC:
METAL HALIDE
MULTI-VAPOR
7,500
20,000
3,000
175
400
1,500
14,000
34,000
1,55,000
FLUORESCENT
18,000
12,000
10,000
30
60
215
1,950
5,850
15,000
2,000
250
4,000
INCANDESCENT:
TUNGSTEN
TUNGSTEN HALOGEN
Table 3-5
67
LUMENS OUT (fc)
Lamp Life vs. Lamp Type
approximately 1000–2000 hours. If changing lamps is
inconvenient or costly, high-pressure sodium lamps should
be used in place of incandescent types. Using highpressure sodium rather than tungsten will save 12 trips to
the site to replace a defective lamp, and having 12 fewer
burned-out lamps will reduce the amount of time the video
surveillance system will be down. High pressure sodium
lamps, however, will not produce good color rendering.
Lamp designs require specifications of wattage, voltage,
bulb type, base type, efficacy, lumen output, color temperature, life, operating cost, and other special features.
Color temperature, power input, and life ratings of a lamp
are closely related and cannot be varied independently.
For a given wattage, the lumen output and the color temperature decrease as the life expectancy increases.
In incandescent lamps, filament power (watts) is roughly
proportional to the fourth power of filament temperature.
So a lamp operated below its rated voltage has a longer
life. A rule of thumb: Filament life is doubled for each 5%
reduction in voltage; conversely, filament life is halved for
each 5% increase in voltage.
3.5.2 Security Lighting Levels
In addition to the lamp parameters and energy requirements, the size and shape of the luminaire, spacing
between lamps, and height of the lamp above the surface illuminated must be considered. Although each video
application has special illumination requirements, primary responsibility for lighting is usually left to architects or illumination engineers. To provide adequate lighting in an industrial security or safety environment in
building hall-ways, stairwells, outdoor perimeters, or parking lot facilities, different lighting designs are needed.
Table 3-6 tabulates recommended light-level requirements
for locations including parking lots, passenger platforms,
building exteriors, and pedestrian walkways.
The video system designer or security director often has
no option to increase or change installed lighting and
must first determine whether the lighting is sufficient for
the CCTV application and then make a judicious choice of
CCTV camera to obtain a satisfactory picture. If lighting is
not sufficient, the existing lighting can sometimes be augmented by “fill-in” lighting at selected locations to provide
the extra illumination needed by the camera. Chapters 4,
5, and 19 cover video lenses, cameras, and LLL cameras
respectively, and offer some options for video equipment
when sufficient lighting is not available.
3.5.3 High-Security Lighting
Lighting plays a key role in maintaining high security in
correctional facilities. Lighting hardware requires special
fixtures to ensure survival under adverse conditions. Highsecurity lamps and luminaires are designed specifically
to prevent vandalism and are often manufactured using
high-impact molded polycarbonate enclosures to withstand vandalism and punishing weather conditions without
breakage or loss of lighting efficiency (Figure 3-17).
These luminaires are designed to house incandescent,
HID, and other lamp types to provide the necessary
light intensity and the full spectrum of color rendition
required for monochrome and color video security systems. Most fixtures feature tamper-proof screws that prevent the luminaire from being opened by unauthorized
personnel. For indoor applications, high-impact polycarbonate fluorescent lamp luminaires offer a good solution.
The molded polycarbonate lenses have molded tabs that
68
CCTV Surveillance
LIGHT LEVEL
TYPE
LOCATION
PARKING AREA
LOADING DOCKS
GARAGES—REPAIR
GARAGES—ACTIVE TRAFFIC
PRODUCTION/ASSEMBLY AREA
ROUGH MACHINE SHOP/SIMPLE ASSY.
MEDIUM MACHINE SHOP/MODERATE
DIFFICULT ASSY.
DIFFICULT MACHINE WORK/ASSY.
FINE BENCH/MACHINE WORK, ASSY.
STORAGE ROOMS/WAREHOUSES
ACTIVE—LARGE/SMALL
INACTIVE
STORAGE YARDS
PARKING-OPEN (HIGH–MEDIUM ACTIVITY)
PARKING-COVERED
(PARKING, PEDESTRIAL AREA)
PARKING ENTRANCES DAY
NIGHT
FtCd
lux
INDOOR
INDOOR
INDOOR
INDOOR
5–50
20
50–100
10–20
50–500
200
500–1000
100–200
INDOOR
INDOOR
20–50
50–100
200–500
500–1000
INDOOR
INDOOR
INDOOR
200–500
200–500
15–30
2000–5000
2000–5000
150–300
INDOOR
OUTDOOR
5
1–20
50
10–200
OUTDOOR
1–2
10–20
OUTDOOR
OUTDOOR
5
5–50
50
50–500
NOTE: 1 FtCd EQUALS APPROXIMATELY 10 lux
Table 3-6
Recommended Light Levels for Typical Security Applications
UNBREAKABLE HIGH
IMPACT RESISTANT
POLYCARBONATE
PRISMATIC DIFFUSER
(A) HIGH PRESSURE SODIUM (HPS)
CAST OR WELDED ALUMINUM
HOUSING PROTECTS
ELECTRICAL COMPONENTS
20
16
12
8
4
0
FIGURE 3-17
High security luminaires
ACRYLIC
POLYSTYRENE
BUTYRATE
IMPACT RESISTANT
ABS
NON-METALLIC
COMPARATIVE
IMPACT
RESISTANCE
(NOTCHED
IZOD TEST)
POLYCARBONATE
TAMPERPROOF SCREWS
(B) FLOURESCENT
Natural and Artificial Lighting
engage special slots in the steel-backed plate and prevent
the luminaire from being opened, thereby minimizing
exposure of the fluorescent lamps to vandalism. Applications include prison cells, juvenile-detention facilities,
high-security hospital wards, parking garages, public housing hallways, stairwells, and underground tunnels.
3.6 SUMMARY
The quality of the final video picture and the intelligence
it conveys depend heavily on the natural and/or artificial light sources illuminating the scene. For optimum
results, an analysis of the lamp parameters (spectrum,
69
illumination level, beam pattern) must be made and
matched to the spectral and sensitivity characteristics of
the camera. Color systems require careful analysis when
they are used with natural illumination during daylight
hours and with broad-spectrum color-balanced artificial
illumination sources. Using multiple light sources having
different color balances in the same scene can produce
poor color rendition in the video image. If the illumination level is marginal, measure it with a light meter
(Chapter 25) to quantify the actual light reaching the camera from the scene. If there is insufficient light for the
standard solid-state video camera, augment the lighting
with additional fill-in sources or choose a more sensitive
ICCD camera (Chapter 19). As with the human eye, lighting holds the key to clear sight.
This page intentionally left blank
Chapter 4
Lenses and Optics
CONTENTS
4.1
4.2
Overview
Lens Functions and Properties
4.2.1 Focal Length and Field of View
4.2.1.1 Field-of-View Calculations
4.2.1.1.1 Tables for Scene Sizes vs.
FL for 1/4-, 1/3-, and
1/2-Inch Sensors
4.2.1.1.2 Tables for Angular FOV
vs. FL for 1/4-, 1/3-, and
1/2-Inch Sensor Sizes
4.2.1.2 Lens and Sensor Formats
4.2.2 Magnification
4.2.2.1 Lens–Camera Sensor Magnification
4.2.2.2 Monitor Magnification
4.2.2.3 Combined Camera and Monitor
Magnification
4.2.3 Calculating the Scene Size
4.2.3.1 Converting One Format to Another
4.2.4 Calculating Angular FOV
4.2.5 Lens Finder Kit
4.2.6 Optical Speed: f-number
4.2.7 Depth of Field
4.2.8 Manual and Automatic Iris
4.2.8.1 Manual Iris
4.2.8.2 Automatic-Iris Operation
4.2.9 Auto-Focus Lens
4.2.10 Stabilized Lens
4.3 Fixed Focal Length Lens
4.3.1 Wide-Angle Viewing
4.3.2 Narrow-Angle Telephoto Viewing
4.4 Vari-Focal Lens
4.5 Zoom Lens
4.5.1 Zooming
4.5.2 Lens Operation
4.5.3 Optical Speed
4.5.4 Configurations
4.5.5 Manual or Motorized
4.6
4.7
4.8
4.9
4.5.6 Adding a Pan/Tilt Mechanism
4.5.7 Preset Zoom and Focus
4.5.8 Electrical Connections
4.5.9 Initial Lens Focusing
4.5.10 Zoom Pinhole Lens
4.5.11 Zoom Lens–Camera Module
4.5.12 Zoom Lens Checklist
Pinhole Lens
4.6.1 Generic Pinhole Types
4.6.2 Sprinkler Head Pinhole
4.6.3 Mini-Pinhole
Special Lenses
4.7.1 Panoramic Lens—360
4.7.2 Fiber-Optic and Bore Scope Optics
4.7.3 Bi-Focal, Tri-Focal Image Splitting Optics
4.7.4 Right-Angle Lens
4.7.5 Relay Lens
Comments, Checklist and Questions
Summary
4.1 OVERVIEW
The function of the camera lens is to collect the reflected
light from a scene and focus it onto a camera sensor.
Choosing the proper lens is very important, since its choice
determines the amount of light received by the camera
sensor, the FOV on the monitor, and the quality of the
image displayed. Understanding the characteristics of the
lenses available and following a step-by-step design procedure simplifies the task and ensures an optimum design.
A CCTV lens functions like the human eye. Both collect light reflected from a scene or emitted by a luminous light source and focus the object scene onto some
receptor—the retina or the camera sensor. The human
eye has a fixed-focal-length (FFL) lens and variable iris
71
72
CCTV Surveillance
diaphragm, which compares to an FFL, automatic-iris
video lens. The eye has an iris that opens and closes just
like an automatic-iris camera lens and automatically adapts
to changes in light level. The iris—whether in the eye or in
the camera—optimizes the light level reaching the receptor, thereby providing the best possible image. The iris in
the eye is a muscle-controlled membrane; the automatic
iris in a video lens is a motorized device.
Of the many different kinds of lenses used in video security applications the most common is the FFL lens, which
is available in wide-angle (90 ), medium-angle (40 ), and
narrow-angle (5 ) FOVs. To cover a wide scene and also
obtain a close-up (telephoto) view with the same camera,
a variable-FOV vari-focal or zoom lens is used. The varifocal lens is used to “fine tune” the focal length (FL) to
a specific FL for the application. To further increase the
camera’s FOV a zoom lens mounted on a pan/tilt platform
is used.
The pinhole lens is used for covert video surveillance
applications since it has a small front diameter and can
easily be hidden. There are many other specialty lenses,
including split-image, fiber optic, right-angle, and automatic focus.
A relatively new lens—the panoramic 360 lens—is used
to obtain a 360 horizontal by up to 90 vertical FOV. This
lens must be used with a digital computer and software
algorithm to make use of the donut-shaped image it produces on the camera sensor. The software converts the
image to a 360 panoramic display.
4.2 LENS FUNCTIONS AND PROPERTIES
A lens focuses an image of the scene onto the CCTV camera sensor (Figure 4-1). The sensor can be a CCD, CMOS,
ICCD, or thermal IR imager.
The lens in a human and a camera have some similarities: they both collect light and focus it onto a receptor
(Figure 4-2).
They have one important difference: the human lens
has one FFL and the retina is one size, but the camera
lens may have many different FLs and the sensor may
have different sizes. The unaided human eye is limited
to seeing a fixed and constant FOV, whereas the video
system can be modified to obtain a range of FOVs. The eye
has an automatic-iris diaphragm to optimize the light level
reaching the retina. The camera lens has an iris (either
manual or automatic) to control the light level reaching
the sensor (Figure 4-3).
NATURAL OR ARTIFICIAL
ILLUMINATION SOURCE
SCENE VIEWED
BY CAMERA /LENS
D
V
REFLECTED LIGHT FROM SENSOR
HEIGHT
LENS
H
WIDTH
C OR CS
MOUNT
CAMERA
CCD, CMOS, IR
SENSOR
LENS FIELD
OF VIEW (FOV)
V = VERTICAL HEIGHT
VIDEO OUT
H = HORIZONTAL WIDTH
D = DIAGONAL
H × V = CAMERA SENSOR FOV
FIGURE 4-1
CCTV camera/lens, scene, and source illumination
POWER IN
Lenses and Optics
SENSOR FORMAT
SCENE
2/3"
1/2"
1/3"
1/4"
SCENE
CAMERA SENSOR
FIELD OF VIEW (FOV)
LENS
IRIS
EYE RETINA
CAMERA SENSOR
EYE FIELD OF VIEW
EYE MAGNIFICATION = 1
17 mm
EYE LENS FOCAL LENGTH ≈ 17 mm (0.67")
FIGURE 4-2
Comparing the human eye to a CCTV lens and camera sensor
EYE
AUTOMATIC IRIS
IRIS ALMOST CLOSED
WHEN VIEWING BRIGHT
SCENE (SUN)
METAL LEAVES
OPEN AND CLOSE
BY MOVING LENS
IRIS RING
IRIS HALF CLOSED
WHEN VIEWING NORMAL
SCENE (INDOORS)
IRIS WIDE OPEN
WHEN VEIWING DARK
SCENE (NIGHT TIME)
IRIS OPEN
HALF CLOSED
IRIS NEARLY
CLOSED
CCTV LENS
AUTOMATIC IRIS
MOTOR DRIVEN IRIS
DRIVE MOTOR/GEAR
MOTOR
DRIVE
FIGURE 4-3
Comparing the human eye and CCTV camera lens iris
VIDEO
SIGNAL
73
74
CCTV Surveillance
4.2.1 Focal Length and Field of View
In the human eye, magnification and FOV are set by the
lens FL and retina size. When the human eye and the
video camera lens and sensor see the same basic picture,
they are said to have the same FOV and magnification. In
practice, a lens that has an FL and FOV similar to that
of the human eye is referred to as a normal lens with a
magnification M = 1. The human eye’s focal length—the
distance from the center of the lens at the front of the
eye to the retina in the back of the eye—is about 17 mm
(0.67 inch) (Figure 4-2).
Most people see approximately the same FOV and magnification (M = 1). Specifically, the video lens and camera format corresponding to the M = 1 condition is a
25 mm FL lens on a 1-inch (diagonal) format camera, a
16 mm lens on a 2/3-inch format camera, a 12.5 mm lens
on a 1/2-inch camera, an 8 mm lens on a 1/3-inch camera, and a 6 mm lens on a 1/4-inch sensor. The 1-inch
format designation was derived from the development of
the original vidicon television tube, which had a nominal tube diameter of 1 inch (25.4 mm) and an actual
scanned area (active sensor size) of approximately 16 mm
in diameter. Figure 4-4 shows the FOV as seen with a lens
having magnifications of 1, 3, and 1/3 respectively.
MONITOR
MONITOR
MONITOR
FIGURE 4-4
NARROW
ANGLE
M=3
NORMAL
M=1
WIDE
ANGLE
M = 1/3
Lens FOV for magnifications of 3, 1, and 1/3
Lenses with much shorter FL used with these sensors
are referred to as wide-angle lenses and lenses with much
longer FL are referred to as narrow-angle (telephoto)
lens. Between these two are medium FL lenses. Telephoto
lenses used with video cameras act like a telescope: they
magnify the image viewed, narrow the FOV, and effectively
bring the object of interest closer to the eye. While there
is no device similar to the telescope for the wide-angle
example, if there were, the device would broaden the FOV,
allowing the eye to see a wider scene than is normal and
at the same time causing objects to appear farther away
from the eye. One can see this condition when looking
through a telescope backwards. This also occurs with the
automobile passenger side-view mirror, a concave mirror
that causes the scene image to appear farther away, and
therefore smaller than it actually is (de-magnified).
Just as your own eyes have a specific FOV—the scene
you can see—so does the video camera. The camera FOV
is determined by the simple geometry shown in Figure 4-5.
The scene has a width (W ) and a height (H ) and is at a
distance (D) away from the camera lens. Once the scene has
been chosen, three factors determine the correct FL lens
to use: (1) the size of the scene (H W ), (2) the distance
between the scene and camera lens (D), and (3) the camera
image sensor size (1/4-, 1/3-, or 1/2-inch format).
Lenses and Optics
75
SENSOR GEOMETRY
SENSOR
TUBE
VERTICAL = 3 UNITS HIGH
SOLID STATE
(CCD)
d = DIAGONAL
v = VERTICAL
v
HORIZONTAL = 4 UNITS WIDE
CCTV
CAMERA
d
h
h = HORIZONTAL
CAMERA SENSOR
FOV
LENS
D
D = DISTANCE FROM SCENE
TO LENS
SCENE
SENSOR SIZE
W
H
FORMAT
HORIZONTAL
WIDTH
VERTICAL HEIGHT
VERTICAL(v )
mm
inch
mm
inch
mm
inch
1"
16
0.63
12.8
0.50
9.6
0.38
2/3"
11
0.43
8.8
0.35
6.6
0.26
1/2"
8
0.31
6.4
0.25
4.8
0.19
1/3"
6
0.24
4.8
0.19
3.6
0.14
1/4"
4
3
0.16
0.12
3.2
2.4
0.13
2.4
1.8
0.1
0.07
1/6"
FIGURE 4-5
DIAGONAL(d ) HORIZONTAL(h)
0.09
Camera/lens sensor geometry and formats
4.2.1.1 Field-of-View Calculations
There are many tables, graphs, monographs, and linear
and circular slide rules for determining the angles and
sizes of a scene viewed at varying distances by a video
camera with a given sensor format and FL lens. One convenient aid in the form of transparent circular scales, called
a “Lens Finder Kit,” eliminates the calculations required
to choose a video camera lens (Section 4.2.5). Such kits
are based on the simple geometry shown in Figure 4-6.
Since light travels in straight lines, the action of a lens
can be drawn on paper and easily understood. Bear in
mind that while commercial video lenses are constructed
from multiple lens elements, the single lens shown in
Figure 4-6 for the purpose of calculation has the same
effective FL as the video lens. By simple geometry, the
scene size viewed by the sensor is inversely proportional to
the lens FL. Shown in Figure 4-6 is a camera sensor of horizontal width (h) and vertical height (v). For a 1/2-inch
CCD sensor, this would correspond to h = 64 mm and
v = 48 mm. The lens FL is the distance behind the lens at
which the image of a distant object (scene) would focus.
The figure shows the projected area of the sensor on the
scene at some distance D from the lens. Using the eye
analogy, the sensor and lens project a scene W wide × H
high (the eye sees a circle as did the original vidicon). As
with the human eye, the video lens inverts the image, but
the human brain and the electronics re-inverts the image
in the camera to provide an upright image. Figure 4-6
shows how to measure or calculate the scene size (W × H )
as detected by a rectangular video sensor format and lens
with horizontal and vertical angular FOVs H and V ,
respectively.
4.2.1.1.1 Tables for Scene Sizes vs. FL for 1/4-, 1/3-,
and 1/2-Inch Sensors
Tables 4-1, 4-2, and 4-3 give scene-size values for the 1/4-,
1/3-, and 1/2-inch sensors, respectively, as a function of
the distance from the camera to the object and the lens
FL. The tables include scene sizes for most available lenses
ranging from 2.1 to 150 mm FL.
To find the horizontal FOV H , we use the geometry of
similar triangles:
h
FL
=
W
D
W =
h
×D
FL
(4-1)
The horizontal angular FOV H is then derived as follows:
tan
h/2
H
=
2
FL
h
H
= tan−1
2
2 FL
h
H = 2 tan−1
2 FL
(4-2)
76
CCTV Surveillance
CAMERA LOCATION
SCENE LOCATION
SCENE
v = SENSOR VERTICAL HEIGHT
h = SENSOR HORIZONTAL WIDTH
θH = HORIZONTAL
ANGLE OF VIEW
FIXED FOCAL
LENGTH LENS
H
W
CAMERA
SENSOR
θV/2
θV = VERTICAL ANGLE OF VIEW
D
v
h
θH/2
FL
H = SCENE HEIGHT
W = SCENE WIDTH
FIGURE 4-6
Sensor, lens and scene geometry
LENS
FOCAL
LENGTH
(mm)
2.1
2.2
2.3
2.6
3.0
3.6
3.8
4.0
4.3
6.0
8.0
12.0
16.0
25.0
ANGULAR
FIELD OF
VIEW: H × V
(DEG.)
81.2 × 60.9
78.6 × 59.0
76.1 × 57.1
69.4 × 52
61.9 × 46.4
53.1 × 39.8
50.7 × 38.0
48.5 × 36.4
45.4 × 34.1
33.4 × 25.0
25.4 × 19.0
17.1 × 12.8
12.8 × 9.6
8.2 × 6.2
5
W ×H
8.6 × 6.4
8.2 × 6.1
7.8 × 5.9
6.9 × 5.2
6.0 × 4.5
5.0 × 3.8
4.7 × 3.6
4.5 × 3.4
4.2 × 3.1
3 × 2.3
2.3 × 1.7
1.5 × 1.1
1.1 × .8
.72 × .54
1/4 - INCH SENSOR FORMAT LENS GUIDE
CAMERA TO SCENE DISTANCE (D) IN FEET
WIDTH AND HEIGHT OF AREA (W × H ) IN FEET
10
20
30
40
50
75
W ×H
W ×H
W ×H
W ×H
W ×H
W ×H
69 × 51
86 × 64 129 × 96
17 × 12.9
51 × 39
34 × 26
65 × 49
16.4 × 12.2
82 × 63 123 × 92
49 × 37
33 × 25
15.6 × 11.8
78 × 59 117 × 86
62 × 47
31 × 23
47 × 35
13.9 × 10.4
69 × 52 104 × 78
55 × 42
28 × 21
42 × 31
36 × 27
48 × 36
60 × 45
90 × 68
12 × 9
24 × 18
10 × 7.5
40 × 30
50 × 38
75 × 57
30 × 23
20 × 15
28 × 21
9.5 × 7.1
38 × 28
47 × 36
71 × 54
19 × 14
9 × 6.8
36 × 27
45 × 34
68 × 51
27 × 20
18 × 14
8.4 × 6.3 16.7 × 12.5 25 × 19
33 × 25
42 × 31
63 × 47
6 × 4.5
24 × 18
30 × 23
45 × 35
18 × 13.5
12 × 9
4.5 × 3.4
23 × 17
35 × 26
13.5 × 10.1 18 × 13.5
9 × 6.8
3 × 2.2
12 × 9
15 × 11
23 × 17
9.0 × 6.8
6 × 4.4
2.3 × 1.7
9 × 6.8 11.2 × 8.4 17 × 13
6.8 × 5.1
4.5 × 3.4
1.4 × 1.1
5.8 × 4.3
7.2 × 5.4 10.8 × 8.1
4.3 × 3.2
2.9 × 2.1
NOTE: 1/4 - INCH LENSES ARE DESIGNED FOR 1/4 - INCH SENSOR FORMATS ONLY AND WILLNOT WORK ON
1/3 - INCH OR 1/2 - INCH SENSORS. LENS FOCAL LENGTHS ARE NOMINAL PER MANUFACTURERS’
LITERATURE.
ANGULAR FOV AND W × H ARE DERIVED FROM EQUATIONS 4 - 1 TO 4 - 4 AND VERTICAL FOV
FROM STANDARD 4:3 MONITOR RATIO: V = 0.75H.
Table 4-1
1/4-Inch Sensor FOV and Scene Sizes vs. FL and Camera-to-Scene Distance
100
W ×H
171 × 129
164 × 123
157 × 117
138 × 104
120 × 90
100 × 76
94 × 72
90 × 68
84 × 62
60 × 46
46 × 34
30 × 23
22 × 17
14.4 × 10.8
Lenses and Optics
1/3-INCH SENSOR FORMAT LENS GUIDE
LENS
FOCAL
LENGTH
(mm)
ANGULAR
FIELD OF
VIEW: H × V
(DEG.)
2.3
2.6
2.8
3.6
3.8
4.0
4.5
6.0
8.0
12.0
16.0
25.0
50.0
75.0
CAMERA TO SCENE DISTANCE (D) IN FEET
WIDTH AND HEIGHT OF AREA (W × H ) IN FEET
92.4 × 69.3
5
W ×H
10.4 × 7.8
10
W ×H
20.8 × 15.6
20
W ×H
41.6 ×3 1.2
30
W ×H
63 × 47
40
W ×H
83 × 62
85.4 × 64.1
81.2 × 60.9
67.4 × 50.5
64.6 × 48.4
61.9 × 46.4
56.1 × 42.1
43.6 × 32.7
33.4 × 25.0
26.6 × 20.0
17.1 × 12.8
11.0 × 8.2
5.5 × 4.1
3.7 × 2.8
9.2 × 6.9
8.6 × 6.5
6.7 × 5.0
6.3 × 4.7
6.0 × 4.5
5.3 × 4.0
4.0 × 3.0
3.0 × 2.3
2.0 × 1.5
1.5 × 1.2
.96 × .72
.48 × .36
.32 × .24
18.5 × 13.8
17.2 × 13
13.3 × 10
12.6 × 9.5
12 × 9
10.6 × 8
8.0 × 6
6 × 4.5
4.0 × 3.0
3.0 × 2.3
1.9 × 1.4
.96 × .72
.64 × .50
36.8 × 27.6
34.4 × 26
26.7 × 20
25 × 18.9
24 × 18
21.2 × 15.9
16 × 12
12 × 9
8.0 × 6.0
6.0 × 4.5
3.8 × 2.9
1.9 × 1.4
1.3 × .96
55 × 41
51 × 39
40 × 30
37.9 × 28.4
36 × 27
31.8 × 23.9
24 × 18
18 × 13.5
12.0 × 9.0
9.0 × 6.8
5.8 × 4.4
2.9 × 2.2
1.9 × 1.4
77 × 58
69 × 52
53 × 40
50.5 × 37.9
48 × 36
42.4 × 31.8
32 × 24
24 × 18
16 × 12
12.0 × 9.0
7.7 × 5.8
3.8 × 2.8
2.6 × 1.9
50
W ×H
104 × 78
75
W ×H
156 × 117
100
W ×H
208 × 156
92 × 69
86 × 65
67 × 50
63 × 47
60 × 45
53 × 40
40 × 30
30 × 22.5
20 × 15
15.0 × 11.3
9.6 × 7.2
4.8 × 3.6
3.2 × 2.4
138 × 104
129 × 98
101 × 75
95 × 71
90 × 68
80 × 60
60 × 45
45 × 34
30 × 23
23 × 17
14.4 × 10.8
7.2 × 5.4
4.8 × 3.6
184 × 138
172 × 130
134 × 100
123 × 92
120 × 90
106 × 80
80 × 60
60 × 45
40 × 30
30 × 22.5
19.2 × 14.4
9.6 × 7.2
6.4 × 4.8
NOTE: MOST 1/3 - INCH LENSES WILL NOT WORK ON 1/2 - INCH SENSORS BUT ALL WILL WORK ON ALL 1/4 - INCH SENSORS.
LENS FOCAL LENGTHS ARE NOMINAL PER MANUFACTURERS’ LITERATURE.
ANGULAR FOV AND W × H ARE DERIVED FROM EQUATIONS 4 - 1 TO 4 - 4 AND VERTICAL FOV FROM STANDARD 4:3 MONITOR
RATIO: V = 0.75H.
Table 4-2
1/3-Inch Sensor FOV and Scene Sizes vs. FL and Camera-to-Scene Distance
1/2-INCH SENSOR FORMAT LENS GUIDE
LENS
FOCAL
LENGTH
(mm)
ANGULAR
FIELD OF
VIEW: H × V
(DEG.)
1.4
2.6
3.5
3.6
3.7
4.0
4.2
4.5
4.8
6.0
7.5
8.0
12.0
16.0
25.0
50.0
75.0
150.0
133 × 100
101.8 × 76.4
84.9 × 63.7
83.3 × 62.5
81.7 × 61.3
77.3 × 58.0
74.6 × 56.0
70.8 × 53.1
67.4 × 50.5
56.1 × 42.1
46.2 × 34.7
43.6 × 32.7
29.9 × 22.4
22.6 × 17.0
14.6 × 10.9
7.3 × 5.5
4.9 × 3.7
2.4 × 1.8
CAMERA TO SCENE DISTANCE (D) IN FEET WIDTH AND HEIGHT
OF AREA (W × H ) IN FEET
5
W ×H
23 × 17
12.3 × 9.2
9.1 × 6.9
8.9 × 6.7
8.6 × 6.5
8.0 × 6.0
7.6 × 5.7
7.1 × 5.3
6.7 × 5.0
5.3 × 4.0
4.3 × 3.2
4.0 × 3.0
2.7 × 2.0
2.0 × 1.5
1.3 × 1.0
.64 × .48
.43 × .32
.21 × .16
10
W ×H
46 × 34
24.6 × 18
18.2 × 13.8
17.8 × 13.4
17.2 × 13.0
16.0 × 12.0
15.2 × 11.4
14.2 × 10.6
13.4 × 10.0
10.6 × 8.0
8.6 × 6.4
8.0 × 6.0
5.3 × 4.0
4.0 × 1.5
2.6 × 2.0
1.3 × 1.0
.85 × .64
.43 × .32
20
W ×H
91 × 69
49 × 37
37 × 28
36 × 27
35 × 26
32 × 24
30 × 23
28 × 21
27 × 20
21 × 16
17.1 × 12.8
16 × 12
10.7 × 8
8×6
5.1 × 3.8
2.6 × 1.9
1.7 × 1.3
.85 × .64
30
W ×H
137 × 103
74 × 55
55 × 41
53 × 40
52 × 39
48 × 36
48 × 34
43 × 32
40 × 30
32 × 24
26 × 19
24 × 18
16 × 12
12 × 9
7.7 × 5.8
3.8 × 2.9
2.6 × 1.9
1.3 × .96
40
W ×H
183 × 137
98 × 74
73 × 55
71 × 53
69 × 52
64 × 24
61 × 46
57 × 43
53 × 40
43 × 32
34 × 26
32 × 24
21.3 × 16
16 × 12
10.2 × 7.7
5.1 × 3.8
3.4 × 2.6
1.7 × 1.3
50
W ×H
228 × 171
123 × 92
91 × 69
89 × 67
86 × 65
80 × 60
76 × 57
71 × 53
67 × 50
53 × 40
43 × 32
40 × 30
27 × 20
20 × 15
12.8 × 9.6
6.4 × 4.8
4.3 × 3.2
2.1 × 1.6
NOTE: ALL 1/2-INCH FORMAT LENSES WILL WORK ON 1/3- AND 1/4 -INCH SENSORS.
LENS FOCAL LENGTHS ARE NOMINAL PER MANUFACTURERS’ LITERATURE.
ANGULAR FOV AND W × H ARE DERIVED FROM EQUATIONS 4-1 TO 4-4.
Table 4-3
1/2-Inch Sensor FOV and Scene Sizes vs. FL and Camera-to-Scene Distance
75
W ×H
342 × 257
185 × 138
137 × 104
134 × 101
129 × 98
120 × 90
114 × 86
107 × 80
101 × 75
80 × 60
65 × 48
60 × 45
41 × 30
30 × 23
19 × 14
6.5 × 4.8
3.2 × 2.4
9.6 × 7.2
100
W ×H
457 × 348
246 × 184
182 × 138
178 × 134
172 × 130
160 × 120
156 × 114
142 × 107
134 × 100
106 × 80
86 × 64
80 × 60
53 × 40
40 × 30
25.6 × 19.2
12.8 × 9.6
8.6 × 6.4
4.3 × 3.2
77
78
CCTV Surveillance
For the vertical FOV, similar triangles give:
v
FL
=
H
D
H=
v
×D
FL
(4-3)
The vertical angular FOV V is then derived from the
geometry:
tan
v/2
v
=
2
FL
v
v
= tan−1
2
2 FL
v
v = 2 tan−1
2 FL
will fit onto the sensor. Lens magnification is measured
relative to the eye which is defined as a normal lens. The
eye has approximately a 17-mm FL and is equivalent to a
25-mm FL lens on a 1-inch format camera sensor.
Therefore, the magnification of a 1-inch (16-mm format) sensor is
Lens focal length (mm)
Sensor diagonal (mm)
Ms =
Ms
(4-4)
4.2.1.1.2 Tables for Angular FOV vs. FL for 1/4-, 1/3-,
and 1/2-Inch Sensor Sizes
Table 4-4 shows the angular FOV obtainable with 1/4 -,
1/3-, 1/2-, and 2/3-inch sensors with some standard lenses
from 1.4 to 150 mm FL. The values of angular FOV in
Table 4-4 can be calculated from Equations 4-2 and 4-4.
4.2.1.2 Lens and Sensor Formats
Fixed focal length lenses must be used with either the
image sensor size (format) for which they were designed
or with a smaller sensor size. They cannot be used with
larger sensor sizes because unacceptable image distortion
and image darkening (vignetting) at the edges of the
image occurs. When a lens manufacturer lists a lens for
a 1/3-inch sensor format, it can be used on a 1/4-inch
sensor but not on a 1/2-inch sensor without producing
image vignetting. This problem of incorrect lens choice
for a given format size occurs most often when a C or
CS mount 1/3-inch format lens is incorrectly used on a
1/2-inch format camera. Since the lens manufacturer does
not “over design” the lens, that is, make glass lens element
diameters larger than necessary, check the manufacturer’s
specifications for proper choice.
4.2.2 Magnification
The overall magnification from a specific camera, lens,
and monitor depends on three factors: (1) lens FL,
(2) camera sensor format, and (3) the monitor size (diagonal). Video magnification is analogous to film magnification: the sensor is equivalent to the film negative, and the
monitor is equivalent to the photo print.
4.2.2.1 Lens–Camera Sensor Magnification
The combination of the lens FL and the camera sensor
size defines the magnification Ms at the camera location.
For a specific camera, the sensor size is fixed. Therefore,
no matter how large the image from the lens is at the
sensor, the camera will see only as much of the image as
1
FL
inch =
16 mm
(4-5)
For 2/3 inch (11-mm format) the magnification is
Ms
FL
11 mm
=
2/3 inch
(4-6)
For 1/2 inch (8-mm format) the magnification is
Ms
1/2 inch
=
FL
8 mm
(4-7)
For 1/3 inch (5.5-mm format) the magnification is
Ms
1/3 inch
=
FL
5.5 mm
(4-8)
For 1/4 inch (4-mm format) the magnification is
Ms
1/4 inch
=
FL
4 mm
(4-9)
Example: From Equation 4-7, a 16-mm FL lens on a 21 -inch
format camera would have a magnification of
Ms
1/2 inch
=
FL
16 mm
=
=2
8 mm
8 mm
4.2.2.2 Monitor Magnification
When the camera image is displayed on the CCTV monitor, a further magnification of the object scene takes
place. The monitor magnification Mm is equivalent to the
ratio of the monitor diagonal (dm ) to the sensor diagonal
(ds ) or
Mmonitor = Mm =
dm
ds
(4-10)
Example: From Equation 4-10, for a 9-inch diagonal monitor (dm = 9 inches) and a 1/2 sensor format (ds = 8 mm =
0315 inch)
Mm =
9
= 2857
0315
LENS
FOCAL
LENGTH
(mm)
1.4
2.1
2.2
2.3
2.6
2.8
3.0
3.5
3.6
3.7
3.8
4.0
4.2
4.3
4.5
4.8
6.0
7.5
8.0
12.0
16.0
25.0
50.0
75.0
150.0
MAXIMUM
IMAGE
FORMAT
OPTICAL
SPEED:
f/#
1/2
1/4
1/3
1/3
1/2
1/3
1/4
1/2
1/2
1/2
1/3
1/2
1/2
1/4
1/2
1/2
1/2
2/3
2/3
2/3
2/3
2/3
2/3
2/3
1/2
1.4
1.0
1.2
1.4
1.6
1.2
1.0
1.4
1/6
1.6
1.4
1.2
1.6
1.4
1.4
1.4
1.0
1.4
1.2
1.2
1.4
1.4
1.4
1.4
1.6
LENS
MOUNT
TYPE
CS
CS
CS
CS
CS
CS
CS
CS, C
CS, C
CS
CS
CS
CS
CS
CS, C
CS, C
CS
CS, C
CS
CS, C
CS, C
CS, C
CS, C
CS, C
CS, C
1/4 INCH SENSOR
HORIZONTAL
101
91
93
89
72
71
65
59
54
53
51
50
49
42
44
39
33
26
25
28
13
8
4.1
2.8
1.4
VERTICAL
76
70
69
67
54
53
49
44
41
40
39
37
36
35
34
29
25
20
19
13
10
6
3.1
2.1
1.1
CAMERA ANGULAR FIELD OF VIEW (FOV) (DEGREES)
1/3 INCH SENSOR
1/2 INCH SENSOR
HORIZONTAL
VERTICAL
HORIZONTAL
VERTICAL
135
101
180
135
113
100
96
85
75
72
128
96
78
72
71
68
65
64
59
54
53
51
50
49
104
93
94
78
71
70
89
87
67
65
59
52
57
35
33
24
17
12
5.5
3.7
1.8
45
39
43
26
25
28
13
9
4.1
2.8
1.4
79
69
57
46
45
30
22
15
7.3
4.8
2.4
59
52
43
35
34
23
17
11
5.5
3.6
1.8
2/3 INCH SENSOR
HORIZONTAL
VERTICAL
96
74
58
39
31
20
10
6.8
3.3
45
29
23
15
7.5
5
2.5
NOTE: ALL FOCAL LENGTHS AND ANGULAR FOVs BASED ON MANUFACTURER’S LITERATURE.
ALL THE LARGER FORMAT LENSES CAN BE USED ON SMALLER FORMAT SENSORS.
LENSES ARE ALSO AVAILABLE HAVING SMALLER FORMATS AND LOWER f/#S THAN THOSE LISTED.
Lenses and Optics
Table 4-4 Representative Fixed Lenses Angular FOV vs. Sensor Format and Lens Focal Length
79
80
CCTV Surveillance
For vertical scene height, using Equation 4-1:
4.2.2.3 Combined Camera and Monitor
The combined lens, sensor, and monitor magnification is
v
×D
FL
4.8 mm
× 25 ft = 9.6 ft
H=
12.5 mm
Scene height = H =
M = Ms × Mm
For the example above and Equation 4-11, the overall magnification of the 8-mm FL lens, 1/2-inch format camera,
and a 9-inch monitor is
M = Ms × Mm = 2 × 2857 = 5714
Table 4-5 summarizes the magnification for the entire
video system, for a 9- and 17-inch monitor and various
lenses and camera formats. It should be noted that increasing the magnification by using a larger monitor does not
increase the information in the scene; it only increases
the size of the displayed picture and permits viewing the
monitor from a greater distance.
4.2.3 Calculating the Scene Size
Equations 4-1 and 4-3 are used to calculate scene size. For
example, calculate the horizontal and vertical scene size as
seen by a 1/2-inch CCD sensor using a 12.5 mm FL lens at
a distance D = 25 ft. A 1/2-inch sensor is 6.4 mm wide and
4.8 mm high. From Equation 4-1 for horizontal scene width:
h
Scene width = W =
×D
FL
6.4 mm
× 25 ft = 12.8 ft
W =
12.5 mm
CAMERA FORMAT
(inch/mm)
MONITOR SIZE
(inch)
4.2.3.1 Converting One Format to Another
To obtain scene sizes (width and height) for a 1/6-inch
sensor, divide all the scene sizes in the 1/3-inch table
(Table 4-2) by 2. For a 2/3-inch sensor, multiple all the
scene sizes in the 1/3-inch table (Table 4-2) by 2.
Understanding Tables 4-1, 4-2, and 4-3 makes it easy to
choose the right lens for the required FOV coverage. As
an example, choose a lens for viewing all of a building
15 feet high by 20 feet long from a distance of 40 feet
with a 1/2-inch format video camera (Figure 4-7). From
Table 4-3, a 12-mm FL lens will just do the job.
If a 1/4-inch format video camera were used, a lens with
an FL of 16 mm would be needed (from Table 4-4, a scene
16.7 feet high by 22.5 feet wide would be viewed).
If a 1/3-inch format video camera were used, a lens with
an FL of 9 mm would be used (from Table 4-2, a scene
15.2 feet high by 20 feet wide would be viewed).
4.2.4 Calculating Angular FOV
Equations 4-2 and 4-4 are used to calculate the horizontal
and vertical angular FOV of the lens–camera combination.
Table 4-4 shows the angular FOV obtainable with some
LENS FOCAL LENGTH
mm
9
1/6
(0.11/2.75)
17
9
1/4
(0.15/4.0)
17
9
1/3
(0.22/5.5)
17
9
1/2
(0.31/8.0)
17
TOTAL MAGNIFICATION
2.4
72.7
30
909.1
2.4
137.4
30
1717.2
2.6
37.3
25.0
358.3
2.6
70.4
25.0
676.8
3.8
28.7
50.0
377.0
3.8
54.0
50.0
712.2
4.8
17.1
75.0
267.8
4.8
32.4
75.0
506.0
ALL VALUES BASED ON SENSOR AND MONITOR DIAGONAL
MAGNIFICATION = Ms × Mm, WHERE Ms =
MONITOR DIAGONAL
LENS FL
and, Mm =
SENSOR DIAGONAL
SENSOR DIAGONAL
EXAMPLE: 1/3-inch FORMAT SENSOR, 3.8 mm FL LENS (0.15 inch), AND 17-inch MONITOR
M=
Table 4-5
3.8 mm
17 inch
= 54
×
0.22 inch
5.5 mm
Monitor Magnification vs. Camera/Monitor Size and Lens Focal Length
Lenses and Optics
81
SCENE LOCATION
CAMERA LOCATION
H = SCENE HEIGHT
LENS
SENSOR
SIDE VIEW
ANGULAR
WIDTH
v
H = 22.5 ft
θV
FL
D = 40 ft
VERTICAL FOV: v/FL = H/D
W = SCENE WIDTH
LENS
SENSOR
TOP(PLAN)
VIEW
ANGULAR
WIDTH
h
θH
W = 30 ft
FL
D = 40 ft
HORIZONTAL FOV: h/FL = W/D
FIGURE 4-7
Calculating the focal length for viewing a building
standard lenses from 2.6 to 75 mm focal length. For the
previous example, calculate the horizontal and vertical
angular FOVs H and V for a 1/2-inch CCD sensor using
a 12.5 mm FL lens. The distance need not be supplied,
since an angular measure is independent of distance.
From Equation 4-2, for horizontal angular FOV:
ratio of the sensor size. Rule of thumb: for a given lens,
angular FOV increases for larger sensor size, decreases for
smaller sensor size.
4.2.5 Lens Finder Kit
h/2 66 mm/2
=
= 0264
tan H =
2
FL
125 mm
H
= 148
2
H = 296
From Equation 4-4 for vertical angular FOV:
tan
v/2 48 mm/2
v
=
=
= 0192
2
FL
125 mm
v
= 109
2
v = 218
Table 4-4 summarizes angular FOV values for some standard lenses from 1.4 to 150 mm FL lenses used on the
1/4-, 1/3-, 1/2- and 2/3-inch sensors. To obtain the angular FOV for sensor sizes, multiply or divide the angles by the
Tables and slide rules for finding lens angular FOVs
abound. Over the years many charts and devices have been
available to simplify the task of choosing the best lens for
a particular security application. Figure 4-8 shows how to
quickly determine the correct lens for an application using
the Lens Finder Kit (copyright H. Kruegle).
There is a separate scale for each of the three camerasensor sizes: 1/4-, 1/3-, 1/2-inch (the 1/4- and 1/3-inch
are shown). The scale for each camera format shows the
FL of standard lenses and the corresponding angular horizontal and vertical FOVs that the camera will see.
To use the kit, the plastic disk is placed on the facility
plan drawing and the lens FL giving the desired camera
FOV coverage is chosen. For example, a 1/4-inch format
camera is to view a horizontal FOV (H ) in a front lobby
30 feet wide at a distance of 30 feet from the camera
(Figure 4-9). What FL lens should be used?
82
CCTV Surveillance
THE LENS FINDER KIT USES THREE TRANSPARENT PROTRACTOR DISKS TO HELP
CHOOSE THE BEST LENS WHEN USING THE 1/4-, 1/3- AND 1/2-INCH CCTV CAMERA
FORMATS WITH C OR CS MOUNTS. THE DISKS ARE UNIVERSAL AND CAN BE USED
ON ANY SCALE DRAWING. HOW TO USE:
1. SELECT THE DISK TO MATCH THE CAMERA FORMAT: 1/4-, 1/3- OR 1/2-INCH.
2. USING A SCALE DRAWING OF THE FLOOR PLAN (ANY SCALE), PLACE THE CENTER HOLE OF
THE DISK AT THE PROPOSED CAMERA LOCATION ON THE FLOOR PLAN.
3. ROTATE THE DISK UNTIL ONE SEGMENT (PIE SECTION) TOTALLY INCLUDES THE HORIZONTAL
FIELD OF VIEW REQUIRED.
4. USE THE FOCAL LENGTH LENS DESIGNATED IN THE SEGMENT ON THE DISK.
5. IF THE SCALE DRAWING INCLUDES AN ELEVATION VIEW, FOLLOW STEPS 1 THROUGH 4 AND
USE THE VERTICAL ANGLE DESIGNATED IN EACH PIE SEGMENT FOR THE VERTICAL FIELD
OF VIEW OF THE LENS.
88
138
8
17
22
DA WORKS
Y
328
8 mm
13
6
mm
6.0
A
7
7 28
3.
458
© H. KRUEGLE
200
AVID
SECURITY A
W
IZ
HO
RI
Z
6m
m
5 T.
R
VE
648 HOR
IZ
48
4.
37
3.6 mm
54 HORIZ
4
VER 1
T.
VE 488
RT
.
8
4 T.
R
VE
4.0 m
m
18
8m
m
26
KS
OR
TIC
VER
R
6 mm
m
718L
DE
IN
m
0m
.8
V
64
HO
R
3.0
m
49
HO
RI
ZO
N
2
LE N C C
SF
m
"
1 /3
R
DE
IN
AL
TIC
VER
© H. KRUEGLE 2007
12m
1m6m
T
T
70
8
948
2mm5
m
LEN C C
SF
11
17
L
TA
2.2
m
"
1 /4
35
HO
RI
ZO
N
m
m V
24
93
L
TA
12m
8
1m6m 2m5m
SECU AVI
RI
T
13
348
17
248
10
1/3 INCH
FORMAT
m
1/4 INCH
FORMAT
NOTE: FOR 2/3- AND 1/2-INCH FORMATS MULTIPLY THE 1/3- AND 1/4-INCH SCALE FOV'S BY 2
FIGURE 4-8
Choosing a lens with the Lens Finder Kit
To find the horizontal angular FOV H , draw the following lines to scale on the plan: one line to a distance 30 feet
from the camera to the center of the scene to be viewed,
a line 30 feet long and perpendicular to the first line, and
two lines from the camera location to the endpoints of
the second 30-foot line. Place the 1/4-inch Lens Finder
Kit on the top view (plan) drawing with its center at the
camera location and choose the FL closest to the horizontal angle required. A 3.6 mm FL lens is closest. This lens
will see a horizontal scene width of 30 feet. Likewise for
scene height: using the side-view (elevation) drawing, the
horizontal scene height is 22.5 feet.
4.2.6 Optical Speed: f-number
The optical speed or f-number (f/#) of a lens defines its
light-gathering ability. The optical speed of a lens—how
much light it collects and transmits to the camera sensor—
is defined by a parameter called the f-number (f/#).
As the FL of a lens becomes longer, its optical aperture
or diameter (d) must increase proportionally to keep the
f-number the same. The f-number is related to the FL
and the lens diameter (clear aperture) d by the following
equation:
f/# =
FL
d
(4-11)
For example, an f/2.0 lens transmits four times as much
light as an f/4.0 lens. The f-number relationship is analogous to water flowing through a pipe. If the pipe diameter
is doubled, four times as much water flows through it.
Likewise, if the f-number is halved (i.e. if the lens diameter
is doubled), four times as much light will be transmitted
through the lens.
In practice the f-number obtained is worse than this
because of various losses caused by imperfect lens transmission, reflection, absorption, and other lens imaging properties. The amount of light (I ) collected and transmitted
Lenses and Optics
83
WALL
30.6 ft
LENS
HORIZONTAL
FOV
PLAN
TOP
VIEW
W = 30.6 ft
30.6 ft
θH
30 ft
30 ft
H = 23.4 ft
CAMERA
D = 30 ft
LENS
CAMERA
SIDE
VIEW
θV
23.4 ft
FORMAT: 1/4 INCH
30 ft
FIGURE 4-9
Determining lobby lens horizontal and vertical FOVs
through the lens system varies inversely as the square of
the lens f-number (K = constant):
I=
K
f /#2
Long-FL lenses are larger (and costlier) than short-FL
lenses, due to the cost of the larger optical elements. It
can be seen from Equation 4-11 that the larger the d is
made, the smaller the f/# is, i.e. more light gets to the
camera sensor. The more light the lens can collect and
transfer to the camera image sensor the better the picture quality: a larger lens permits the camera to operate
at lower light levels. This light-gathering ability depends
on the size (diameter) of the optics: the larger the optics
the more the light that can be collected.
Most human eyes have the same size lens (approximately 7 mm lens diameter). In video systems, however,
the lens size (the diameter of the front lens) varies over
a wide range. The optical speed of video lenses varies significantly: it varies as the square of the diameter of the
lens. This means a lens having a diameter twice that of
another will pass four times as much light through it. Like
a garden hose, when the diameter is doubled, the flow is
quadrupled (Figure 4-10).
The more the light passing through a lens and reaching
the video sensor the better the contrast and picture image
quality. Lenses with low f-numbers, such as f/1.4 or f/1.6,
pass more light than lenses with high f-numbers. The lens
optical speed is related to the FL and diameter by the
equation f/# = focal length/diameter. So the larger the
FL given the same lens diameter, the “slower” the lens
(less light reaches the sensor). A slow lens might have an
f-number of f/4 or f/8.
Most lenses have an iris ring usually marked with numbers such as 1.4, 2.0, 2.8, 4.0, 5.6, 8.0, 11, 16, 22, C, representing optical speed, f-numbers, or f-stops. The difference
between each of the iris settings represents a difference of
a factor of 2 in the light transmitted by the lens. Opening
the lens from, say, f/2.0 to f/1.4 doubles the light transmitted. Only half the light is transmitted when the iris
opening is reduced from, say, f/5.6 to f/8. Changing the
iris setting two f-numbers changes the light by a factor of
4 (or 1/4), and so on. Covering the f/# range from f/1.4
to f/22 spans a light-attenuation range of 256 to 1. The
C designation on the lens indicates when the lens iris is
closed and no light is transmitted.
In general, faster lenses collect more light energy
from the scene, are larger, and are more expensive. In
calculating the overall cost of a video camera lens system,
a more expensive, fast lens often overrides the higher cost
incurred if a more sensitive camera is needed or additional
lighting must be installed.
84
CCTV Surveillance
WATER FLOW ANALOGY
LENS LIGHT TRANSMISSION
D = DIAMETER
f / # = OPTICAL SPEED
HOSE
LENS
f/4
D=1
LIGHT TRANSMISSION: 4X
f/2
WATER FLOW: 4X
D=2
FIGURE 4-10
Light transmission through a lens
4.2.7 Depth of Field
The depth of field in an optical system is the distance
that an object in the scene can be moved toward or away
from the lens and still be in good focus. In other words,
it is the range of distance toward and away from the camera lens in which objects in the scene remain in focus.
Ideally this range would be very large: say, from a few
feet from the lens to hundreds of feet, so that essentially
all objects of interest in the scene would be in sharp
focus. In practice this is not achieved because the depth
of field is: (1) inversely proportional to the focal length,
and (2) directly proportional to the f-number. Medium to
long FFL lenses operating at low f-numbers—say, f/1.2
to f/4.0—do not focus sharp images over their useful
range of from 2 or 3 feet to hundreds of feet. Long
focal length lenses—say, 50–300 mm—have a short depth
of field and can produce sharp images only over short
distances and must be refocused manually or automatically (auto-focus) when viewing objects at different scene
distances.
When these lenses are used with their iris closed down
to, say, f/8 to f/16, the depth of field increases significantly
and objects are in sharp focus at almost all distances in the
scene. Short focal length lenses (27–5 mm) have a long
depth of field. They can produce sharp images from a few
feet to 50–100 feet even when operating at low f-numbers
4.2.8 Manual and Automatic Iris
The lens iris is either manually or automatically adjusted to
optimize the light level reaching the sensor (Figure 4-3).
The manual iris is adjusted with a ring on the lens. The
auto-iris uses an internal mechanism and motor (or galvanometer) to adjust the iris.
4.2.8.1 Manual Iris
The manual-iris video lens has movable metal “leaves”
forming the iris. The amount of light entering the camera is determined by rotating an external iris ring, which
opens and closes these internal leaves. Figure 4-11 shows
a manual iris FFL lens and the change in light transmitted
through it at different settings of the iris.
Solid-state CCD and CMOS camera sensors can operate over wide light-level changes with manual-iris lenses
but require automatic-iris lenses when used over their
full light-level range, that is, from bright sunlight to lowlevel nighttime lighting. Some solid-state cameras use
electronic shuttering (Section 5.5.3) and do not require
an automatic-iris lens.
4.2.8.2 Automatic-Iris Operation
Automatic-iris lenses have an electro-optical mechanism
whereby the amount of light passing through the lens is
adjusted depending on the amount of light available from
the scene and the sensitivity of the camera.
The camera video signal provides the information used
for adjusting the light passing through the lens. The system works something like this: if a scene is too bright for
the camera, the video signal will be strong (large in amplitude). This large signal will activate a motor or galvanometer that causes the lens iris circular opening to become
smaller in diameter, thereby reducing the amount of light
reaching the camera. When the amount of light reaching the camera produces a predetermined signal level,
the motor or galvanometer in the lens stops and maintains that light level through the lens. Likewise if too little
light reaches the camera, the video camera signal level is
small and the automatic-iris motor or galvanometer opens
up the iris diaphragm, allowing more light to reach the
camera. In both the high and the low light level conditions the automatic-iris mechanism produces the best
Lenses and Optics
85
FOCUSING RING
MANUAL IRIS
f/# MARKINGS
LENS
f/#
LIGHT TRANSMISSION
1
1.4
2.0
2.8
4.0
5.6
8.0
11.0
16.0
22.0
C
1
1/2
1/4
1/8
1/16
1/32
1/64
1/128
1/256
1/512
0
NOTE: REFERENCE LIGHT LEVEL = 1 AT f/1.4
C = LENS IRIS FULLY CLOSED
EACH INCREASE OR DECREASE IN f/# REPRESENTS ONE f STOP
FIGURE 4-11
Lens f/# vs. light transmission
contrast picture. Automatic-iris lenses are available to compensate the full range of light, from bright sunlight to
darkness.
There are two types of automatic-iris lenses: direct drive
and video drive. The two methods used to control these
two lens types are: DC motor (or galvanometer) drive or
video signal drive. With the DC drive method the camera
has all the electronics and directly drives the DC motor
in the lens with a positive or a negative signal to open
and close the iris depending on the light level. With the
video method the camera video signal drives the electronics in the lens which then drives the DC motor (or galvanometer) in the lens. Figure 4-12 shows some common
automatic-iris lenses.
A feature available on some automatic-iris lenses is
called average-peak response weighting which permits optimizing the picture still further based on the variation
in lighting conditions within the scene. Scenes with
high-contrast objects (bright headlight, etc.) are better compensated for by setting the automatic-iris control to peak, so that the lens ignores (compensates
for) the bright spots and highlights in the scene (see
Section 5.5.5). Low-contrast scenes are better compensated by setting the control to average. Figure 4-13 illustrates some actual scenes obtained when these adjustments are made. Automatic-iris lenses should only be used
with cameras having a fixed video gain in their system.
Automatic-iris lenses are more expensive than their manual counterparts, with the price ratio varying by about two
or three to one.
4.2.9 Auto-Focus Lens
Auto-focus lenses were originally developed for the consumer camcorder market and are now available to the
security market (similar to the solid-state sensor evolution). There are two types of auto-focus techniques in use.
One auto-focus system uses a ranging (distance measuring) means to automatically focus the scene image onto
the sensor. A second type analyzes the video signal in by
means of DSP electronics, and forces the lens to focus
on the target in the scene. The function of both these
types of systems is to keep objects of interest in focus on
the camera sensor even though they move toward or away
from the lens. These lenses are particularly useful when
a person (or vehicle) enters a camera FOV and moves
toward or away from the camera. The auto-focus lens
changes focus from the surrounding scene and focuses
on the moving object (automatically) to keep the moving object in focus. Various types of automatic-focusing
techniques are used, including: (1) active IR ranging, (2)
ultrasonic wave, (3) solid-state triangulation, and (4) video
signal DSP.
86
CCTV Surveillance
(A) DC MOTOR IN LENS, ELECTRONICS IN CAMERA
FIGURE 4-12
(B) VIDEO ELECTRONICS IN CAMERA
Automatic-iris fixed focal length (FFL) lenses
(A) HIGH CONTRAST SCENES OPTIMIZED USING
MEDIUM RESPONSE WEIGHTING
(B) NORMAL CONTRAST SCENES OPTIMIZED USING
PEAK RESPONSE WEIGHTING
(C) LOW CONTRAST SCENES OPTIMIZED USING
AVERAGE RESPONSE WEIGHTING
FIGURE 4-13
Automatic-iris enhanced video scenes
4.2.10 Stabilized Lens
A stabilized lens is used when it is necessary to remove
unwanted motion of the lens and camera with respect to
the scene being viewed. Applications for stabilized lenses
include handheld cameras, cameras on moving ground
vehicles, airborne platforms, ships, and cameras on towers
and buildings. Stabilized lenses can remove significant
image vibration (blurring) in pan/tilt mounted cameras
that are buffeted by wind or the motion caused by the moving vehicle. The stabilized lens system has movable optical
components and/or active electronics that compensate for
(move in the opposite direction to) the relative motion
between the camera and the scene. An extreme example
Lenses and Optics
of a stabilized video camera system is the video image from
a helicopter. The motion compensation results in a steady,
vibration free scene image on the monitor. Figure 4-14
shows a stabilized security lens and samples of pictures
taken with and without the stabilization on.
4.3 FIXED FOCAL LENGTH LENS
Video lenses come in many varieties, from the simple,
small, and inexpensive mini pinhole and “bottle cap”
lenses to complex, large, expensive, motorized, automaticiris zoom lenses. Each camera application requires a specific scene to be viewed and a specific intelligence to be
extracted from the scene if it is to be useful in a security application. Monitoring a small front lobby or room
to see if a person is present may require only a simple
lens. It is difficult, however, to determine the activity of
a person 100–200 feet away in a large showroom. Apprehending a thief or thwarting a terrorist may require a
high-quality, long FL zoom lens and a high resolution
camera mounted on a pan/tilt platform. Covert cameras
using pinhole lenses are often used to uncover internal
theft, shoplifting, or other inappropriate or criminal activity. They can be concealed in inconspicuous locations,
installed quickly, and moved on short notice. The following sections describe common and special lens types used
in video surveillance applications.
The majority of lenses used in video applications are
FFL lenses. Most of these lenses are available with a manual focusing ring to adjust the amount of light passing
through the lens and reaching the image sensor. The very
short FL lenses (less than 6 mm) often have no manual
iris. FFL lenses are the workhorses of the industry. Their
attributes include low cost, ease of operation, and long life.
Most FFL lenses are optically fast and range in speed from
f/1.2 to f/1.8, providing sufficient light for most cameras
to produce an excellent quality picture. The manual-iris
lenses are suitable for medium light-level when used with
most solid-state cameras.
Most FFL lenses have a mount which in the industry for
attaching the lens to the camera is called a C or CS mount
(Figure 4-15).
The C mount has been a standard in the CCTV industry
for many years while the CS mount was introduced in the
mid-1990s to match the trend toward smaller camera sensor formats and their correspondingly smaller lens requirements. The C and CS mount has a 1 inch 32 threads per
inch thread. Most security surveillance cameras are now
manufactured with a CS mount and supplied with a 5 mm
thick spacer adapter ring which allows a C mount lens to
be attached to the CS mount camera. The C mount focuses
the scene image 0.69 inches (17.5 mm) behind the lens
onto the camera sensor. The CS mount focuses the scene
image 0.394 inches (10 mm) behind the lens onto the
sensor. Commonly used CS mount FLs vary from 2.5 mm
(A) LENS
(C) UNSTABILIZED IMAGE
(B) LENS
FIGURE 4-14
Stabilized lenses and results
87
(D) STABILIZED IMAGE
88
CCTV Surveillance
C MOUNT
CS MOUNT
1"-32 TPI
1"-32 TPI
CCD/CMOS
SENSOR
CCD/CMOS
SENSOR
MECHANICAL LENS GEOMETRY
BACK FOCAL
DISTANCE
IMAGE
PLANE
SCENE
MECHANICAL BACK
FOCAL DISTANCE
17.526 mm
(0.69")
12.5 mm
(0.492")
FLANGE BACK
FOCAL LENGTH (F,f )
TPI = THREADS PER INCH
FIGURE 4-15
C and CS mount lens mounting dimensions
(wide-angle) to 200 mm (telephoto). The C mount is also
used for long FL lenses having large physical dimensions.
Large optics are designed to be used with almost any sensor format size from 1/4 to 1 inch. Lenses with FLs longer
than approximately 300 mm are large and expensive. As
the FL becomes longer the diameter of the lens increases
and costs escalate accordingly. Most FFL lenses are available in a motorized or automatic-iris version. These are
necessary when they are used with LLL ICCD cameras in
daytime and nighttime applications where light level must
be controlled via an automatic iris or neutral density filters
depending on the scene illumination.
With the widespread use of smaller cameras and lenses
a new set of lens–camera mounts developed. They were
not given any special name but are referred to as 11 mm,
12 mm (the most common), and 13 mm mounts. The
dimensions refer to the diameter of the thread on the lens
and the camera mount. Figure 4-16 shows these lens
mounts. Note that the threads are not all the same (the
13 mm mount is different from the 11 mm and 12 mm).
4.3.1 Wide-Angle Viewing
While the human eye has peripheral vision and can detect
the presence and movement of objects over a wide angle
THREAD: 13 mm DIA. × 1.0 mm PITCH
(160 ), the eye sees a focused image in only about the
central 10 of its FOV. No video camera has this unique eye
characteristic, but a video system’s FOV can be increased
(or decreased) by replacing the lens with one having a
shorter (or longer) FL. The eye cannot change its FOV
without the use of external optics.
Choosing different FL lenses brings trade-offs: reducing
the FL increases the FOV but reduces the magnification,
thereby making objects in the scene smaller and less discernible (i.e. decreasing resolution). Increasing the FL has
the opposite effect.
To increase the FOV of a CCTV camera, a short-FL
lens is used. The FOV obtained with wide-angle lenses
can be calculated from Equations 4-1, 4-2, 4-3, and 4-4, or
by using Table 4-4, or the Lens Finder Kit. For example,
substituting an 8 mm FL, wide-angle lens for a 16 mm lens
on any camera doubles the FOV. The magnification is
reduced to one-half, and the camera sees “twice as much
but half as well.” By substituting a 4 mm FL lens for the
16 mm lens, the FOV quadruples. We see sixteen times as
much scene area but one-fourth as well.
A 2.8 mm FL lens is an example of a wide-angle lens; it
has an 82 horizontal by 67 vertical FOV on a 1/3-inch
sensor. A super wide FOV lens for a 1/2-inch sensor is
the 3.5 mm FL lens, with an FOV approximately 90 horizontal by 75 vertical. Using a wide-angle lens reduces
THREAD: 12 mm DIA. × 0.5 mm PITCH
THREAD: 10 mm DIA. x 0.5 mm PITCH
SENSOR
BACK FOCAL
LENGTH RANGES
FROM 3 mm TO 9 mm
MOST COMMON MOUNT: 12 mm DIA. × 0.5 mm PITCH
FIGURE 4-16
Mini-lens mounting dimensions
Lenses and Optics
89
1/4"
TELEPHOTO
(SHADED)
1/3"
1/2"
FL = 4.0 mm
WIDE ANGLE
FL = 25 mm
NORMAL FOR 1" SENSOR
FL = 75 mm
NARROW ANGLE
(TELEPHOTO)
FL = 8 MM
NORMAL FOR 1/3" SENSOR
NOTE: ANGULAR FOV SHOWN FOR 1/3-, 1/2-, AND 2/3-INCH FORMAT SENSORS
FIGURE 4-17
Wide-angle, normal, and narrow-angle (Telephoto) FFL lenses vs. format
the resolution or ability to discern objects in a scene.
Figure 4-17 shows a comparison of the FOV seen on 1/4-,
1/3-, and 1/2-inch format cameras with wide-angle, normal, and telephoto lenses.
4.3.2 Narrow-Angle Telephoto Viewing
When the lens FL increases above the standard M = 1
magnification condition the FOV decreases and the magnification increases. Such a lens is called a medium- or
narrow-angle (telephoto) lens. The lens magnification is
determined by Equations 4-5, 4-10, 4-8, and 4-9 for the
reference (1 inch) and three commonly used sensor sizes
(see also Table 4-4 and the Lens Finder Kit).
Outdoor security applications often require viewing
scenes hundreds and sometimes thousands of feet away
from the camera. To detect and/or identify objects,
persons, or activity at these ranges requires very longFL lenses. Long-FL lenses between 150 and 1200 mm
are usually used outdoors to view these parking lots or
other remote areas. These large lenses require very stable
mounts and rugged pan-tilt drives to obtain good picture
quality. The lenses must be large (3–8 inches in diameter)
to collect enough light from the distant scene and have
usable f-numbers (f/2.5 to f/8) for the video camera to
produce a good picture on the monitor.
Fixed focal length lenses having FLs from 2.6 mm up to
several hundred millimeters are refractive- or glass-type.
Above approximately 300 mm FL, refractive glass lenses
become too large and expensive, and reflective mirror
optics or mirror and glass optics are used to achieve optically fast (low f-number) lenses with lower weight and
size. These long FL telephoto lenses, called “Cassegrain”
or “catadioptric lenses,” cost hundreds to thousands of
dollars. Figure 4-18 shows a schematic of these lenses,
a 700 mm f/8.0 and a 300 mm f/5.6 lens used for longrange outdoor surveillance applications.
4.4 VARI-FOCAL LENS
The vari-focal lens is a variable focal length lens developed
to be used in place of an FFL lens (Figure 4-19).
In general it is smaller and costs much less than a zoom
lens. The advantage of the vari-focal lens over an FFL lens
is that its focal length and FOV can be changed manually
90
CCTV Surveillance
WINDOW
M1
SENSOR
LIGHT
FROM
DISTANT
SCENE
L
M2
FOCAL
PLANE
L = CORRECTING
LENSES
FIGURE 4-18
M1 = PRIMARY MIRROR
M2 = SECONDARY MIRROR
Long-range, long-focal length catadioptric lenses
8 mm FOV
3 mm FOV
FOCAL LENGTH
3–8 mm
SENSOR
FIGURE 4-19
VARI-FOCAL
LENS
Vari-focal lens configuration
by rotating the barrel on the lens. This feature makes it
convenient to adjust the lens FOV to a precise angle while
installed on the camera. The lenses were developed to
be used in place of FFL lenses to “fine tune” the FL for
a particular application. Having the ability to adjust the
FL “on the job” makes it easier for the installer and at
the same time permits the customer to select the exact
FOV necessary to observe the desired scene area. One
minor inconvenience of the vari-focal lens is that it must
be refocused each time the FL is changed. Typical varifocal lenses are available with focal lengths of: 3–8 mm,
5–12 mm, 8–50 mm, 10–120 mm (Table 4-6). With just
these few lenses focal lengths of from 1.8–120 mm and
144–16 FOVs can be covered continuously (i.e.—any
focal length in the range). The vari-focal lenses are a subset and simplified version of zoom lenses but they are not
a suitable replacement for the zoom lens in a variable FOV
pan/tilt application.
Lenses and Optics
FOCAL
LENGTH
(mm)
1.4–3.1
1.6–3.4
1.8–3.6
2.2–6
2.7–12
2.8–12
3.0–8
3.0–8
3.5–8
3.6–18
4.5–12.5
5.0–50
5.5–82.5
6–12
6–15
6–60
7–70
8–16
8–80
10–30
10–40
20–100
ZOOM
RATIO
FORMAT
(INCH)
2.2:1
2.1:1
2:1
2.7:1
4.4:1
4.3:1
2.7:1
2.7:1
2.3:1
5:1
2.8:1
10:1
15:1
2:1
2.5:1
10:1
10:1
2:1
10:1
3:1
4:1
5:1
1/3
1/3
1/3
1/4
1/3
1/3
1/3
1/3
1/3
1/2
1/2
1/3
1/3
1/2
1/2
1/3
1/2
1/2
1/2
1/2
1/3
1/3
OPTICAL
SPEED
f/#
1.4
1.4
1.6
1.2
1.2
1.4
1.4
1.0
1.4
1.8
1.2
1.3
1.8
1.4
1.4
1.6
1.8
1.6
1.6
1.4
1.4
1.6
HORIZONTAL ANGULAR FOV (DEG.)
1/4 INCH
1/3 INCH
1/2 INCH
WIDE
TELE
WIDE
TELE
WIDE
TELE
121
69.5
185
94.5
135
101
180
84.3
144
79.0
90.0
34.7
75
56
97.4
23.8
73
54.7
97.4
24.1
67.0
26.0
89.5
34.0
67.0
27
91
36
58.7
26.5
79.8
35.4
54.4
11.5
72.3
54.3
95.9
20.0
45.9
34.4
61.2
45.9
81.6
30.0
39.0
4.2
52
5.6
35.3
2.5
47.1
3.3
31.6
16.8
42.1
22.4
56.1
29.9
33.1
14.4
44.1
19.2
59.1
25.7
32.7
3.5
43.6
4.7
29.0
3.0
38.2
4.0
50.0
5.1
24.5
12.6
33.6
16.8
43.5
22.4
25
26
33.0
3.5
42.9
4.6
20
7.1
27
9.4
36
12.5
20.6
5.3
27.5
7.0
10.2
2.1
13.6
2.8
91
2/3 INCH
WIDE
TELE
59.8
30.8
NOTE: HORIZONTAL ANGULAR FOV FROM MANUFACTURERS’ SPECIFICATIONS
Table 4-6 Representative Vari-Focal Lenses—Focal Length, Vari-Focal Zoom Ratio vs. Sensor Format,
Horizontal FOV
4.5 ZOOM LENS
Zoom and vari-focal lenses are variable FL lenses. The
lens components in these assemblies are moved to change
their relative physical positions, thereby varying the FL and
angle of view through a specified range of magnifications.
Prior to the invention of zoom optics, quick conversion
to different FLs was achieved by mounting three or four
different FFL lenses on a turret with a common lens mount
in front of the CCTV camera sensor and rotating each lens
into position, one at a time, in front of the sensor. The
lenses usually had wide, medium, and short FLs to achieve
different angular coverage. This turret lens was obviously
not a variable-FL lens and had limited use.
4.5.1 Zooming
Zooming is a lens feature that permits seeing detailed
close-up views (high magnification) of a subject (scene
target) or a broad (low magnification), overall view of
an area. Zoom lenses allow a smooth, continuous change
in the angular FOV. The angle of view can be made
narrower or wider depending on the zoom setting. As
a result, a scene can be made to appear close-up (high
magnification) or far away (low magnification), giving the
impression of camera movement toward or away from the
scene, even though the camera remains in a fixed position. Figure 4-20 shows the continuously variable nature
of the zoom lens and how the FOV of the video camera
can be changed without replacing the lens.
To implement zooming, several elements in the lens
are physically moved to vary the FL and thereby vary the
angular FOV and magnification. Tables 4-1, 4-2, 4-3, and
4-4, and the Lens Finder Kit can be used to determine the
FOV for any zoom lens. By adjusting the zoom ring setting,
one can view narrow-, medium-, or wide-angle scenes. This
allows a person to view a scene with a wide-angle perspective and then close in on one portion of the scene that is of
specific interest. The zoom lens can be made significantly
more useful and providing the camera a still wider FOV
by mounting it on a pan/tilt platform controlled from a
remote console. The pan/tilt positioning and the zoom
lens variable FOV from wide to narrow angle and anywhere
in between provide a large dynamic FOV capability.
4.5.2 Lens Operation
The zoom lens is a cleverly designed assembly of lens
elements that can be moved to change the FL from
a wide angle to a narrow angle (telephoto) while the
image on the sensor remains in focus (Figure 4-21).
This is a significant difference from the vari-focal lens
92
CCTV Surveillance
CAMERA
WIDE-ANGLE LIMIT
NARROW-ANGLE
TELEPHOTO
LIMIT
ZOOM LENS
VERTICAL
FIELD OF VIEW
HORIZONTAL
FIELD OF VIEW
FIGURE 4-20
Zoom lens variable focal length function
FOCUS
RING
ZOOM
RING
IRIS
RING
IRIS
MOVABLE
ZOOM
GROUP
FRONT
FOCUSING
OBJECTIVE
GROUP
FIGURE 4-21
Zoom lens configuration
CAMERA
SENSOR
REAR
STATIONARY
RELAY
GROUP
Lenses and Optics
which must be re-focused each time its FL is changed
(Section 4.4).
A zoom FL lens combines at least three moveable groups
of elements:
1. The front focusing objective group that can be adjusted
over a limited distance with an external focus ring
to initially fine-focus the image onto the camera
sensor.
2. A movable zoom group located between the front and
the rear group that moves appreciably (front to back)
using a separate external zoom ring. The zoom group
also contains corrective elements to optimize the image
over the full zoom range. Other lenses are also moved
a small amount to automatically adjust and keep the
image on the sensor in sharp focus, thereby eliminating
subsequent external adjustment of the front focusing
group.
3. The rear stationary relay group at the camera end of the
zoom lens that determines the final image size when it
is focused on the camera sensor.
93
faster (more light throughput, lower f-number) than at
the telephoto setting. For example, a 11–110 mm zoom
lens may be listed as f/1.8 when set at 11 mm FL and
f/4 when set at 110 mm FL. The f-number for any other
FL in between the two settings lies in between these two
values.
4.5.4 Configurations
Each lens group normally consists of several elements.
When the zoom group is positioned correctly, it sees the
image produced by the objective group and creates a
new image from it. The rear relay group picks up the
image from the zoom group and relays it to the camera sensor. In a well-designed zoom lens a scene in focus
at the wide-angle (short-FL) setting remains in focus at
the narrow-angle (telephoto) setting and everywhere in
between.
Many manufacturers produce a large variety of manual
and motorized zoom lenses suitable for a wide variety of
applications. Figure 4-22 shows two very different zoom
lenses used for surveillance applications.
The manual zoom lens shown has a 85–51 mm FL (6:1
zoom ratio) and has an optical speed of f/1.6. The long
range lens shown has a large zoom ratio of 21:1. This lens
has an FL range of 30–750 mm and speed of f/4.6.
Figure 4-23 shows the FOVs obtained from a 11–110 mm
FL zoom lens on a 1/2-inch sensor camera at three zoom FL
settings.
Table 4-7 is a representative list of manual and motorized zoom lenses, from a small, lightweight, inexpensive
8–48 mm FL zoom lens to a large, expensive, 13.5–600 mm
zoom lens used in high-risk security areas by industry, military, and government agencies.
Zoom lenses are available with magnification ratios from
6:1 to 50:1. Many have special features, including remotely
controlled preset zoom and focus positions, auto-focus and
stabilization.
4.5.3 Optical Speed
4.5.5 Manual or Motorized
Since the FL of a zoom lens is variable and its entrance
aperture is fixed, its f-number is not fixed (see Equation 4-11). For this reason, zoom lens manufacturers often
list the f-number for the zoom lens at the wide and narrow FLs, with the f-number at the wide-angle setting being
The FL of a zoom lens is changed by moving an external zoom ring either manually or with an electric motor.
When the zoom lens iris, focus, or zoom setting must be
adjusted remotely, a motorized lens with a remote controller is used. The operator can control and change these
(A) MANUAL
FIGURE 4-22
Manual and motorized zoom lenses
(B) MOTORIZED
94
CCTV Surveillance
11 mm FL
SCENE
24 mm FL
HORIZONTAL ZOOM ANGLES
110 mm FL
WIDE: 46°
MEDIUM: 24°
NARROW: 5°
5°
24°
2 /3-INCH
SENSOR
FORMAT
46°
ZOOM LENS
11–110 mm
FIGURE 4-23
FOCAL
LENGTH
(mm)
4.5–54
4.6–28
5.5–77
5.5–187
5.7–34.2
5.8–121.8
6–72
6–90
7.5–105
8–48
8–96
8–160
9–180
10.5–105
10–140
10–200
10–300
12.5–75
12–120
12–240
16–160
10–500
NOTE: THE LENS HAS A CIRCULAR FOV
BUT THE SENSOR 4:3 ASPECT
RATIO FOV IS SHOWN
Zoom lens FOV at different focal length settings
ZOOM
RATIO
FORMAT
(INCH)
OPTICAL
SPEED:
f/#
12:1
6:1
14:1
34:1
6:1
21:1
12:1
15:1
14:1
6:1
12:1
20:1
20:1
10:1
14:1
20:1
30:1
6:1
10:1
20:1
10:1
50:1
1/4
1/4
1/3
1/3
1/3
1/3
1/3
1/3
1/2
1/2
1/2
1/2
1/3
2/3
2/3
1/2
1/2
2/3
1/2
1/2
1
1/2
1.1
1.0
1.4
1.8
1.0
1.6
1.5
1.2
1.4
1.0
2.0
2.0
1.2
1.4
1.9
2.5
1.5
1.6
1.8
1.6
1.8
4.0
HORIZONTAL ANGULAR FOV (DEG.)
1/4 INCH
1/3 INCH
1/2 INCH
WIDE
T ELE
WIDE
TELE
WIDE
TELE
43.5
3.7
41.6
7.5
36.2
2.6
47.1
3.5
35.0
1.1
46.6
1.5
34.5
6.1
46
8.1
33.8
1.7
45
2.3
33.4
2.8
43.6
3.8
33.0
2.3
43.8
3.1
27.0
2.0
35.5
2.6
46.2
3.5
24.9
4.4
33.0
5.9
43.2
7.7
25.3
2.2
33.4
2.8
43.5
3.7
25.1
1.3
33.4
1.7
43.6
2.3
22.6
1.2
30.3
1.5
18.6
2.0
24.8
2.6
33.0
3.5
19.9
1.5
26.4
2.0
35.0
2.7
20.3
1.1
27.0
1.4
35.5
1.8
20.0
0.7
26.6
0.9
35.5
1.25
16.1
2.8
21.4
3.7
28.4
4.9
16.6
1.7
22.1
2.3
29.4
3.1
17.2
0.9
23.0
1.2
30.8
1.6
16.8
1.8
22.4
2.4
30.8
3.2
13.7
0.3
18.2
0.4
35.5
0.7
2/3 INCH
WIDE
TELE
45.5
47.5
4.8
3.6
38.8
6.7
44.9
4.6
NOTE: NOMINAL HORIZONTAL ANGULAR FOV FROM MANUFACTURERS’ SPECIFICATION
Table 4-7 Representative Motorized Zoom Lenses—Focal Length, Zoom Ratio vs. Sensor Formats,
Horizontal FOV
Lenses and Optics
settings remotely using toggle switch controls on the console or automatically through preprogrammed software.
The motor and gear mechanisms effecting these changes
are mounted within the zoom lens. Manual zoom lenses
are not very practical for surveillance since an operator is
not located at the camera location and cannot manually
adjust the zoom lens.
95
but in some cases the operator can choose a manualor automatic-iris setting on the lens or the controller. In
surveillance applications, one shortcoming of a pan-tilt
mounted zoom lens is the existence of “dead zone” viewing
areas since the lens cannot point and see in all directions
at once.
4.5.7 Preset Zoom and Focus
4.5.6 Adding a Pan/Tilt Mechanism
A zoom lens and camera pointed in a fixed direction provides limited viewing. When the zoom lens is viewing wide
angle it sees a large FOV, but when it is zoomed to a narrow angle it will zoom in and magnify only in one pointing
direction—straight ahead of the camera. This is of limited
use unless it is pointing at a area of importance such as an
entrance door, entry/exit gate, receptionist, i.e. one single
location.
To fully utilize a zoom lens it is mounted on a pan-tilt
mechanism so that the lens can be pointed in almost any
direction (Figure 4-24).
By varying the lens zoom control and moving the
pan/tilt platform, a wide dynamic FOV is achieved. The
pan-tilt and lens controller remotely adjusts pan, tilt,
zoom, and focus. The lens usually has an automatic iris,
CAMERA
LENS
PAN/ TILT
PLATFORM
In a computer-controlled surveillance system a motorized
zoom lens with electronic preset functions is used. In this
mode of operation as a preset zoom lens, the zoom and
focus ring positions are monitored electrically and memorized by the computer during system setup. These settings
(presets) are then automatically repeated on command by
the computer software at a later time. In this surveillance
application, this feature allows the computer to point the
camera–lens combination according to a set of predetermined conditions and adjust pointing and the zoom lens
FL and focus: i.e. (1) azimuth and elevation angle, (2)
focused at a specific distance and (3) iris set to a specific f-number opening. When a camera needs to turn to
another preset set of conditions in response to an alarm
sensor or other input, the preset feature eliminates the
need for human response and significantly reduces the
time to acquire a new target.
PAN/ TILT
POINTING
RANGE
ZOOM LENS NARROW
FOV
ZOOM LENS CAN BE
POINTED ANYWHERE
WITHIN PAN/ TILT
POINTING RANGE
PAN/ TILT
POINTING
RANGE
FIGURE 4-24
Dynamic FOV of pan/tilt-mounted zoom lens
SCENE
ZOOM LENS WIDE
FOV
96
CCTV Surveillance
4.5.8 Electrical Connections
The motorized zoom lens contains electronics, motors,
and clutches to control the movement of the zoom, focus,
and iris adjustment rings, and end-of-travel limit switches
to protect the gear mechanism.
Since the electrical connections have not been standardized among manufacturers, the manufacturer’s lens wiring
diagram must be consulted for proper wiring. Figure 4-25
shows a typical wiring schematic for zoom, focus, and iris
mechanisms. The zoom, focus, and iris motors are controlled with positive and negative DC voltages from the
lens controller, using the polarity specified by the manufacturer.
4.5.9 Initial Lens Focusing
To achieve the performance characteristics designed into
a zoom lens, the lens must be properly focused onto the
camera sensor during the initial installation. Since the
lens operates over a wide range of focal lengths it must
be tested and checked to ensure that it is in focus at
the wide-angle and telephoto settings. To perform a critical focusing of the zoom lens the aperture (iris) must
be wide open (set to the lowest f-number) for all backfocus adjustments of the camera sensor. This provides
the conditions for a minimum depth of field, and the conditions to perform the most critical focusing. Therefore,
adjustments must be performed in subdued lighting, or
with optical filters in front of the lens, to reduce the light
and allow the lens iris to open fully to get minimum depth
of field. The following steps should be followed to focus
the lens:
1. With the camera operating, view an object at least 50
feet away.
2. Make sure the lens iris is wide open so that focusing is
most critical.
3. Set the lens focus control to the extreme far position.
4. Adjust the lens zoom control to the extreme wide-angle
position (shortest FL).
5. Adjust the camera sensor position adjustment control
to obtain the best focus on the monitor.
6. Move the lens zoom to the extreme telephoto (longest
FL) setting.
7. Adjust the lens focus control (on the controller) for the
best picture.
8. Re-check the focus at the wide-angle (position of shortest FL).
CONNECTOR
CONTROL CONSOLE
SUPPLY VOLTAGE *
VIDEO SIGNAL
IRIS
REMOTE CONTROL **
AUTO
MANUAL
FOCUS + 12 V NEAR ***
ZOOM
LENS
FOCUS
NEAR
CONTROL
FAR
IRIS
CAMERA
FOCUS
ZOOM
ZOOM
TELE
CONTROL
WIDE
ZOOM + 12 V TELE ***
COMMON
* DEPENDS ON MANUFACTURER: RANGES FROM 8 TO 12 VDC TYPICAL
** REQUIRES POSITIVE AND NEGATIVE VOLTAGE: ±6 VDC MAX TYPICAL
*** DEPENDING ON THE CONTROLLER THIS MAY BE ±6 OR ±12 VDC MAX
REQUIRES POSITIVE AND NEGATIVE VOLTAGE
FIGURE 4-25
Motorized zoom lens electrical configuration
Lenses and Optics
4.5.10 Zoom Pinhole Lens
Pinhole lenses with a small front lens element are common place in covert video surveillance applications. Zoom
pinhole lenses while not as common as FFL pinhole lenses
are available in straight and right-angle configuration. One
lens has an FL range of 4−12 mm and an optical speed
of f/4.0.
4.5.11 Zoom Lens–Camera Module
The requirement for a compact zoom lens and camera
combination has been satisfied with the zoom lens–camera
module. This module evolved out of a requirement for a
lightweight, low-inertia camera lens for use in high-speed
pan-tilt dome installations in casinos and retail stores. The
camera–lens module has a mechanical cube configuration
(Figure 4-26) so that it can easily be incorporated into
small pan-tilt dome housings and be pointed in any direction at high speeds.
The module assembly includes the following components and features: (1) rugged, compact mechanical
structure suitable for high-speed pan-tilt platforms, (2)
large optical zoom ratio, typically 20:1, and (3) sensitive
1/4- or 1/3-inch solid-state color camera with excellent
sensitivity and resolution. Options include: (1) automaticfocus capability, (2) image stabilization, and (3) electronic zoom.
4.5.12 Zoom Lens Checklist
The following should be considered when applying a zoom
lens:
97
• What FOV is required? See Tables 4-1, 4-2, 4-3, 4-4, and
the Lens Finder Kit.
• Can a zoom lens cover the FOV or must a pan-tilt platform be used?
• Is the scene lighting constant or widely varying? Is a
manual or automatic iris required?
• What is the camera format: 1/4-, 1/3-, 1/2-inch?
• What is the camera lens mount type: C, or CS?
• Is auto-focus or stabilization needed?
• Is electronic zoom required to extend the FL range?
Zoom lenses on pan-tilt platforms significantly increase
the viewing capability of a video system by providing a large
range of FLs all in one lens. The increased complexity
and precision required in the manufacture of zoom lenses
makes them cost three to ten times as much as an FFL lens.
4.6 PINHOLE LENS
A pinhole lens is a special security lens with a relatively
small front diameter so that it can be hidden in a wall,
ceiling, or some object. Covert pinhole lens–camera assemblies have been installed in emergency lights, exit signs,
ceiling-mounted lights, table lamps, and even disguised
as a building sprinkler head fixture. Any object that can
house the camera and pinhole lens and can disguise or
hide the front lens element is a candidate for a covert
installation. In practice the front lens is considerably larger
than a pinhole, usually 006−038 inch in diameter, but
nevertheless it can be successfully hidden from view. Variations of the pinhole lens include straight or right-angle,
manual or automatic iris, narrow-taper or stubby-front
shape (Figure 4-27). The lenses shown are for use with
C or CS mount cameras. Whether to use the straight or
right-angle pinhole lens depends on the application. A
detailed description and review of covert camera and pinhole lenses are presented in Chapter 18.
4.6.1 Generic Pinhole Types
FIGURE 4-26
Compact zoom lens–camera cube
A feature that distinguishes two generic pinhole lens
designs from each other is the shape and size of the front
taper (Figure 4-28). The slow tapering design permits easier installation than the fast taper and also has a faster
optical speed, since the larger front lens collects more
light.
The optical speed (f-number) of the pinhole lens is
important for the successful implementation of a covert
camera system. The lower the f-number of the lens, the
more the light reaching the camera and the better the
video picture. An f/2.2 lens transmits 2.5 times more light
than an f/3.5. The best theoretical f-number is equal to
98
CCTV Surveillance
(A) MANUAL IRIS FAST TAPER
(C) MANUAL IRIS SLOW TAPER
FIGURE 4-27
(B) AUTOMATIC IRIS FAST TAPER
(D) RIGHT ANGLE MANUAL IRIS
SLOW TAPER
(E) RIGHT ANGLE AUTOMATIC IRIS
SLOW TAPER
Straight and right-angle pinhole lenses
the FL divided by the entrance lens diameter (d). From
Equation 4-11:
f/# =
FL
d
For a pinhole lens, the light getting through the lens to
the camera sensor is limited primarily by the diameter of
the front lens or the mechanical opening through which
it views. For this reason, the larger the lens entrance diameter, the more light gets through to the image sensor,
resulting in a better picture quality, if all other conditions
remain the same.
4.6.2 Sprinkler Head Pinhole
There are many types of covert lenses available for the
security industry: pinhole, mini, fiber optic, camera-lenses
covertly concealed in objects. The sprinkler head camera is a unique pinhole lens hidden in a ceiling sprinkler fixture which makes it extremely difficult for an
observer standing at floor level to detect or identify. This
unique device provides an extremely useful covert surveillance system. Figure 4-29 shows the configuration and
two versions of the sprinkler head lens, the straight and
right-angle.
This pinhole lens and camera combination is concealed in and above a ceiling using a modified sprinkler
head to view the room below the ceiling. For investigative purposes, fixed pinhole lenses pointing in one specific direction are usually suitable. To look in different
directions there is a panning sprinkler head version. An
integral camera-lens-sprinkler head design is shown in
Section 18.3.5.
4.6.3 Mini-Pinhole
Another generic family of covert lenses is the mini-lens
group (Figure 4-30). They are available in conventional onaxis and special off-axis versions. Their front mechanical
shape can be flat or cone shaped.
These lenses are very small, some with a cone-shaped
front, typically less than 1/2 inch diameter by 1/2 inch
long and mount directly onto a small video camera. The
front lens in these mini-lenses ranges from 1/16 inch to
3/8 inch diameter. The cone-shaped mini-lens is easier to
install in many applications. These mini-lenses are optically fast having speeds of f/1.4 to f/1.8 and can be used in
places unsuitable for larger pinhole lenses. An f/1.4 minipinhole lens transmits five times more light than an f/3.5
pinhole lens. Mini-pinhole lenses are available in FLs from
2.1 to 11 mm and when combined with a good camera
Lenses and Optics
FAST-TAPER BARREL
SLOW-TAPER BARREL
LARGE
DIAMETER
SMALL
DIAMETER
CEILING TILE
30°
FIGURE 4-28
55°
Short vs. long tapered pinhole lenses
STRAIGHT
RIGHT-ANGLE
CAMERA
RIGHT-ANGLE
PINHOLE
SPRINKLER
LENS
STRAIGHT
PINHOLE
SPRINKLER
LENS
MANUAL
IRIS
SPRINKLER
HEAD
CEILING TILE
SPRINKLER
HEAD
LENS/CAMERA
FIELD OF VIEW
MOVEABLE
MIRROR
FIGURE 4-29
Sprinkler head pinhole assembly installation
CAMERA
CEILING TILE
99
100
CCTV Surveillance
FIGURE 4-30 Mini-pinhole
lenses
8 mm IN WATEC
OFFSET MOUNT
8 mm IN
C MOUNT
11 mm IN WATEC
OFFSET MOUNT
2.1 mm
8 mm
3.8 mm
11 mm
3.7 mm PINHOLE
result in the fastest covert cameras available. A useful variation of the standard mini-lens is the off-axis mini-lens. This
lens is mounted offset from the camera axis, which causes
the camera to look off to one side, up, or down, depending on the offset direction chosen. Chapter 18 describes
pinhole and mini-lenses in detail.
4.7 SPECIAL LENSES
There are several special video security lenses and lens functions that deserve consideration. These include: (1) a new
panoramic 360 lens, (2) fiber-optic and bore scope, (3)
split-image, (4) right-angle, (5) relay, (6) automatic-focus,
(7) stabilized, and (8) long-range. The new panoramic lens
must be integral with a camera and used with computer
hardware and software (Section 5.10). The other special
lenses are used in applications when standard FFL, varifocal or zoom lenses are not suitable. The auto-focus and
stabilizing functions are used to enhance the performance
of zoom lenses, vari-focal, fixed focus lenses, etc.
4.7.1 Panoramic Lens—360
There has always been a need to see “all around” i.e.
an entire room or other location, seeing 360 with one
panoramic camera and lens. To date, a 360 FOV camera system has only been achieved with multiple cameras
and lenses and combining the images on a split-screen
monitor. This lens is usually mounted in the ceiling of
a room or on a tower. Panoramic lenses have been available for many years but have only recently been combined
with powerful digital electronics, sophisticated mathematical transformations and compression algorithms to take
advantage of their capabilities. The availability of high resolution solid-state cameras has made it possible to map
a 360 by 90 hemispherical FOV into a standard rectangular monitor format with good resolution. Figure 4-31
shows two panoramic lens having a 360 horizontal FOV
and a 90 vertical FOV.
In operation the lens collects light from the 360
panoramic scene and focuses it onto the camera sensor
as a donut-shaped image (Figure 4-32). The electronics
and mathematical algorithm convert this donut-shaped
panoramic image into the rectangular (horizontal and vertical) format for normal monitor viewing. (Section 2.6.5
describes the panoramic camera in detail.)
4.7.2 Fiber-Optic and Bore Scope Optics
Coherent fiber-optic bundle lenses can sometimes solve
difficult video security applications. Not to be confused
with the single or multiple strands of fiber commonly used
to transmit the video signal over long distances, the coherent fiber-optic lens has many thousands of individual glass
fibers positioned adjacent to each other. These thousands
of fibers transmit a coherent image from an objective lens,
over a distance of several inches to several feet, where the
image is then transferred again by means of a relay lens to
the camera sensor. A high-resolution 450 TV lines coherent fiber bundle consists of several hundred thousand glass
fibers that transfer a focused image from one end of the
fiber bundle to the other. Coherent optics means that each
point in the image on the front end of the fiber bundle
corresponds to a point at the rear end. Since the picture
quality obtained with fiber-optic lenses is not as good as
that obtained with all glass lenses, such lenses should only
Lenses and Optics
360° LENS
360°
HORIZ. FOV
RAW DONUT IMAGE
FROM 360° LENS
90°
VERT. FOV
LENS SEES FULL HEMISPHERE:
360° × 180°
FIGURE 4-31
Panoramic 360 lens camera module
360°
IMAGE FORMED ON SENSOR
DONUT RING IMAGE REPRESENTING
HEMISPHERE FOV: 360 H × 180 V
SENSOR
PANORAMIC CAMERA /SENSOR
360°
HORIZ. FOV
CD
A
B
B
A
90°
VERT. FOV
ALGORITHM TO CONVERT
DONUT SHAPED PANORAMIC
IMAGE TO NORMAL LINEAR X,Y
90°
0°
A
B
D
C
DC
LENS SEES FULL HEMISPHERE:
360° × 180°
NORMAL 360° FORMAT IMAGE
0°
360°
LENS FEATURES:
• SEES ENTIRE AREA INCLUDING DIRECTLY BELOW
• INFINITE DEPTH OF FIELD: NO FOCUSING REQUIRED
• NO MOVING PARTS FOR P/ T/ Z–ALL ELECTRONIC
• FLUSH CEILING (OR WALL) MOUNTING
FIGURE 4-32
Panoramic lens layout and description
PC SOFTWARE /
HARDWARE
NORMAL
DISPLAY
101
102
CCTV Surveillance
(A) OBJECTIVE LENS: 8 mm OR 11 mm FL
FIBER TYPE: RIGID CONDUIT
RELAY LENS: M = 1:1
MOUNT: C OR CS
FIGURE 4-33
(B) OBJECTIVE LENS: ANY C OR CS MOUNT
FIBER TYPE: FLEXIBLE BUNDLE
RELAY LENS: M = 1:1
MOUNT: C OR CS
Rigid and flexible fiber optic lenses
be used when no other lens–camera system will solve the
problem. Fiber-optic lenses are expensive and available in
rigid or flexible configurations (Figure 4-33).
In the complete fiber-optic lens, the fiber bundle is to
be preceded by an objective lens, FFL or other, which
focuses the scene onto the front end of the bundle and
RELAY LENS
CAMERA
SENSOR
followed by a relay lens that focuses the image at the rear
end of the bundle onto the sensor (Figure 4-34).
Fiber-optic lenses are used in security applications for
viewing through thick walls (or ceilings), or any installation where the camera must be a few inches to several
feet away from the front lens, for example the camera
OBJECTIVE LENS
SCENE
COHERENT
FIBER OPTIC BUNDLE
G
6 to 12 INCHES LONG
G
IMAGE
ON
SENSOR
G
FIGURE 4-34
Fiber-optic lens configuration
G
FIBER
OPTIC
OUTPUT
SCENE
SCENE
IMAGE
Lenses and Optics
on an accessible side of a wall and the front of the lens
on the inaccessible scene side. In this situation the lens
is a foot away from the camera sensor. Chapter 18 shows
how coherent fiber-optic lenses are used in covert security
applications.
The bore scope lens is another class of viewing optics for
video cameras (Figure 4-35). This lens has a rigid tube of
6−30 inches long and a diameter of 004–05 inches. The
two generic designs, single rod lens and multiple small
lenses, transmit the image from the front objective lens
to the rear lens and onto a camera sensor. The single
rod lens uses a unique graded index (GRIN) glass rod, to
refocus the image along its length.
Bore scope lenses can only transmit a small amount
of light because of the small rod or lens diameters. This
results in high f-numbers, typically f/11 to f/30. The slow
speed limits the bore scope application to well-illuminated
environments and sensitive cameras. The image quality
of the bore scope lens is better than that of the fiberoptic lens, since it uses all glass lenses. Figure 4-35 shows
the diagram of a GRIN bore scope lens 0.125 inches in
diameter and 12 inches long, and an all-lens bore scope
with a diameter of 0.187 inches and a length of 18 inches.
The latter has a mirror at the tip to allow viewing at rightangles to the lens axis.
4.7.3 Bi-Focal, Tri-Focal Image Splitting Optics
A lens for imaging two independent scenes onto a single
video camera is called an image-splitting or bi-focal lens.
The split-image lens has two female C or CS lens ports for
two objective lenses. The lens views two different scenes
with two separate lenses, and combines the scenes onto
one camera sensor (Figure 4-36).
Each of the two objective lenses can have the same or
different FLs and will correspondingly produce the same
or different magnifications. The split-image lens accomplishes this with only one camera. Depending on the orientation of the bifocal lens system on the camera, the image
is split either vertically or horizontally. Any fixed-focus,
pinhole, vari-focal, zoom, or other lens that mechanically
fits onto the C or CS mount can be used. The adjustable
mirror mounted on the side lens allows the camera to
look in almost any direction. This external mirror can
point at the same scene as the front lens. In this case,
if the front lens is a wide-angle lens (4 mm FL) and the
side lens is a narrow-angle lens (50 mm FL), a bifocal lens
system results: one camera views a wide-field and narrowfield simultaneously (Figure 4-36). Note that the horizontal
scene FOV covered by each lens is one-half of the total lens
FOV. For example, with the 4 mm and 50 mm FL lenses
on a 1/3-inch camera and a vertical split (as shown), the
SINGLE GRADED INDEX (GRIN) LENS
CAMERA
RELAY
LENS
OBJECTIVE
SECTION
GRADIENT INDEX LENS
SENSOR
STRAIGHT
FRONT LOOKING
VIEWING HEAD
MULTIPLE DISCRETE LENSES
CAMERA
RELAY
LENS
RELAY LENSES
SENSOR
OBJECTIVE
SECTION
RIGHT ANGLE
VIEWING HEAD
WORKING LENGTH
FIGURE 4-35
Bore scope lens system
103
104
CCTV Surveillance
VERTICAL SPLIT
B
A
M
SCENE A
HORIZONTAL SPLIT
A
MOVEABLE
MIRROR
B
SCENE B
M
SCENE A
NARROW FOV
SCENE B
M = ADJUSTABLE MIRROR
WIDE FOV
FIGURE 4-36
Bi-focal split-image optics
4 mm lens displays a 30 × 45 feet scene, and the 50-mm
lens displays a 24 × 36 feet scene at a distance of 50 feet.
The horizontal FOV of each lens has been reduced by onehalf of what each lens would see if the lens were mounted
directly onto the camera (60 × 45 feet for the 4 mm lens
and 4.8 × 3.6 feet for the 50 mm lens). By rotating the splitimage lens 90 about the camera optical axis a horizontal
split is obtained. In this case the vertical FOV is halved. It
should be noted that the bifocal lens inverts the picture
on the monitor, a condition that is simply corrected by
inverting the camera.
A three-way or tri-split optical image-splitting lens views
three scenes (Figure 4-37). The tri-split lens provides
the ability to view three different scenes with the same
or different magnifications with one camera. Each scene
occupies one-third of the monitor screen. Adjustable optics
on the lens permit changing the pointing elevation angle
of the three front lenses so that they can look close-in for
short hallway applications and all the way out (near horizontal) for long hallways. Like the bi-split lens, this lens
inverts the monitor image, which is corrected by inverting
the camera. Both the bi-split and the tri-split lenses work
on 1/4-, 1/3-, or 1/2-inch camera formats.
The image splitting is accomplished without electronic
splitters and is useful when only one camera is installed
but two or three scenes need to be monitored.
4.7.4 Right-Angle Lens
The right-angle lens permits mounting a camera parallel
to a wall or ceiling while the lens views a scene at 90 to
the camera axis and wall or ceiling (Figure 4-38).
When space is limited behind a wall, a ceiling, in an
automatic teller machine (ATM) or an elevator cab, the
right-angle lens is a solution. The right-angle optical system permits use of wide-angle lenses (2.6 mm, 110 FOV)
looking at right angles to the camera axis. This cannot
be accomplished by using a mirror and a wide-angle lens
directly on the camera since the entire scene will not be
reflected by the mirror to the lens on the camera. The
edges of the scene will not appear on the monitor because
of picture vignetting (Figure 4-39).
The right-angle adapter permits the use of any FL lens
that will mechanically fit into its C or CS mount and works
with 1/4-, 1/3-, or 1/2-inch camera formats.
Lenses and Optics
MONITOR DISPLAY
1
2
3
SCENE 1
CAMERA
LENS 1
LENS 3
LENS 2
FRONT LENS
SECTION MOVES
UP AND DOWN TO
CHANGE VERTICAL
POINTING ANGLE
SCENE 3
SCENE 2
FIGURE 4-37
Tri-split lens views three scenes
CAMERA
C OR CS MOUNT
RIGHT ANGLE
RELAY LENS
ANY C OR CS
MOUNT LENS
NARROW OR WIDE ANGLE
MONITOR DISPLAY
FULL
SCENE
FIGURE 4-38
Right angle lens
105
106
CCTV Surveillance
CAMERA
FRONT
SURFACE
MIRROR
WIDE-ANGLE
LENS
SENSOR
THIS PART OF
SCENE IS NOT
REFLECTED OFF
MIRROR
THIS PART OF SCENE
IS BLOCKED BY LENS
FIGURE 4-39
Picture vignetting from wide-angle lens and mirror
4.7.5 Relay Lens
The relay lens is used to transfer a scene image focused
by any standard lens, fiber-optic or bore scope lens onto
the camera sensor (Figure 4-40).
CAMERA
The relay lens must always be used with some other
objective lens and does not produce an image in and
of itself. When used at the fiber bundle output end,
a fiber-optic lens re-images the scene onto the sensor.
When incorporated into split-image or right-angle optics,
RELAY
OBJECTIVE
LENS
LENS
SCENE
IRIS
FOCUSED
SCENE
IMAGE
FOCUSED
SCENE
IMAGE
AUTO IRIS RELAY LENS
WITH C MOUNT LENS
FIGURE 4-40
Relay lens adapter
Lenses and Optics
it re-images the “split” scene or right-angle scene onto
the sensor. The relay lens can be used with a standard
FFL, pinhole, zoom, or other lens as a lens extender with
unit magnification (M = 1), for the purpose of optically
“moving” the sensor out in front of the camera.
4.8 COMMENTS, CHECKLIST AND QUESTIONS
• A standard objective lens inverts the picture image, and
the video camera electronics re-invert the picture so that
it is displayed right-side-up on the monitor.
• The 25 mm FL lens is considered the standard or reference lens for the 1-inch (actually 16 mm diagonal)
format sensor. This lens–camera combination is defined
to have a magnification of M = 1 and is similar to the
normal FOV of the human eye. The standard lens for a
2/3-inch format sensor is 16 mm; for a 1/2-inch sensor
is 12.5 mm; for a 1/3-inch sensor is 8 mm; and for a 1/4inch sensor is 6 mm. All these combinations produce a
magnification M = 1. They all have the same angular
FOV and therefore view the same size scene.
• A short-FL lens has a wide FOV (Table 4-3; 4.8 mm, 1/2inch sensor sees a 13.4 feet wide × 10.0 feet high scene
at 10 feet).
• A long-FL lens has a narrow FOV (Table 4-3; 75 mm,
1/2-inch sensor sees a 4.3 feet wide × 3.2 feet high scene
at 50 feet).
• To determine what FOV is required for an application,
consult Tables 4-1, 4-2, 4-3, 4-4 and the Lens Finder Kit.
• If the exact FOV desired cannot be obtained with an
FFL, use a vari-focal lens.
• Does the application require a manual or motorized
zoom lens, a pan-tilt mount?
• Is the scene lighting constant or widely varying? Is a
manual or automatic iris required?
• What is the camera format (1/4-, 1/3-, 1/2-inch)?
• What type of camera lens mount: C, CS, mini 11 mm,
12 mm, 13 mm, bayonet, or other?
4.9 SUMMARY
The task of choosing the right lens for a security application is an important aspect in designing a video security
107
system. The large variety of focal lengths and lens types
make the proper lens choice a challenging one. The lens
tables and Lens Finder Kit provide convenient tools for
choosing the optimum FL lens to be used and the resulting angular FOV obtained with the chosen sensor size.
The common FFL lenses used in most video systems
have FLs in the range of 2.8–75 mm. Super wide-angle
applications may use a 2.1 mm FL. Super telephoto applications may use FLs from 100 to 300 mm. Most in the
range of 2.8–75 mm are available with a manual or automatic iris.
Vari-focal lenses are often chosen when the exact FL
desired cannot be obtained with the FFL. The vari-focal
lens can “fine tune” the focal length exactly to obtain the
angular FOV required. Vari-focal lenses are available in
manual and auto-iris configurations. Vari-focal lenses must
be re-focused when their FL is changed.
Zoom lenses are used when the camera and lens must
be scanned over a large scene area and the magnification of the scene must be changed. This is accomplished
by mounting the camera-lens on a pan-tilt platform capable of remotely panning and tilting the camera-lens assembly and zooming the zoom lens. Zoom lenses are available
with FLs from 8 to 200 mm with zoom ratios from 6 to 50
to 1. The zoom lens is available with a manual or automatic iris.
When the video security application requires that the
camera and lens be hidden, covert pinhole lenses and
mini-pinhole lenses are used. The pinhole lenses are
mounted to cameras having a C or CS mounts. The minipinhole lenses are mounted directly onto a small single board camera (with or without housing) and hidden
behind walls or ceilings or mounted into common objects:
PIR motion sensor, clock, emergency light, sprinkler head,
etc.). Chapter 18 describes covert video lenses and systems
in more detail.
Special lenses like the bi- and tri-split, fiber-optic or bore
scope lenses are only used when other simpler techniques
can not be used.
The newly implemented 360 panoramic lens is used
with a computer system and can view a 360 horizontal
FOV and up to a 90 vertical FOV. This lens has taken an
important place in digital video surveillance systems. The
computer transforms the complex donut-shaped image
into a useful rectangular image on the monitor.
This page intentionally left blank
Chapter 5
Cameras—Analog, Digital, and Internet
CONTENTS
5.1
5.2
5.3
5.4
5.5
Overview
Camera Function
5.2.1 The Scanning Process
5.2.2 The Video Signal
5.2.2.1 Monochrome Signal
5.2.2.2 Color Signal
Camera Types
5.3.1 Analog Camera
5.3.1.1 Monochrome
5.3.1.2 Color—Single Sensor
5.3.1.3 Color—Monochrome
Switchover
5.3.1.4 Color—Three Sensor
5.3.2 Digital Camera
5.3.2.1 Digital Signal Processing (DSP)
5.3.2.2 Smart Camera
5.3.2.3 Legal Considerations
5.3.3 Internet Camera
5.3.3.1 The IP Camera ID
5.3.3.2 Remote Viewing
5.3.3.3 Compression for Transmission
5.3.4 Low Light Level ICCD
5.3.5 Thermal IR
5.3.6 Universal System Bus (USB)
Basic Sensor Types
5.4.1 Solid State—Visible
5.4.1.1 Charge Coupled Device (CCD)
5.4.1.2 Complementary Metal Oxide
Semiconductor (CMOS)
5.4.2 ICCD, SIT, ISIT—Visible/Near IR
5.4.3 Thermal IR
5.4.4 Sensor Fusion—Visible/IR
Camera Features—Analog/Digital
5.5.1 Video Motion Detection (VMD)
5.5.2 Electronic Zooming
5.5.3 Electronic Shuttering
5.5.4 White Balance
5.5.5 Video Bright Light Compression
5.5.6 Geometric Accuracy
5.6 Camera Resolution/Sensitivity
5.6.1 Vertical Resolution
5.6.2 Horizontal Resolution
5.6.3 Static vs. Dynamic Resolution
5.6.4 Sensitivity
5.7 Sensor Formats
5.7.1 Solid-State
5.7.2 Image Intensifier
5.7.3 Thermal IR
5.8 Camera Lens Mounts
5.8.1 C and CS Mounts
5.8.2 Mini-Lens Mounts
5.8.3 Bayonet Mount
5.8.4 Lens–Mount Interferences
5.9 Zoom Lens–camera Module
5.10 Panoramic 360 Camera
5.11 High Definition Television (HDTV)
5.12 Summary
5.1 OVERVIEW
The function of a video camera is to convert the focused
visual (or IR) light image from the camera lens into a
time-varying electrical video signal for later presentation
on a monitor display or permanent recording on a video
recorder. The lens collects the reflected light from the
scene and focuses it onto the camera image sensor. The
sensor converts the light image into a time-varying electronic signal. The camera electronics process the information from the sensor and via the video signal sends it
to a viewing monitor by way of coaxial cable, fiber optics,
two-wire unshielded twisted-pair (UTP), wireless, or other
transmission means. Figure 5-1 shows a simple video camera/lens and monitor system.
109
110
CCTV Surveillance
TRANSMISSION MEANS
• COAXIAL
• UNSHIELDED TWISTED PAIR (UTP)
• FIBER OPTIC
• WIRELESS
CAMERA
MONITOR
LENS
SCENE
SCENE
RASTER
SCANNING
1 2 3 ...
1
2
1
2
CRT SCAN
FIGURE 5-1
ROW/COLUMN
PIXEL SCANNING
1
2
3
.
.
.
SOLID STATE
Video system with lens, camera, transmission means, and monitor
The monochrome or color, solid-state or thermal IR
cameras analyze the scene by scanning an array of horizontal and vertical pixels in the camera sensor. This process
generates an electrical signal representing the light and
color information in the scene as a function of time, so
that the scene can be reconstructed on the monitor or
recorded for later use.
Unlike film cameras, human eyes, and LLL image intensifiers that see a complete picture continuously, a video camera scans an image—point by point —until it has scanned
the entire scene, i.e. one frame. In this respect the camera scan is similar to the action of a typewriter: the type
element starts at the left corner of the page and moves
across to the right corner, completing a single line of type.
The typewriter carriage then returns to the left side of the
paper, moves down to the next line, and starts again.
In most video cameras, interlaced scanning is like a
typewriter that adds a second carriage return after each
line, repeating the lines until it reaches the bottom of
the page. This is how it completes one field, or half the
video image. The scanner/typewriter then moves back up
the page and begins typing on the second line at the left
or in the middle of the line just below the first line. It
continues this way, moving down and filling in the lines
between the original lines, until the entire page is complete. In this way the scanning completes the second field
and produces one full video frame. This electronic process
repeats (like putting in a new sheet of paper) for each
frame. Some cameras produce the video signal using progressive scanning in which every line is scanned one after
the other rather than skipping a line. Computer monitors
use progressive scanning.
In the mid-1980s the solid-state CCD video sensor
became a commercial reality. This new device replaced
the vidicon tube and silicon tube image sensors and represented a significant advance in camera technology. The
use of the solid-state “chip” sensor made the camera 100%
solid-state offering significant advantages over any and all
tube cameras: long life, no aging, no image burn-in, geometric accuracy, excellent sensitivity and resolution, low
power consumption, and small size.
Several different sensor types are available for video
security applications with the most prominent and widely
used being the CCD, CMOS, ICCD, and thermal IR. The
CCD and CMOS are used in daylight and some nighttime
applications and respond to visible and near-IR energy.
The ICCD is used in low-light-level nighttime applications.
The thermal IR camera is used in nighttime applications
when there is no visible or near-IR radiation and/or there
is a smoke, dust, or fog environment.
Solid-state cameras use a silicon array of photo-sensor
sites (pixels) to convert the input light image into an
Cameras—Analog, Digital, and Internet
electronic video signal that is then amplified and passed
on to a monitor for display. Most solid-state sensors
are charge-transfer devices (CTD) that are available in
three types depending on the manufacturing technology:
(1) the CCD, (2) the charge-priming device (CPD), and
(3) the charge-injection device (CID). A fourth sensor
type introduced more recently to the security market is the
CMOS. By far the most popular devices used in security
camera applications are the CCD and CMOS. The CID is
reserved primarily for military and industrial applications.
Solid-state and thermal cameras are significantly smaller
weigh less, and consume less power than the prior tube
cameras. A packaged solid-state image sensor is typically
3/4 inch × 3/4 inch × 1/4 inch or smaller while its prior,
tube predecessor was 3/4 inch in diameter and 5 inches
long or larger. Solid-state cameras consume from a fraction of a watt to several watts compared to 8–20 watts for
the tube camera.
The security field began using color cameras after the
technology for solid-state color cameras developed in the
consumer camcorder market. These color cameras have
a single solid-state sensor with an integral three-color filter and an automatic white-balancing circuit to provide a
reliable and sensitive device.
To produce a noise-free monochrome or color picture
with sufficient resolution to identify objects of interest, the
sensor must have sufficient sensitivity to respond to available natural daytime or artificial lighting. As mentioned,
security video cameras sensitive to visible and/or nearIR lighting can be represented by two general categories:
(1) CCD and CMOS solid-state, and (2) LLL ICCD. Separate from these visible and near-IR cameras is a third
category operating in the thermal (heat energy) IR region
which is responsive to the difference in temperature in the
scene rather than reflected light from the scene. The LLL
and thermal cameras are described in Chapter 19.
In subsequent sections each parameter contributing to
the function and operation of these security cameras is
described.
All security cameras have a lens mount in front of the
sensor to mechanically couple an objective lens or optical
system to the camera. The original C mount was designed
for the larger 1/2-, 2/3-, and 1-inch tube and solid-state
sensor formats and still accounts for many camera installations. Currently the most popular mount is the CS mount.
It is designed for the 1/4-, 1/3-, and 1/2-inch format sensor cameras and their correspondingly smaller objective
lenses. The CS mount configuration evolved from the original C mount as cameras sensors became smaller. Small
printed circuit board cameras used for covert surveillance
use a mini-mount with 10, 12, and 13 mm thread diameters
(see Section 5.8).
5.2 CAMERA FUNCTION
This section describes the functioning of the major parts
of a solid-state analog and digital video camera and the
video signal. Figure 5-2 is a generalized block diagram for
the analog and digital video camera electronics.
The camera sensor function is to convert a visual or IR
light image into a temporary sensor image which the camera scanning mechanism successively reads, point by point
or line by line, to produce a time-varying electrical signal representing the scene light intensity. In a color camera this function is accomplished threefold to convert the
three primary colors—red, green, and blue—representing
the scene, into an electrical signal.
The analog video camera consists of: (1) image sensor, (2) electronic scanning system with synchronization,
(3) timing electronics, (4) video amplifying and processing electronics, and (5) video signal synchronizing and
DIGITAL CIRCUITS
ANALOG CIRCUITS
PIXEL READOUT
DSP
SYNC TIMING
(SCANNING)
SENSOR
LENS
AMPLIFIER
COMBINING
ELECTRONICS
DRIVER
AMPLIFIER
SCENE
LOW VOLTAGE DC
POWER
CONVERTER
FIGURE 5-2
CCTV camera block diagram
111
INPUT POWER:
12 VDC
117 VAC
VIDEO
OUT
75 ohm
112
CCTV Surveillance
combining electronics. The synchronizing and combining
electronics produce a composite video output signal. To
provide meaningful images when the scene varies in realtime, scanning must be sufficiently fast—at least 30 fps—
to capture and replay moving target scenes. The video
camera must have suitable synchronizing signals so that a
monitor, recorder or printer at the receiving location can
be synchronized to produce a stable, flicker-free display
or recording.
The digital video camera (see dotted block) consists
of: (1) image sensor, (2) row and column pixel readout circuitry, (3) DSP circuits, and (4) video synchronizing and combining electronics. The synchronizing and
combining electronics produce a composite video output
signal.
The following description of the video process applies
to all solid-state, LLL and thermal cameras. The lens forms
a focused image on the sensor. The sensor image readout
is performed in a process called “linear” (or raster) scanning. The video picture is formed by interrogating and
extracting the light level on each pixel in the rows and
columns. The brightness and color at each pixel varies as
a function of the focused scene image so that the signal
obtained is a representation of the scene intensity and
color profile.
5.2.1 The Scanning Process
One video frame is composed of two fields. In the US the
NTSC system is based on the 60 Hz power line frequency
and 1/30 second per frame (30 fps), each frame containing 525 horizontal lines. In the European system, based on
a 50 Hz power line frequency and 1/25 second per frame,
each frame has 625 horizontal lines.
This solid-state analog video output signal has the same
format as that from its tube camera predecessor. Two
methods of scanning have been used: 2:1 interlace and
random interlace. Present cameras use the 2:1 interlace
scanning technique to reduce the amount of flicker in
the picture and improve motion display while maintaining the same video signal bandwidth. In both scanning
methods, every other line of pixels is scanned. In the
NTSC system, each field contains 262½ television lines.
This scanning mode is called two-field, odd-line scanning
(Figure 5-3).
START 0
FIELD 1
VIDEO
OUT
VIDEO
CAMERA
262 1/2
END
START 0
FIELD 2
NOTE:
APPROXIMATELY 21 HORIZONTAL LINES
FOR VERTICAL RETRACE IN 1 FIELD OR
42 LINES FOR 1 FRAME
VISUAL LINES IN 1 FRAME =
525 – 2 × 21 = 483 LINES
VIDEO
SIGNAL
FIELD 1
VERTICAL
BLANKING
TIME
525
END
FIELD 2
262 1/2
SYNC
FIGURE 5-3
NTSC two-field/odd-line scanning process
525
TIME
Cameras—Analog, Digital, and Internet
In the NTSC standard, 60 fields and 30 frames are completed per second. With 525 TV lines per frame and 30 fps,
there are 15,750 TV lines per second. In the standard
NTSC system, the vertical blanking interval uses 21 lines
per field, or a total of 42 lines per frame. Subtracting
these 42 lines from the 525-line frame leaves 483 active
picture lines per frame representing the scene. By convention, the scanning function of every camera and every
receiver monitor starts from the upper left corner of the
image and proceeds horizontally across to the right of the
sensor. Each time it reaches the right side of the image
it quickly returns to a point just below its starting point
on the left side. This occurs during what is called the
“horizontal blanking interval” of the video signal. This
process is continued and repeated until the sensor is completely read out and eventually reaches the bottom of the
image, thereby completing one field. At this point the
sensor readout stops (or in the case of the CRT monitor
the beam turns off again) and returns to the top of the
image: this time is called the “vertical blanking interval.”
For the second field (a full frame consists of two fields),
the scan lines fall in between those of the first field. By this
method the scan lines of the two fields are interlaced, which
reduces image flicker and allows the signal to occupy the
same transmission bandwidth it would occupy if it were
performing progressive scanning. When the second field
is completed and the scanning spot arrives at the lower
right corner, it quickly returns to the upper left corner to
repeat the entire process.
For the solid-state camera, the light-induced charge in
the individual pixels in the sensor must be clocked out of
the sensor into the camera electronics (Figure 5-4).
The time-varying video signal from the individual pixels
clocked out in the horizontal rows and vertical columns
likewise generates the two interlaced fields. In the case of
the tube camera, a moving electron beam in the tube does
the scanning similar to the CRT in the tube monitor.
By scanning the target twice (remember the typewriter
analogy), the sensor is scanned, starting at the top left side
of the picture, and a signal representing the scene image is
produced. First the odd lines are scanned, until one field
of 262½ lines is completed. Then the beam returns to the
top left of the sensor and scans the 262½ even-numbered
lines, until a total picture frame of 525 lines is completed. Two separate fields of alternate lines are combined
to make the complete picture frame every 1/30th of a
VERTICAL
TIMING
VERTICAL
VIDEO SIGNAL
SHIFTING
VIDEO
SIGNAL
HORIZONTAL
VIDEO SIGNAL
SHIFTING
0 ,1,2,3 . . .
754
0
1
2
3
.
.
.
1 HORIZONTAL
LINE
SENSOR
OUTPUT
SIGNAL
VERTICAL
SHIFT
REGISTER
SYNC
FIGURE 5-4
VIDEO OUTPUT
HORIZONTAL
SHIFT
REGISTER
HORIZONTAL
TIMING
SOLID
STATE
SENSOR
Solid-state camera scanning process
113
488
TIME
114
CCTV Surveillance
second. This TV camera signal is then transmitted to the
monitor, where it re-creates the picture in an inverse fashion. This base-band video signal has a voltage level from
0 to 1 volt (1 volt peak to peak) and is contained in a
4–10 MHz electrical bandwidth, depending on the system
resolution. The synchronizing signals are contained in the
0.5 volt timing pulses.
5.2.2 The Video Signal
The video signal can be better understood by looking at
the single horizontal line of the composite signal shown
in Figure 5-5.
The signal is divided into two basic parts: (1) the scene
illumination intensity information and (2) the synchronizing pulses. Synchronization pulses with 0.1-microsecond
rise and fall times contain frequency components of up
to 2.5 MHz. Other high frequencies are generated in the
video signal when the image scene detail contains rapidly
changing light-to-dark picture levels of small size and
are represented by about 4.2 MHz, and for good fidelity
must be reproduced by the electronic circuits. These
high-frequency video signal components represent rapid
changes in the scene—either moving targets or very small
objects. To produce a stable image on the monitor, the
synchronizing pulses must be very sharp and the electronic
bandwidth wide enough to accurately reproduce them.
The color signal in addition to a luminance (Y ) intensity component includes an additional chrominance (C)
color component in the form of a “color burst” signal.
The color signal can also be represented by three primary
color components: red, green, and blue (RGB), each having waveforms similar to the monochrome signal (without
the color burst signal).
5.2.2.1 Monochrome Signal
The monochrome camera signal contains intensity information representing the illumination on the sensor. For
the monochrome camera, all color information from the
scene is combined and represented in one video signal.
The monochrome signal contains four components:
1.
2.
3.
4.
horizontal line synchronization pulses
setup (black) level
luminance (gray-scale) level
field synchronizing pulses.
WHITE
LEVEL
V(t)
SCENE
ILLUMINATION
(A) ONE HORIZONTAL LINE
BLACK
LEVEL
TIME (t)
H
63 MICROSECONDS
0
(B) ONE FRAME—2 FIELDS
V(t)
EQUALIZING
PULSE
INTERVAL
H
H
3H
VERTICAL
SYNC
PULSE
INTERVAL
EQUALIZING
PULSE
INTERVAL
3H
(NOT TO SCALE)
FIGURE 5-5
Monochrome NTSC CCTV video signal
3H
HORIZONTAL
SYNC
PULSES
9TH TO 12TH
TIME (t)
Cameras—Analog, Digital, and Internet
5.2.2.2 Color Signal
Any video color signal is made up of three component parts:
1. Luminance—over all (Black and White)
2. Hue—tint or color
3. Saturation—color intensity of the hue.
A black and white video signal describes only the luminance.
The luminance, hue, and saturation of picture information
can be approximated by the unique combination of primary
color information. The primary colors in this additive color
process consist of red, green, and blue (RGB) color signals.
These primary colors can be combined to give the colors
frequently seen in a color bar test pattern. These colors are
described by turning on the primary color component parts,
either full “on” to full “off.” To produce the other colors
needed, the intensity of the individual primary colors must
be continuously variable from full on to full off (like the light
dimmer on a light switch).
The color camera signal contains light intensity and
color information. It must separate the spectral distribution of the scene illumination into the RGB color components (Figure 5-6). The color video signal is far more
complex than its monochrome counterpart, and the timing accuracy, linearity, and frequency response of the
electronic circuits are more critical in order to achieve
high-quality color pictures. The color video signal contains seven components necessary to extract the color and
intensity information from the picture scene and later
reproduce it on a color monitor:
1.
2.
3.
4.
5.
6.
7.
horizontal line synchronization pulses
color synchronization (color burst signal)
setup (black) level
luminance (gray-scale) level
color hue (tint)
color saturation (vividness)
field synchronizing pulses.
Figure 5-7 shows the video waveform with some of these
components.
Horizontal Line Synchronization Pulses. The first component the horizontal line synchronization pulses of the composite video signal has three parts: (1) the front porch, which
isolates the synchronization pulses from the active picture
information of the previous line, (2) the back porch, which
isolates the synchronization pulses from the active picture
information of the next scanned line, and (3) the horizontal
line sync pulse, which synchronizes the receiver, monitor,
or recorder to the camera.
Color Synchronization (Burst Signal). The second component, the color synchronization, is a short burst of color
information used as phase synchronization for the color
information in the color portion of each horizontal line.
The front porch, synchronization pulse, color burst, and
back porch make up the horizontal blanking interval. This
color burst signal, occurring during the back-porch interval of the video signal, serves as a color synchronization
signal for the chrominance signal.
RGB TO COMPOSITE ENCODER
SIGNAL INPUT
RGB CAMERA
SIGNAL OUTPUT
RED(R)
NTSC COMPOSITE VIDEO
GREEN(G)
Y–LUMINANCE
BLUE(B)
C–CHROMA
SYNC *
GROUND
GROUND
S–VIDEO(Y,C)
LENS
ANALOG MONITOR
* EXTERNAL SYNC OR SYNC ON GREEN
SCENE
SCENE
FIGURE 5-6
115
RGB to composite video encoding block diagram
116
CCTV Surveillance
EXPANDED SYNC
PORTION OF TV SIGNAL
IRE
UNITS
PICTURE
VIDEO
HORIZONTAL
PICTURE
BLANKING
100
WHITE
LEVEL
BACK
PORCH
20
7.5
0
BREEZE
FRONT WAY
PORCH
COLOR
SYNC
BURST
ooo
LUMINANCE
BLACK LEVEL
–20
SYNC
–40
WAVEFORM COMPONENTS:
• HORIZONTAL LINE SYNC PULSES (FRONT PORCH, BACK PORCH)
• COLOR SYNC (BURST)
• SETUP (BLACK) LEVEL
• PICTURE VIDEO
FIGURE 5-7
Luminance signal superimposed with sub-carrier color signal
Setup. The third component of the color television waveform is the setup or black level, representing the video
signal amplitude under zero light conditions.
Luminance. The fourth component is the luminance
black-and-white picture detail information. Changes and
shifts of light as well as the average light level are part of
this information.
Color Hue, and Saturation. The fifth and sixth components are the color hue, and color saturation information.
This information is combined with the black-and-white
picture detail portion of the waveform to produce the
color image.
Field Synchronization Pulse. This component maintains
the vertical synchronization and proper interlace.
These seven components form the composite waveform
for a color video signal.
The chrominance and the luminance make up the analog component parts of any video color signal. By keeping
these component parts separated, the interaction between
chrominance and luminance that could produce picture
distortion in the NTSC encoded signal is minimized. By
keeping the chrominance and luminance components
separated, picture quality can be improved dramatically.
The output of high-end video security systems is in a
form of three RGB signals. In most cases for video security, these RGB signals are combined (or encoded) into a
single video signal that is a composite of the primary color
information, or dual video signals: (1) luminance (Y )
and (2) chrominance (C ), representing the intensity and
color information, respectively. For the composite signal
the RGB signals go into an encoder and a single encoded
color signal comes out: the composite video signal. In the
USA, the color encoding standards were established by the
national television systems committee (NTSC). European
and other countries use a color encoding standard called
phase alternation line (PAL) or sequential with memory
(SECAM). Figure 5-6 shows the block diagram for the RGB
to composite video encoding.
In the NTSC system the luminance (Y ) or black-andwhite component of the video signal is used as a base upon
which the color signal is built. The color signal rides on
the base signal as a “sub-carrier” signal. Figure 5-7 shows
this sub-carrier signal superimposed on the base luminance signal which then completely describes the color
and monochrome video signals.
After much experimentation it was found that by combining the three RGB video signals in specific proportions an
accurate rendition of the original color signal was obtained.
These ratios were: 30% of the red video signal, 59% of the
Cameras—Analog, Digital, and Internet
green video signal, and 11% of the blue video signal. To this
signal was added the saturation and hue information. This
involved the generation of two additional combinations of
the RGB video signals. In the NTSC color system the hue
and saturation of a color system are described as a result
of combining the proper proportions of an I modulating
level and a Q modulating level, consisting of specific ratios of
RGB signals. To obtain accurate color rendition, the proper
ratio and phase relationships of the signals (expressed
in degrees) are required. This analysis is explored
in detail in Chapter 25 with the use of vector scopes.
117
5.3.1 Analog Camera
Until about the year 2000, all security cameras were CCD
and CMOS analog types. With the development of higher
density integrated circuits, digital signal processing (DSP)
was added and the use of digital cameras is now common
place. This section describes the monochrome and color
analog cameras.
5.3.1.1 Monochrome
5.3 CAMERA TYPES
Video security cameras are represented by several generic
forms including: (1) analog, (2) digital, (3) Internet,
(4) LLL, and (5) thermal IR. For daytime applications,
monochrome, color, analog, digital, and IP cameras are
used. When remote surveillance is required an IP camera
is used. For low light and nighttime applications the LLL
ICCD image intensified camera is used. For very low light
level or no light level applications, thermal IR cameras
are used.
CYAN
BLUE
MAGENTA
Most CCD and CMOS image sensors have wide spectral ranges covering the entire visible range of 400–700
nanometers (nm) and the near-IR spectral region of
800–900 nm. Figure 5-8 shows the spectral response of a
visible and near-IR CCD sensor with and without filters.
Some monochrome cameras are responsive to near-IR
energy from natural light or IR LED illuminators. These
cameras are operated without IR cutoff filters.
When the CCD or CMOS camera is pointed toward a
strong light source or bright object, the sensor is often
overloaded due to the high sensitivity of the imager in
GREEN
YELLOW
ORANGE
UV
INFRARED (IR)
SPECTRUM
RED
RELATIVE
SENSITIVITY
100
NEAR IR-CCD
WITHOUT IR FILTER
80
60
CMOS
WITHOUT IR
FILTER
40
PHOTOPIC
EYE RESPONSE
CCD
WITH IR
FILTER
20
0
0
FIGURE 5-8
400
500
600
700
800
900
1000
Spectral response of a visible and near-IR CCD, and CMOS sensor
1100
WAVELENGTH
(NANOMETERS)
118
CCTV Surveillance
the near-IR region. This overload produces a bright-light
band above and below the object on the monitor display. If the illuminating source contains a bright spot
of IR radiation, such as from sunlight or a car headlight, the IR cutoff filter should be used to prevent sensor
overload.
Monochrome cameras generally operate in most types
of scene lighting providing the light level is sufficient.
Light sources such as mercury vapor, metal arc, tungsten,
and low- and high-pressure sodium are widely used for
monochrome camera applications.
5.3.1.2 Color—Single Sensor
There are two generic color video camera types: singlesensor and three-sensor with prism. The single color sensor is the by far the most common type used in security
applications (Figure 5-9).
This camera has a complex color-imaging sensor that
contains an overlay of three integral optical filters to produce signals responding to the three primary colors: red
(R), green (G), and blue (B), which are sufficient to reproduce all the colors in the visible spectrum. The three color
filters divide the total number of pixels on the sensor
by three, so that each filter type covers one third of the
pixels. The sensor is followed by video electronics and
clocking signals to synchronize the composite video color
SCENE
LENS
signal. A higher quality alternative to the composite signal
is found in some color cameras having a 3-wire RGB, or a
2-wire Y and C output signal.
Since the single sensor camera has only one sensor, the light from the lens must be split into thirds,
thereby decreasing the overall camera sensitivity by three.
Since each resolution element on the display monitor
is composed of three colors, the resolution likewise is
reduced by this factor of 3. However, because of its relatively low cost the single-sensor camera is still much
more widely used than the more expensive three-sensor
prism type.
Color cameras are supplied with IR blocking filters since
the IR energy does not supply any color information
and would only overload the sensor and/or distort the
color rendition. The IR filter alters the spectral response
of the CCD imager to match the visible color spectrum
(Figure 5-10). The two curves represent the sensor with
and without the IR filter in place.
In order to obtain good color rendition when using
color cameras, the light source must have sufficient energy
between 400 nm (0.4 micron) and 790 nm (0.79 micron)
corresponding to the visible light spectrum. The IR
blocking filters restrict the optical bandwidth reaching the
color sensor to within this range so that color cameras
cannot be used with IR light sources having radiation to
the range of 800–1200 nm.
FULL
SPECTRUM
IMAGE
SINGLE COLOR SENSOR
COLOR VIDEO
SIGNAL OUTPUT
PROCESSING ELECTRONICS
CCD
COLOR PIXEL
TRIPLET
R G
B
STRIPE FILTERS
RED
GREEN
BLUE
FIGURE 5-9
Single-sensor color video camera block diagram
PICTURE
ELEMENT
(PIXEL)
Cameras—Analog, Digital, and Internet
CYAN
BLUE
MAGENTA
119
GREEN
YELLOW
ORANGE
UV
INFRARED (IR)
SPECTRUM
RED
RELATIVE
SENSITIVITY
100
SOLID STATE
WITHOUT IR FILTER
80
60
SOLID STATE
WITH IR FILTER
40
VIDICON
(REF)
20
0
0
FIGURE 5-10
400
500
600
700
800
900
1000
1100
WAVELENGTH
(NANOMETERS)
Spectral response of CCD imagers with and without IR filters
The color tube camera and early versions of the
color CCD camera had external white-balance sensors
and circuits to compensate for color changes. Present
solid-state color cameras incorporate automatic whitebalance compensation as an integral part of the camera
(see Section 5.5.4).
5.3.1.3 Color—Monochrome Switchover
Many applications (particularly outdoor) require cameras
that operate in daytime and nighttime. To accomplish
this, some cameras incorporate automatic conversion
from color to monochrome operation. This automatic
switchover significantly increases effectiveness of the camera in daytime and nighttime operation and reduces
the number of cameras required and the overall cost.
The conversion (switchover) is accomplished electronically and/or optically. Using the optical technique to
switch from the daytime mode to the nighttime mode, an
IR blocking filter is mechanically moved out of the optical path so that visible and near-IR radiation falls onto
the color sensor. Simultaneously the three-component
color signal is combined into one monochrome signal
resulting in a typical tenfold increase in camera sensitivity
(Figure 5-11).
5.3.1.4 Color—Three Sensor
The three-sensor color camera uses a beam-splitting prism
interposed between the lens and three solid-state sensors
to produce the color video signal (Figure 5-12).
The function of the prism is to split the full visible spectrum into the three primary colors, R, G, and B. Each
individual sensor has its own video electronics and clocking signals synchronized together to eventually produce
three separate signals proportional to the RGB color content in the original scene. The display from this camera
when compared with the single-sensor camera has three
times the number of pixels and shows a picture having
almost three times higher resolution and sensitivity, and
a picture with a rendition closer to the true colors in the
scene. This camera is well suited for the higher resolution
analog S-VHS, Hi-8 VCRs, and digital DVRs and digital
versatile disks (DVDs) now available for higher resolution
security applications (Chapter 9). S-VHS, Hi-8 and DVR
recorders can use the higher resolution Y (luminance)
and C (chrominance) signals, or RGB signals representing the color scene. The camera output signals (Y C or
RGB) can be combined to produce a standard composite
video output signal. This optical light combining prism
and three-sensor technique is significantly more costly
than a single-sensor camera, but results in a signal having
120
CCTV Surveillance
SCENE
ILLUMINATION
SENSOR
VIDEO OUT
AMPLIFIER
LENS
VIDEO SIGNAL
LEVEL SENSE
IR BLOCKING FILTER
+
–
OPERATION:
MOTOR
• SWITCHOVER: AUTOMATIC
FROM DAY TO NIGHT
CAMERA
IR BLOCKING FILTER
SENSOR
MOTOR
DAYTIME: FILTER IN
NIGHTTIME: FILTER OUT
FIGURE 5-11
Daytime color to nighttime monochrome camera switchover
significantly superior color fidelity and higher resolution
and sensitivity.
5.3.2 Digital Camera
Although most electronics have moved into a digital computer world, until recently video security was still operating in analog terms. For the camera an analog output
signals is typically recorded downstream as an analog signal. There is presently a strong migration toward a digital
video world using digital electronics in all the components
of the video system. Digital signal processing (DSP) has
been the driving force behind this migration. The initial
first step occurred with the introduction of DSP cameras
and has continued with the development of advanced PCdriven switching devices, digital ID cameras, and DVRs.
Today’s DSP cameras are less expensive than the analog
cameras they are replacing and have more features. Likewise DVRs replacing analog VCRs have increased resolution, improved reliability, and provide easy access to and
routing of the stored video records.
The advancements in digital technology have made
color video more practical, effective, and economical.
Presently color cameras now account for 70–80% of
all video camera sales. This is directly attributable to
higher performance and lower cost provided by digital
technology.
Most average resolution digital video cameras used in
security applications have about 512 by 576 active pixels.
High resolution cameras typically have 752 by 582 active
pixels. The latter is equivalent to SVHS-quality analog
video recording and has a bandwidth of approximately
6–7 MHz. Since VHS quality is sufficient for many applications, the standard full screen image format or fractional
screen common intermediate format (CIF)—with 352 by
240 (NTSC) pixels for the luminance signal Y and 176
by 144 pixels for the chrominance signals U and V—was
defined. The use of CIF resolution considerably reduces
the amount of data being recorded or transmitted while
providing adequate image quality.
Presently the CCD camera is the camera of choice
in digital systems. However, the CCD is being challenged by CMOS technology because of their lower prices,
smaller size, and lower power requirements. While many
customers want to make use of their existing analog
components in a digital system upgrade, replacement of
analog components to digital components makes most
sense. This is particularly true if the system will be used
to send the video signal over the Internet or other digital
networks since analog video signals sent over the Internet
require a high bandwidth than when digital components
Cameras—Analog, Digital, and Internet
121
BLUE
BLUE
IMAGE
VIDEO
ELECTRONICS
CLOCKING
SIGNALS
CCD
SCENE
BLUE
LENS
GREEN
IMAGE
VIDEO
ELECTRONICS
CCD
CLOCKING
SIGNALS
B
G
R
3 SENSOR OUTPUTS
MIXED AND
ASSEMBLED
INTO COLOR PICTURE
GREEN
REFLECTS
RED
RED
REFLECTS
BLUE
RED
IMAGE
CCD
VIDEO
ELECTRONICS
GREEN
CLOCKING
SIGNALS
RED
FIGURE 5-12
Three sensor color camera using prism
are used. Analog signals can be converted to digital signals before sending the signal across the network but this
requires special converters. It is more cost-effective to buy
a digital camera and put it directly on the network.
5.3.2.1 Digital Signal Processing (DSP)
The introduction of DSP cameras and advanced digital
technology has thrown the entire video security industry into a major tailspin—digital video security is here to
stay. The word “digital,” when referring to CCTV cameras,
only means that the camera incorporates digital enhancement or processing of the video signal and not that
the output signal is a digital signal. These cameras offer
improved image quality and features such as back-light
compensation, iris control, shuttering, electronic zoom,
and electronic sensitivity control to improve picture intelligence and overcome large lighting variations and other
problems.
The output signal from most surveillance cameras is still
an analog signal. This is because the required maximum
operating distance needed in most systems is longer than
most digital signals can be transmitted. A camera with
true digital output would have a very limited operating
distance (a few hundred feet) which would not be very
useful in most video security applications. The solution
for this is the use of network cameras and system networking equipment leading to the use of Internet cameras
transmitting over: (1) local area network (LAN), (2) wide
area network (WAN) or WLAN, (3) wireless networks
(WiFi), (4) intranets, and (5) the Internet, as a means for
long-distance monitoring. As mentioned earlier, most DSP
camera outputs are analog and use the communication
channels listed above.
Since the signal-to-noise ratio (SNR) in DSP cameras is
better than in analog cameras, manufacturers can increase
the amplification using automatic gain control (AGC)
resulting in a higher quality video image under poor lighting conditions. The typical SNR for a non-DSP camera is
between 46 and 48 dB. Cameras with DSP have an SNR
of between 50 and 54 dB. Note that every 3 dB change
in signal strength equals a 50% improvement in the signal level.
One new DSP signal processing technology employs circuitry that expands the dynamic range of an image sensor
up to 64 times over that of a conventional CCD camera
and brings camera performance closer to the capabilities
of the human eye. The camera simultaneously views bright
and dark light levels and digitally processes the bright
and dim images independently. In this new technique a
122
CCTV Surveillance
long exposure is used in the dark portions of the scene,
and a short exposure in the bright portions. The signals
are later combined using DSP into an enhanced image
incorporating the best portions of each exposure, and the
composite image is sent as a standard analog signal to the
monitor or recorder.
In the analog video world, if a video signal is weak or
noisy it can be amplified or filtered but the digital video
world is different. The digital video signal is immune to
many external signal disturbances but it can tolerate only
so many errors and then the signal is gone. A sudden signal drop-off is referred to as the cliff effect in which the
video signal is momentarily lost—a complete video picture
break-up or drop-out (see Figure 5-13).
5.3.2.2 Smart Camera
The introduction of smart digital cameras has changed the
architecture of video surveillance systems so that they can
now perform automated video security (AVS). Most analog
video systems allow the security officer to make decisions
based on the information seen on the video monitor. With
the availability of smart digital video cameras and DSP
electronics, decisions are made by the camera rather than
the security personnel.
The evolution from analog to digital cameras has provided the ability to incorporate intelligence into the camera and make the video camera a smart camera. In the
past if a guard saw a person walking the wrong way in a
restricted area, the guard would sound an alarm or alert
someone in the area to investigate the activity. Smart cameras now have VMD algorithms to distinguish different
types of objects and direction of movement. It is a small
task to have software sound an alarm or alert someone
automatically and free the guard for other tasks.
As another example, software algorithms have been
developed that can perform menial tasks. If a store manager wants to know how many people entered the front
door and went to a particular aisle or location, today’s
smart cameras have software that can analyze the video and
provide this information automatically. The camera’s DSP
takes all the incoming video and converts it to a format
that it can use to perform the analysis and make decisions.
The resulting output then interfaces to other devices to
carry out the decisions.
COMPARISON OF ANALOG AND DIGITAL PICTURE QUALITY
PICTURE
QUALITY
DIGITAL SIGNAL
†
PICTURE BREAK-UP
(CLIFF EFFECT)
* ANALOG-SIGNAL QUALITY DROPS GRADUALLY
ANALOG *
SIGNAL
†
DIGITAL-SIGNAL QUALITY DROPS ABRUPTLY.
PICTURE BREAKS UP OR DROPS OUT.
S/N (ANALOG)
ERROR RATE (DIGITAL)
DIGITAL PICTURE DROP-OUT
DIGITAL PICTURE BREAK-UP
REPEATS SAME IMAGE
BLOCKS INSTEAD
OF PICTURE
FIGURE 5-13
FRAME 1
Digital video signal picture break-up or drop-out
FRAME 2
FRAME N
Cameras—Analog, Digital, and Internet
To effectively implement AVS and video intelligence
across multiple cameras requires moving the image analysis into the camera. By running all or part of the video
analysis software in the camera, reliability is improved in
the overall system by eliminating a single point of failure.
There is also improved scalability in the system since additional cameras do not impact the central AVS system. By
performing the video analysis in the camera, the analysis
software takes advantage of the uncorrupted video, and
has the ability to instantly adjust to changes in the scene
to optimize it for the algorithm. Since transmission bandwidth is limited, the smart camera decides what if any
video should be sent from the camera and how much
compression should be applied to the video signal before
transmitting it. This technique reduces signal degradation
since the information has already been acted upon in the
camera.
Cameras can be made smarter through the use of DSP.
Not only can a camera make a record of an event, but
it can now evaluate the importance and relevance of that
event. By processing images at the camera level, the camera electronics can make decisions as to how to capture an
image. When an event occurs, such as movement in the
image, the camera electronics determine if the movement
is in a field of interest. Likewise it can recognize a person
as the main object vs. a dog or piece of paper blowing in
the wind. The camera can determine whether a person
needs to know about the event and alert security personnel automatically. This feature allows a single person to
manage a much greater number of cameras than would
otherwise be possible with an analog system and can significantly reduce employee expenses. If there is no activity
the camera can capture the scene at a lower resolution
or frame rate, thereby reducing the bandwidth required,
minimizing the impact a digital camera will have on a
network, and conserving storage capacity in the recorder.
(A) WEDGE HOUSING
FIGURE 5-14
5.3.2.3 Legal Considerations
There are legal factors to consider when using a digital
camera or digital video system for court and prosecution
purposes. In the digital process the camera image can
be manipulated pixel-by-pixel, with text or other modifications made in the image after the original image has
been recorded. There is a bias in the courts that points
out that compressed video can be manipulated. Therefore, it is suggested that the scene be captured using an
uncompressed, full JPEG image at a high frame rate with
a smart camera when an event is potentially important.
However, without a smart camera, in the time it takes to
alert a person and await a response to capture the event
in JPEG, the important moment could already have taken
place. A smart camera could make this determination itself
and thus be more responsive by increasing the resolution and frame rate automatically. The camera could also
intelligently zoom in on the target to get more detail and
information, something analog cameras cannot do without human intervention.
5.3.3 Internet Camera
The Internet camera using the IP and the Internet has
become a critical component in the use of AVS. Prior to
the Internet, video security was focused on the use of video
to bring the visual scene to the security officer. Using the
power of the Internet and digital IP cameras, the camera
scenes can now be transmitted directly to a security officer
located anywhere on the network (Figure 5-14).
To uniquely identify any specific camera on the network, an ID address and a password are assigned to the
camera. The camera when connected to the network can
be interrogated from any Internet port, anywhere in the
(B) COMPACT
Digital internet protocol (IP) cameras
123
(C) FULL FEATURED
124
CCTV Surveillance
world. During installation the camera is assigned an Internet address so that the user can view the camera scene
using the appropriate password and camera ID number
and commanding it to send the picture over the network
to the monitoring port. Likewise the security operator can
transmit command signals to the camera and platform to
perform pan, tilt, zoom, etc.
5.3.3.1 The IP Camera ID
The IP camera is assigned a digital address so that it can
be accessed from anywhere on the network—locally or
remotely. The network permits direct two-way communications for commanding the camera in pan, tilt, and zoom
while simultaneously receiving the image from the remote
Internet camera. The IP camera is given an Internet protocol (IP) address having the form shown in Table 5-1.
5.3.3.2 Remote Viewing
Remote viewing or AVS is the direction that the video
security industry is taking. This powerful new tool allows
viewing anywhere in the world using the Internet camera and the Internet as its transmission means. This AVS
function means that all security personnel can gain access
to camera control, etc. depending on password authorization, thereby significantly increasing the effectiveness of
the video security system.
The ability to view a scene remotely via the Internet,
intranet, or other long-distance communications path reliably and economically has resulted in the implementation
of AVS. The ability to receive a video picture from the
camera and command the camera to pan, tilt, zoom, etc.
all from the security control room at any remote distance
has significantly increased the functionality and value of
video security systems.
5.3.3.3 Compression for Transmission
Long before digital video transmission was envisioned,
engineers realized the need to compress the color video
signal. The color systems were designed to be compatible
with monochrome video signals already in use. The color
video signal had to fit in the same bandwidth space as the
monochrome signal. This color compatibility was not an
easy engineering task and created many trade-offs that can
only be solved with digital video transmission. In an ideal
system the color signal would be transmitted as three, highresolution primary channels red, green, and blue (RGB),
each with its own luminance and color information. Even
before the analog color video signal is converted to digital
data and compressed using data compression algorithms,
the video signal has been compressed using analog matrix
coding. It has not been possible to send a high-quality,
high-resolution computer digital video signal through a
standard real-time video transmission system. It is the same
as trying to pass high-quality stereo sound through a telephone: even with extensive coding it is not possible. In
the analog system video noise manifests itself as grain in
the color picture, or smearing, or contrast and brightness
problems that cause tint (hue) changes and picture rolling
or breakup. None of these analog problems should occur
in a well-designed digital video system. However, digital
video has a whole new set of problems such as aliasing,
compression artifacts, jagged edges, jumpy motion, and
just plain poor quality due to low data bit rate or compression. There is no digital video system benchmark at
present to accurately compare different video systems.
DISSECTING THE IP ADDRESS AND SUBNET MASK
DECIMAL NOTATION
BINARY NOTATION
IP ADDRESS
154.140.76.45
10011010
10001100
01001100
00101101
SUBNET MASK
255.255.255.0
11111111
11111111
11111111
00000000
THIS OCTET IS PART
OF AN EXTENDED
NETWORK PREFIX
THIS OCTET REPRESENTS
HOST INFORMATION
CAMERA ASSIGNED IP ADDRESS DURING INSTALLATION
Table 5-1
Internet Protocol (IP) Camera Address
Cameras—Analog, Digital, and Internet
To transmit the wide bandwidth video signal over a narrow
bandwidth communication channel requires that the video
signal be compressed at the camera location and decompressed at the monitoring location. The compression algorithms used for video removes redundant signal and picture
information both within each video frame (intra-frame)
and or redundant information from frame to frame (interframe). The techniques (algorithms) used to remove this
redundant information have been developed by several
technical groups and manufacturers. Several of the most
common algorithms are M-JPEG, MPEG-4, and H.264
developed by the Joint Motion Picture Engineers Group.
These compression formats use frame-by-frame compression. A wavelet compression algorithm called JPEG 2000
was created as the successor to the original JPEG format
developed in the late 1980s for still digital video (single
frame) and photography. This algorithm is based on stateof-the-art wavelet techniques, but is designed for static
imaging applications, on the Internet for e-commerce, digital photography, image databases, cell phones, and PDAs,
rather than for real-time video transmission.
There are basically two different types of video compression: (1) lossy, and (2) lossless (Chapter 7). Lossy compression as its name implies means that the final displayed
picture is not an exact replica of the original camera signal. The amount of compression determines how much
the final signal departs from the original. As a rule of
thumb, the more the compression the more the departure
from the original. The compression range for a lossy system can vary from 10 to 1, to 400 to 1 reduction in signal
bandwidth.
Digital video compression is simply a system for reducing the redundancy in the data words that describe every
pixel on the screen. Compression is used to reduce the
data size for a given video frame and de-compression is
used to convert the compressed signal back into a form like
the original video signal. How closely this compressed signal matches the original input video depends on the quality and the power of the compression algorithm. There
are several generic types of compression techniques available to the digital video engineer. Two basic types are:
inter-frame compression, which occurs in between frames,
and intra-frame which occurs within a frame. Inter-frame
compression is based on the fact that for most scenes there
is not a great change in data from one frame to the next.
It takes advantage of the condition that only a part of the
scene changes or has motion and therefore only those
portions which are different are compressed.
5.3.4 Low Light Level ICCD
The most sensitive LLL camera (Chapter 19) is the intensified CCD (ICCD). In special applications the silicon
intensified target (SIT), and intensified SIT (ISIT) are
used, but these prior generation tube cameras have all
125
but been replaced by the ICCD camera. These LLL cameras share many of the characteristics of the monochrome
CCD and CMOS described earlier but include a light intensification means to amplify the light thereby responding
to much lower light levels. The most sensitive solid-state
video camera is the ICCD and is used to view scenes illuminated under very low-light-level artificial lighting, moonlight, and starlight conditions. These LLL cameras have
an image intensifier coupled to an imaging tube or solidstate sensor and can view scenes hundreds to thousands
of feet from the camera under nighttime conditions.
5.3.5 Thermal IR
Thermal IR imaging systems are different from LLL nightvision systems based on ICCD image-intensifying sensors.
The ICCD responds to reflected sunlight, artificial lighting, moonlight, and starlight to form a visual image. It
also responds to the reflected light from near-IR emitting LEDs and filtered IR thermal lamp sources. In contrast, thermal imaging systems respond exclusively to the
heat from warm or hot emitting objects. The availability
of non-cooled (room temperature) thermal IR detector
technology is now driving the IR imaging security market.
The primary reasons are significant cost reduction, room
temperature operation, and improved camera operating
characteristics.
5.3.6 Universal System Bus (USB)
The Universal system bus (USB) is a transmission protocol
developed to permit disparate electronic equipment, cameras, etc. to communicate with a computer. The original
narrower bandwidth USB-1 protocol has been surpassed
by the new wideband USB-2 which interfaces the real-time
video signal with the computer USB port.
5.4 BASIC SENSOR TYPES
Background. Solid-state CCD sensors are a family
of image-sensing silicon semiconductor components
invented at Bell Telephone Laboratories in 1969. The
CCD imagers used in security applications are small,
rugged, and low in power consumption.
The solid-state CID camera was invented at the General
Electric Company in the 1970s. Unlike all other solid-state
sensors, this camera can address or scan any pixel in a
random sequence, rather than in the row and column
sequence used in the others. Although this feature has not
been used in the security field in the past, some new digital
cameras are taking advantage of this capability. When the
CID camera is scanned in the normal NTSC pattern, it has
attributes similar to those of other solid-state cameras.
126
CCTV Surveillance
Most video security installations use visible light
monochrome or color solid-state cameras. Prior to the use
of the solid-state cameras all video cameras used sensors
based on vacuum tube technology. The only instance in
which this technology is now used in video security practice
is in the LLL, SIT, and ISIT camera. Prior to the solid-state
sensor camera, video cameras utilized tube technology for
the sensor and solid-state transistors and integrated circuits for all signal processing. The tube cameras (mostly
monochrome) used a scanning electron beam to convert
the optical image into an electronic signal. The camera
tube consisted of a transparent window, the light-sensitive
target, and a scanning electron beam assembly. In operation, the electron beam scanned across the sensor target
area by means of electromagnetic coils positioned around
the exterior of the tube that deflected the beam horizontally and vertically. The video signal was extracted from
the tube by means of the electron beam with a new picture extracted every 1/30th of a second. Tube cameras
were available in sizes of 1/2-, 2/3-, and 1-inch formats.
Tube cameras were susceptible to image burn-in when
exposed to bright light sources and had a maximum lifetime expectancy of only a few years.
Functionally, the camera lens focuses the scene image
onto the target surface after passing through the sensor
window. The rear surface of the sensitive target area is
scanned by the electronic beam to produce an electrical
signal representative of the scene image. Solid-state electronics then amplified this electrical signal to a level of 1
volt and combined it with the synchronizing pulses. These
electronics produce the composite video signal consisting
of an amplitude-modulated signal representing the instantaneous intensity of the light signal on the sensor and the
horizontal and vertical synchronizing pulses.
Tube monochrome cameras provided excellent resolution because the target was a homogeneous continuous surface. With small electron beam spots sizes, high
resolutions of 500–600 TV lines for a 2/3-inch camera
and 1000 TV lines for a 1 inch diameter vidicon tube
were obtained. The workhorse of the industry was the
monochrome vidicon tube that was sensitive to visible
light. Later the monochrome silicon and Newvicon (Panasonic trademark) types were developed that were sensitive
to visible and near-IR energy. These tube cameras operated with light levels from bright sunlight (10,000 FtCd)
down to 1 FtCd. The vidicon was the least sensitive type
with the silicon or Newvicon tube being a better choice
for dawn to dusk applications having sensitivities between
10 and 100 times higher than the vidicon depending on
the spectral color and IR content of the illumination. The
silicon diode had a high sensitivity in the red region of the
visible spectrum and in the near-IR spectrum and could
“see in the dark” when the scene was illuminated with
an IR source. The silicon camera was the most sensitive
tube-type camera and had the highest resistance to bright
light damage.
5.4.1 Solid State—Visible
The CCD sensor was a new technology that replaced the
tube camera. The CCD solid-state sensor camera reduced
cost, power consumption, and product size, and was considerably more stable and reliable than the tube-type.
The CCD and newer CMOS sensor video cameras operate significantly differently than did their predecessor tube
cameras. No electron beam scans the sensor. Solid-state
sensors have hundreds of pixels in the horizontal and vertical directions equivalent to several hundred thousand
pixels over the entire sensor area. A pixel is the smallest
sensing element on the sensor and converts light energy
into an electrical charge, and then to an electrical current signal. Arranged in a checker-board pattern, sensor
pixels come with a specific number of rows and columns.
The total number determines the resolution of the
camera.
Solid-state image sensors are available in several types,
but all fall into two basic categories: charge transfer device
(CTD) and CMOS. The generic CTD class can further be
divided into CCD, CPD, and CID. Of these three types,
the CCD and CMOS are by far the most popular.
Charge coupled devices provide quality video performance manifesting low noise, wide dynamic range, good
sensitivity, fair anti-blooming and anti-smear reduction
capabilities, and operate at real-time (30 fps) video rates.
5.4.1.1 Charge Coupled Device (CCD)
At approximately the same time the CCD was invented in
1969 at the Bell Telephone Laboratories in New Jersey,
the Philips research laboratory in the Netherlands was also
working on an imaging transfer device. The Philips device
was called a “bucket brigade device” (BBD), which was
essentially a circuit constructed by wiring discrete MOS
transistors and capacitors together. The BBD was never
seriously considered for use as an imaging device, but the
concept of a “bucket brigade” provides a concise functional mechanism similar to the CCD in which charge is
passed from one storage site to the next through a series
of MOS capacitors.
By placing pixels in a line and stacking multiple lines,
an area array detector is created. As the camera lens
focuses the light from a single point in the scene onto
each pixel, the incident light on each pixel generates an
electron charge “packet” whose intensity is proportional
to the incident light. Each charge packet corresponds to a
pixel. Each row of pixels represents one line of horizontal
video information. If the pattern of incident radiation is
a focused light image from the optical lens system, then
the charge packets created in the pixel array are a faithful
reproduction of that image.
In the process, called “charge coupling,” the electrical
charges are collectively transferred from each CCD pixel
Cameras—Analog, Digital, and Internet
to an adjacent storage element by use of external synchronizing or clocking voltages. In the CCD sensor the
image scene is moved out of the silicon sensor via timed
clocking pulses that in effect push out the signal, line by
line, at a precisely determined clocked time. The amount
of charge in any individual pixel depends on the light
intensity in the scene, and represents a single point of the
intelligence in the picture. To produce the equivalent of
scanning, a periodic clock voltage is applied to the CCD
sensor causing the discrete charge packets in each pixel
to move out for processing and transmission. The image
sensor has both vertical and horizontal transfer clocking
signals as well as storage registers, to deliver an entire
field of video information once, during each integration
period, 1/30th of a second in the NTSC system. CCD
sensors require other timing circuits, clocks, bias voltages
made by standard manufacturing processes, and five or
more support chips.
All CCD image sensors consume relatively low power
and operate at low voltages. They are not damaged by
intense light but suffer some saturation and blooming
under intense illumination. Most recent devices contain
anti-blooming geometry and exposure control (electronic
shuttering) to reduce optical overload. Typical device
LIGHT
PIXELS
INPUT
parameters for a 1/3-inch format CCD available today are:
771 × 492 pixels (horizontal by vertical) for monochrome
and 768 × 494 for color cameras. They have horizontal
resolutions of 570 TV lines for monochrome and 480 TV
lines for color. Sensitivities are 0.05 lux (F/1.2 lens) for
monochrome and 0.5 lux (F/1.0 lens) for color. The CCD
sensors are available in formats of 1/4-, 1/3-, and 1/2inch, and in some special cameras in a 1/5-, 1/6-, or
2/3-inch format. All have the standard 4 × 3 aspect ratio.
Typical dynamic ranges for monochrome and color are
100 to 1 without shuttering, and 3000 to 1 with electronic
shuttering times range from 1/16–1/10,000 second.
Interline Transfer. There are several different CCD sensor pixel architectures used by different manufacturers.
The two most common types are the inter-line transfer
(ILT) and frame transfer (FT). Figure 5-15 shows the
pixel organization and readout technique for the ILT CCD
image sensor.
The pixel organization has precisely aligned photosensors with vertical inter-linearly arrayed shift registers,
and a horizontal shift register linked with the vertical shift
registers as shown. The photo-sensor sites respond to light
variations that generate electronic charges proportional to
VERTICAL
STORAGE/
READOUT
PHOTO SENSOR
SITES (PIXELS)
HORIZONTAL
READOUT
OUTPUT
AMPLIFIER
VERTICAL
SHIFT REGISTER
LIGHT IN-SIGNAL PATH
EVEN LINE
ODD LINE
EVEN LINE
ODD LINE
EVEN LINE
ODD LINE
EVEN LINE
SIGNAL
OUTPUT
AMPLIFIER
HORIZONTAL
SHIFT REGISTER
FIGURE 5-15
Interline transfer CCD sensor layout
127
OUT
128
CCTV Surveillance
the light intensity. The charges are passed into the vertical
shift registers simultaneously and then transferred to the
horizontal shift registers successively until they reach the
sensor output amplifier. The camera electronics further
amplify and process the signal. Each pixel and line of information in the ILT device is transferred out of the sensor
array line-by-line, eventually clocking out all 525 lines and
thereby scanning the entire sensor to produce a frame of
video. This sequence is repeated to produce a continuous
video signal.
Frame Transfer. In the FT CCD, the entire 525 lines
are transferred out of the light sensitive array and simultaneously, and stored temporarily in an adjacent nonilluminated silicon buffer array (Figure 5-16).
The basic FT CCD structure is composed of two major
elements: a photo-plane and a companion memory section.
First the photo-plane is exposed to light. After exposure
the charge produced is quickly transferred to the companion memory and then read out of memory—one line at a
time for the entire frame time. While this memory is being
read out, the photo-plane is being exposed for the next
image. Although full-pixel storage memory is required for
this structure, it has the big advantage of having all the
pixels exposed at the same time. CMOS technology on the
other hand exposes a line until it is time to read out that
line, then that line is transferred to the output register.
Consequently the beginning and end of each exposure
time of each line is different for every line, i.e. all pixels
are not exposed at the same time. The difference between
CCD and CMOS is seen when there is motion in the scene.
The CCD works better whenever the scene consists of significant motion relative to a line time.
The FT CCD imager has photo-sites (pixels) arranged in
an X-Y matrix of rows and columns. Each site has a lightsensitive photodiode and an adjacent charge site which
receives no light. The pixel photodiode converts the light
photons into charge (electrons). The number of electrons produced is proportional to the number of photoelectrons (light intensity). The light is collected over the
entire sensor simultaneously and then transferred to the
adjacent site, and then each row is read out to a horizontal transfer register. The charge packets for each row are
read out serially and then sensed by a charge-to-voltage
converter and amplifier section.
LIGHT IN-SIGNAL PATH
IMAGING AREA
LIGHT SENSITIVE
PIXELS
STORAGE AREA
(MEMORY)
LIGHT INSENSITIVE
HORIZONTAL OUTPUT REGISTER
FIGURE 5-16
Frame transfer CCD sensor pixel organization
OUTPUT
AMPLIFIER
SIGNAL
OUTPUT
Cameras—Analog, Digital, and Internet
5.4.1.2 Complementary Metal Oxide Semiconductor
(CMOS)
For more than two decades solid-state CCD has been the
technology of choice for security cameras. However, they
are now being challenged by the CMOS sensor. CMOS
research sponsored by NASA and has led to many commercial applications of the CMOS imagers.
In the past CMOS image sensors were relegated to low
resolution applications but now they have sufficient pixels
for serious security applications. Charge coupled device
sensors will still have a place in the high resolution, high
sensitivity applications but the CMOS has found increasing
application for main-stream video security.
The holy grail in most CMOS imager ventures has
been the “camera-on-a-chip” in which a single CMOS chip
includes the imaging sensor, timing and control, as well as
post-processing circuitry. The CMOS-type sensor exhibits
high picture quality but has a lower sensitivity than the
CCD. In the CMOS device, the electric signals are read
out directly through an array of transistor switches rather
than line by line as in the CCD sensor.
The CMOS sensor has come into vogue because of the
advantage of incorporating on-board analog to digital converters, timing circuits, clocks, and synchronization circuits
on the chip. The sensor is manufactured using standard
silicon processes, the same as those used in computer chip
fabrication, resulting in lower fabrication costs. A CMOS
sensor uses about 10–20% as much power as a comparable CCD.
Digital signals from CMOS sensors are always transmitted (not stored as in the CCD sensor) and therefore do not
need a DSP. Significant improvements have been made in
CMOS cameras for low light level indoor applications. The
typical CMOS camera requires a light level of 05–1 FtCd.
In general, CCD cameras operate in lower light conditions
than CMOS cameras.
Using the standard semiconductor production lines it is
possible to add a microprocessor or DSP, random access
memory (RAM), read only memory (ROM), and a USB
controller to the same IC.
Complementary metal oxide semiconductor sensors are
lower-priced than CCD and will likely remain so because
they are manufactured using the most common silicon
processing techniques and are also easier to integrate with
other electronic circuitry. CMOS sensors are inherently
better than their CCD counterpart in light overload situations and exhibit far less blooming than the CCD. When
the CCD is pointed at a bright lamp (100 watt incandescent or other) light source, a white blob is seen around
the bulb which obscures the fixture and ceiling scene adjacent to it. With the CMOS the fixture and ceiling detail
is seen.
Active Pixel Sensor (APS). The CMOS APS digital
camera-on-a-chip technology has progressed rapidly since
its invention by the scientists at the NASA Jet Propulsion
Laboratory (California).
In the 1990s Stanford University developed a new technology to improve CMOS sensors called the “active pixel
sensor” (APS). This digital pixel system (DPS) technology
produced higher quality, sharper images, and included
an amplifier and analog-to-digital converter (ADC) within
each image sensor pixel. The ADCs convert the light signal values into digital values at the point of light capture.
Figure 5-17a shows how the DPS works, illustrating that
B) ACTIVE COLUMN SENSOR (ACS)
A) ACTIVE PIXEL SENSOR (APS)
LARGE
OPEN LOOP
AMPLIFIER ON
EACH PIXEL
RESET BIAS
TO NEXT
COLUMN
ROW RESET
ROW SELECT
SMALL
AMPLIFIER
CLOSED LOOP
UNITY GAIN AMP.
TO NEXT
ROW RESET
COLUMN
ROW SELECT
ACTIVE SENSOR AREA
~70% OF PIXEL AREA
TECHNOLOGY: AMPLIFIER INSIDE EACH PIXEL
WEAKNESSES: INTERNAL AMPLIFIER LOWERS
“FILL FACTOR” TO ~30%
REDUCED DYNAMIC RANGE:
VARIATION IN AMPLIFIER GAIN
FROM PIXEL TO PIXEL
TECHNOLOGY: SHARED UNITY GAIN AMPLIFIER
FOR EACH COLUMN
ATTRIBUTES: HIGH “FILL FACTOR” 70%
ONLY 1 TRANSISTOR IN EACH PIXEL
HIGH DYNAMIC RANGE: 80 dB
UNITY GAIN AMPLIFIER SHARED
BY EACH PIXEL IN EACH COLUMN
FIXED PATTERN
NOISE
NOISIER THAN ACS
FIGURE 5-17
OUTPUT
RESET BIAS
OUTPUT
ACTIVE SENSOR AREA
~30% OF PIXEL AREA
129
CMOS active pixel sensor (APS) and active column sensor (ACS)
130
CCTV Surveillance
the charge is removed just before saturation of the pixel
occurs, thereby insuring that each pixel is neither under
nor over exposed.
Because each pixel has its own ADC, each pixel in effect
acts as its own camera. These sensors have in effect thousands of “cameras” which are combined to create highquality video frames and pictures. One disadvantages of
the APS technology is that it reduces the “fill factor” (sensitivity, dynamic range) and produces fixed pattern noise.
A salient advantage of the technology is that high-lighted
areas do not saturate and cause blooming or smearing as
when illuminated by a street light or automobile light for
applications in nighttime highway surveillance or vehicle
license plate identification. The CMOS APS devices are
immune to smear and have 30–40% fill factors.
To increase sensor sensitivity, modern on-chip microlenses are formed by an inexpensive process. These lenses
act as “funnels” to direct light incident across an entire pixel
toward the sensitive portions of the pixel (not an imaging
lens). Microlenses increase the responsivity of some low-fillfactor sensors by a factor of two to three. The fill factor is
the ratio of optically illuminated area of the sensitive silicon
area to the total silicon area in a particular pixel.
Active Column Sensor (ACS). To overcome some of the
disadvantages of the APS CMOS sensor (sensitivity, noise),
suppliers have developed active column sensor (ACS)
CMOS sensors (Figure 5-17b).
The CMOS sensors have had limitations for the video
security industry but the ACS process has the potential of
overcoming these limitations.
The ACS CMOS imager technology eliminates nonuniformity of gain by using a unity gain amplifier at each
pixel site. Active column sensor also increases the 30%
fill factor for APS technology to 70% for ACS. These sensors can also operate at much faster clock speeds and
therefore produce no smear for fast motion in the image
or fast pan/tilt applications. They offer outstanding antiblooming capability in both rows and columns which
makes them well suited for high- and low-lighted scenes.
They rank high in video quality as do CCD imagers.
The ACS technology, CMOS imager could do to the
CCD sensor what the CCD did to the vidicon.
The Internet requires the best image quality at very low
cost for video graphics array (VGA) and common intermediate format (CIF) display resolution. The ACS CMOS
sensor sensitivity has so improved that CMOS sensors are
now comparable to the CCD.
Prior to the use of ACS imager technology, most CMOS
imagers used the APS technology, the technique of placing
an amplifier inside each pixel. This reduced the fill factor
and therefore the sensitivity and the dynamic range of the
sensor. The ACS process uses a unity gain amplifier which
reduces the non-uniformity of the individual pixels and
results in a higher fill factor and higher dynamic range.
In the coming years CMOS sensors should exhibit no
limitation whatsoever regarding frame speed, resolution,
sensitivity, and noise in comparison with CCD sensors.
Most available CCD sensors have a signal-to-noise ratio
(S/N) of no greater than 58 dB. Some advanced CMOS
sensor arrays already have a 66 dB sensitivity and from
1024 × 1034 to 4096 × 4096 pixels.
Table 5-2 compares the sensitivity of different types of
CCD and CMOS solid-state sensors (see also section 5.6).
5.4.2 ICCD, SIT, ISIT—Visible/Near IR
For dawn and dusk outdoor illumination only the best
CCD cameras can produce a usable video picture. ICCD
cameras can operate under the light of a quarter-moon
with 0.001 FtCd. The ISIT camera can produce an image
with only 0.0001 FtCd, which is the light available from
stars on a moonless night. These LLL cameras offer a
100–1000 times improvement in sensitivity over the best
monochrome CCD or CMOS cameras. They intensify light,
whereas the CCD and CMOS detect light. The ICCD uses a
light intensifier tube or micro-channel plate (MCP) intensifier to amplify the available light up to 50,000 times.
The resulting sensitivity approaches that of the SIT camera, is much smaller, requires much less power, and eliminates the blurring characteristics of the SIT under very
low light level conditions.
The ICCD camera system has sufficient sensitivity and
automatic light compensation to be used in surveillance
applications from full sunlight to quarter moonlight conditions. The cameras are provided with automatic lightlevel compensation mechanisms having a 100 million to 1
light-level range and built-in protection to prevent sensor
degradation or overload when viewing bright scenes.
For viewing the lowest light levels, the ISIT camera
provides the widest dynamic range from full sunlight to
starlight conditions, having a 4 billion to 1 automatic lightlevel range control. Though large, these cameras have
been used in critical LLL security applications. The ISIT
camera uses an SIT tube with an additional light amplification stage and is still the lowest (and most expensive)
LLL camera available. A description of these LLL cameras
is given in Chapter 19.
5.4.3 Thermal IR
The infrared spectrum is generally defined as followings: the near IR or short-wave IR covers from 700 to
3000 nm (075–3 microns (m)), the mid-wave IR from 3
to 5 microns, and the long-wave IR from 8 to 14 microns.
Short-wave IR camera systems use the natural reflection and
emission from targets and are used in applications making
use of available LLL radiation from reflected moonlight,
sky glow (near cities or other nighttime lighted facilities),
or artificially generated radiation from IR LEDs or filtered IR lamp sources. Mid-wave IR systems use the energy
Cameras—Analog, Digital, and Internet
FORMAT
TYPE
DESCRIPTION
HORIZONTAL
RESOLUTION
(TV LINES)
SENSITIVITY*
(Lux)
COLOR
COMMENTS
B/W
1/6 CCD
COLOR (NTSC), B/W
480
5.0
1/6 CCD
COLOR (NTSC), B/W
470
2.5
0.1
ULTRA-FAST IP SPEED DOME
1/6 CCD
COLOR (PAL), B/W
460
2.5
0.1
ULTRA-FAST IP SPEED DOME
1/4 CCD
1/4 CCD
COLOR (NTSC), B/W
COLOR (PAL), B/W
COLOR (NTSC), B/W
470
470
510
0.5
0.5
1.0
0.01
0.01
0.06
SPEED DOME, SURVEILLANCE
SPEED DOME, SURVEILLANCE
COLOR (NTSC)
380
3.0
1/4 CCD
1/4 CMOS
REMOTE HEAD 7 mm DIAMETER
SURVEILLANCE
GENERAL SURVEILLANCE
1/3 CCD
COLOR (NTSC), B/W0
480/570 (B/W)
0.8
0.1
SURVEILLANCE-DAY/NIGHT
1/3 CCD
MONOCHROME (NTSC)
380
—
0.5
SURVEILLANCE
1/3 CCD
COLOR (NTSC), B/W
480
1.0
0.05
SURVEILLANCE
1/3 CMOS
COLOR (NTSC)
380
2.0
1/3 CMOS
MONOCHROME (NTSC)
400
—
0.05
SURVEILLANCE
1/3 CMOS
COLOR (NTSC)
380
1.0**
0.05**
COVERT SURVEILLANCE
1/2 CCD
1/2 CCD
1/2 CCD
COLOR (NTSC), B/W
COLOR (PAL), B/W
MONOCHROME (NTSC)
480
0.15
0.015
SURVEILLANCE
480
570
0.15
0.015
0.07
SURVEILLANCE
HIGH RESOLUTION B/ W
—
131
COVERT SURVEILLANCE
* SENSITIVITY IS A MEASURE OF THE LIGHT LEVEL AT 3200 Degrees KELVIN COLOR TEMPERATURE NECESSARY TO PRODUCE A FULL
1 VOLT PEAK TO PEAK VIDEO SIGNAL.
** MINIMUM ILLUMINATION = THE LIGHT LEVEL TO OBTAIN A RECOGNIZABLE VIDEO SCENE.
B/ W = BLACK / WHITE (MONOCHROME)
Table 5-2
Sensitivity of Representative CCD and CMOS Image Sensors
from hot sources (fires, bright lamps, gun barrel emission,
explosives and very hot, red hot, white hot objects) that
provide good thermal emission. Long-wave IR systems use
the differences in radiation from room temperature emitters
like humans, animals, vehicles, ships and aircraft (engine
areas), warm buildings, and other hot objects as compared
to their surroundings. The IR thermal camera is the only
system that can “see” when the visible or near-IR radiation
suitable for visible, near- or mid-IR sensors is too low to
detect. These systems see in total darkness and can often
“see” through smoke and fog.
The use of IR cameras relies on thermal differences (contrast)—heat emitted by target vs. heat emitted by the background surrounding it—thereby providing
images with better contrast than using ICCD image
intensification. Thermal sensors require very little temperature difference between the target and background for
the sensor to detect the target.
Thermal IR cameras look like video cameras in their
mechanical and electrical characteristics but the lenses
required are different in that the glass in standard visible light or near-IR cameras is replaced by a lens using
an infrared transmitting material such as germanium.
Thermal systems are readily available for security application but cost between 10 and 100 times more than standard
video cameras. These lower resolution IR cameras have a
comparatively small number of pixels that result in a pixilated picture, but there is often sufficient intelligence to
determine the objects or activity in the scene. Electronic
smoothing of the picture is often used to improve the
displayed scene. The use of pseudocolors, i.e. different
colors representing different temperatures, is a significant
aid in interpreting the scene. Medium resolution systems
typically have 320 × 256 pixel arrays and high resolution
systems have 640 × 512 arrays (military, very expensive).
See Chapter 19 for examples of thermal IR imagers.
The human body glows (radiates energy) like a 100
watt bulb in the IR spectrum but only if it is viewed in
the correct spectrum, i.e. the long-wave IR spectrum. The
wavelengths of the radiation emitted by most terrestrial
objects lie between about 3 and 12 m in the mid- and
far-IR region of the spectrum. The peak of the human
body radiation (at 98 Fahrenheit) is at about 9 m.
Infrared detectors fall into two different categories: photovoltaic and thermal. The photovoltaic detectors generate
an electrical current directly proportional to the number of photons incident on the detector. Thermal detectors respond to the change in resistance or some other
temperature-dependent parameter in the material. As the
absorbed light heats up, the material (pixels) changes in
132
CCTV Surveillance
resistance or capacitance producing a change in the electrical circuit.
Pyroelectric and bolometric detectors are the two types
of detectors that form the basis of most non-cooled thermal IR camera designs.
5.4.4 Sensor Fusion—Visible/IR
A technique called “multi-spectral imaging” in which an
image is displayed from two different detectors operating
at different wavelengths is finding increased use in the
security field.
Displaying the images from two different wavelength
regions (sensor fusion) on the same monitor significantly
increases intelligence obtained from the combined scene.
In the 3–5 micron region some targets and backgrounds
“reverse” their energy levels. This change can be detected
when the two signals are subtracted. In normal single
detector systems this signal reversal is averaged out and
not detected, thereby reducing detection capability.
A powerful sensor fusion technique uses the
combination of an image-intensified camera and a thermal
IR camera to significantly improve seeing under adverse
nighttime conditions having smoke, dust, and fog. The
fusion of near-IR and far-IR cameras with combined overlay display results in a significantly improved night vision
system. The system combines the strengths of image intensification (a clear sharp picture) with the advantages of
thermal IR (high detection capability). This provides the
ability to see in practically any non-illuminated nighttime
environmental condition.
5.5 CAMERA FEATURES—ANALOG/DIGITAL
Analog cameras are limited to a few automatic compensating functions: (1) automatic gain control (AGC), (2) white
light balance (WB). Digital cameras with DSP on the
other hand can have many automatic functions. Some are
described below.
5.5.1 Video Motion Detection (VMD)
The second-most often used intrusion-detection device
(first is the pyroelectric infrared (PIR)) in the security
industry is the VMD. The digital VMD uses an analog-todigital device to convert the analog video signal to a digital
signal. The DSP circuits then respond to the movement
or activity in the image as recognized as a specific type
and rate change within a defined area using a preset minimum sensitivity for size and speed. While PIR intrusion
sensors detect the change in the temperature of a particular part of the viewed area, the VMD senses a change
in the contrast within the camera scene from the normal quiescent video image. These digital VMD modules
are now small enough to be incorporated directly into a
video camera housing or larger, more sophisticated ones
connected in between the camera and the video monitor. These digital VMDs are much more immune to RFI
and EFI interferences and temperature changes that can
cause false alarms in the PIR devices. Prior analog VMD
technology exhibited an array of false alarm problems
related to changes in scene lighting, shadows, cable or
wireless transmission noise, etc. With the advancement of
CCD cameras and DSP circuitry, the reliability and false
alarm rate have been managed, resulting in reliable VMD
detectors with the CCD and CMOS cameras scene contrast analysis replaced by localized pixel analysis. It is now
possible to digitally analyze changes in individual or small
groups of pixels, resulting in increased levels of reliability
and reduced false alarm rate. Recent improvements in
the digital VMD have addressed problems associated with
false alarms due to foreign objects moving through the
field of view at rates of speeds too fast or too slow rates
to be of interest. The products available have automatic
adjustments (algorithms) to process the video signal data
to exclude these false alarms. Other false alarms caused by
natural weather changes, i.e. clouds coming into the field
of view, or small animals and birds or other debris passing
through the camera field of view have for the most part
been eliminated. These new digital systems have resulted
in low false alarm rates and systems that only respond to
intruders.
Digital VMDs do not require a computer for operation
and are usually provided with an RS232 interface for computer integration and remote programming and reporting. This approach to operation and control provides a
user-friendly interface to most users that are familiar with a
menu-driven screen and mouse operation. Physically they
consist of modular units or are designed in the form of
plug-in boards for easy installation into existing camera
equipment.
5.5.2 Electronic Zooming
Prior to video cameras incorporating DSP electronics, the
only option for zooming the video camera system was
through the use of zoom lens optics. Electronic zoom was
first perfected in consumer CCD and CMOS camcorders
and still cameras and then in the security industry. The
electronic zooming technique makes use of magnifying
the image electronically by selecting a portion of the sensor area and presenting its magnified video image onto
the monitor screen. Zoom ratios of from 5:1 to 20:1 are
available depending on the basic resolution of the sensor.
Since only a selected portion of the entire sensor is used,
electronic zooming can often be combined with electronic
pan and tilt by moving the area used in the sensor over
Cameras—Analog, Digital, and Internet
different parts of the entire sensor area. This results in
electronically panning and tilting while the camera and
lens are held stationary.
5.5.3 Electronic Shuttering
It is essential to match the camera sensor sensitivity to the
lighting in the scene. In general the more the lighting
available the less sensitive the camera has to be. Digital
signal processing technology permits the camera to adapt
to the scene illumination through the use of electronic
shuttering of the camera. The camera electronics adapt
so that it is optimally adjusted for the scene light level,
which changes the sensitivity of the sensor to compensate for varying light levels. This electronic sensitivity control (ESC) allows for small changes in light levels found
in indoor applications such as lobby areas, hallways with
external windows, storage areas, or where an outside door
is occasionally opened. It is not for use in outdoor applications having large light level changes (due to circuitry
limitations), where the use of an automatic-iris lens is usually required. It often permits the use of a manual-iris lens
assembly, which reduces the overall cost of the camera–
lens combination, rather than an auto-iris.
5.5.4 White Balance
Automatic white balance is required so that when the camera is initially turned on, it properly balances its color
circuits to a white background, which in turn is determined by the type of illumination at the scene. The camera
constantly checks the white-balance circuitry and makes
any minor compensation for variations in the illumination color temperature, i.e. the spectrum of colors in the
viewed scene.
Color cameras are sensitive to the color temperature of
light as defined by the color rendering index (CRI) of
light sources. A common problem for many color camera
systems is their inability to reproduce the exact color of
an object when using different light sources with different
CRIs. Color rendering is the term used to describe how
well a light source is able to produce the actual color of
the viewed object without causing a shift or error in color.
The color temperature determines the white component
of the light source composed of the totality of all the colors in the light source spectrum. Different types of lamps
produce different ranges of “white” light and these differences must be compensated for. This compensation is
performed by the WB circuits of the camera. Today’s DSP
cameras have automatic WB electronics that can adjust
between color temperatures from 2800 to 7600 K which
encompasses most lighting conditions. Chapter 3 shows
the spectral output from common light sources and video
camera spectral sensitivities used in security applications.
133
5.5.5 Video Bright Light Compression
One major improvement resulting from the use of DSP
in cameras is back light compensation (BLC). The DSP
camera with BLC adjusts to and can simultaneously view
dark and bright scene areas thereby increasing the camera
dynamic range by more than thirty times over conventional cameras. This technique is ideal for many applications where there are highly contrasted lighting conditions or where contrast conditions change throughout the
course of viewing. The camera accomplishes this by digitizing the image signal, at two different rates. Short times
(faster speed) register the bright image areas, and long
times (slower speed) register dark image areas. The two
signals are processed together in the camera and combined into a single signal at the output. Until the use of
BLC these conditions did not permit a clear view of the
entire image and required the use of high-end cameras
with digital back-light masking capabilities.
Back light compensation allows cameras to be pointing
at brightly lighted building entrances and exits, ATMs,
or underground parking facilities. Other applications
include casinos where interior lighting is designed to
brighten gaming and cash areas and to soften lounges,
seating areas, and aisles. Another application is a loading dock that is illuminated with different light levels
and poses a similar problem during the course of any
given day. Exterior lighting conditions in these areas
vary from dark to blinding sunshine. In another interior
application, jewelry counters often feature brightly illuminated display areas with subdued lighting in the surrounding areas. Now cameras with DSP compensation can
be used to continuously monitor both interior and exterior areas under virtually any lighting condition, applications that were previously not possible with analog camera
designs.
5.5.6 Geometric Accuracy
One of the significant advantages solid-state image sensors
have over their tube sensor predecessors is the precise geometric location of the pixels with respect to one another.
In a CCD, CMOS, or thermal IR sensor, the locations of
the individual photo-sensor pixel sites are known exactly
since they are determined during manufacture of the sensor and never move.
5.6 CAMERA RESOLUTION/SENSITIVITY
When classifying a video camera the two specifications
that are most important are the resolution and sensitivity. Unfortunately in many data sheets there is confusion
surrounding these terms.
134
CCTV Surveillance
Resolution. Resolution is the quality of definition and
clarity of the picture, and is defined in discernible TV
lines; the more the lines the higher the resolution and
the better the picture quality. Resolution is a function of
the number of pixels (picture elements) in the CCD chip.
In other words, the resolution is directly proportional to
the number of pixels in the CCD sensor. In some data
sheets, two types of resolution are defined: vertical and
horizontal. Vertical resolution is equal to the number of
discernible horizontal lines in the picture and is limited by
either the 525 or the 625 line resolution as defined in the
NTSC or CCIR standards. Horizontal resolution relates
to the number of lines reproduced in the picture in the
vertical direction, and depends on the bandwidth.
Sensitivity. Sensitivity is a measure of how low a light
level a camera can respond to and still produce a usable
or minimum quality picture. It is measured in FtCd or lux
for CCD, CMOS, and ICCD cameras operating and the
visible and near-IR wavelength range, and in delta-temp
(t ) in the mid- and far-IR. One FtCd equals approximately 9.3 lux. The smaller the number (FtCd, lux or t )
the more sensitive the camera. Typical values for state-ofthe-art cameras are: (1) monochrome camera 01−0001
lux, (2) color camera (single sensor) 1 FtCd–5 FtCd, (3)
thermal IR 0.1 t .
5.6.1 Vertical Resolution
Vertical resolution in the analog scanning system is derived
from the 504 effective scanning lines in the 525-line NTSC
television system. The camera scanning dissects a vertical
line appearing in the scene into 483 separate segments.
Since each scanning line on the monitor has a discrete
width, some of the scene detail between the lines is lost.
As a general rule approximately 30% of any scene is lost
(called the “Kell factor”). Therefore, the standard 525-line
NTSC television system produces 340 vertical TV lines of
resolution (483 effective lines × 07). In any standard 525line CCTV system, the maximum achievable vertical resolution is approximately 350 TV lines. In a 625-line system,
the maximum achievable vertical resolution is approximately 408 TV lines.
Vertical resolution in the digital system is just the number of vertical camera pixels. However, if a digital camera
is displayed on a 525 (or 625) line analog CRT display,
then the resolution is limited to the 350 (or 408) TV lines
of the analog system.
5.6.2 Horizontal Resolution
The NTSC standard provides a full video frame composed
of 525 lines, with 483 lines for the image and two vertical blanking intervals composed of 21 retrace lines each.
The TV industry adopted a viewing format with a widthto-height ratio of 4:3 and specifies horizontal resolution
in TV lines per picture height. The horizontal resolution on
the monitor tube depends on how fast the video signal
changes its intensity as it traces the image on a horizontal
line. The traditional method for testing and presenting
video resolution test results is to use the Electronic Industries Association (EIA) resolution target (Figure 5-18).
If only one resolution is defined in a camera data sheet,
the manufacturer is referring to the horizontal resolution.
There are several ways for measuring the horizontal resolution. The most common is to use a video resolution chart
which has horizontal and vertical lines as the target scene.
The camera resolution is the point where the lines start
to merge and cannot be separated. This chart-measuring
technique can be subjective since different people perceive, when the lines merge, differently. The resolution of
the monitor must be higher than the camera.
The minimum-spaced discernible black-and-white transition boundaries in the two wedge areas are the vertical
limiting (horizontal wedge) and horizontal limiting (vertical wedge) resolution values. Various industries using
electronic imaging devices have specified resolution criteria dependent on the particular discipline involved. In the
analog video security industry the concept of TV lines is
defined as the resolution parameter.
A more scientific technique for measuring the horizontal resolution is by measuring the bandwidth of the signal.
The bandwidth of the video signal from the camera is measured on an oscilloscope (see Chapter 25). Multiplying
the bandwidth by 80 TV lines/MHz gives the resolution
of the camera. For example if the bandwidth is 6 MHz the
camera resolution will be 6 × 80 or 480 TV lines.
The horizontal resolution is determined by the maximum
speed or frequency response (bandwidth) of the video electronics and video signal. While the vertical resolution is
determined solely by the number of lines or pixels chosen—
and thus not variable under the US standard of 525 lines—
the horizontal resolution depends on the electrical performance of the individual camera, transmission system,
and monitor. Most standard cameras with a 6 MHz bandwidth produce a horizontal resolution in excess of 450 TV
lines. The horizontal resolution of the system is therefore
limited to approximately 80 lines/MHz of bandwidth.
The solid-state-imaging industry has adopted pixels as its
resolution parameter. To obtain TV-line resolution equivalent when the number of pixels are specified, multiply
the number of pixels by 0.75. In photography, line pairs
or cycles per millimeter is the resolving power notation.
While all these parameters are useful, they tend to be confusing. For the purposes of CCTV security applications,
the TV line notation is used. For reference, the other
parameters are defined as follows:
• One cycle is equivalent to one line pair.
• One line pair is equivalent to two TV lines.
• One TV line is equivalent to 1.25 pixels.
Cameras—Analog, Digital, and Internet
135
INDICATES
VERTICAL
RESOLUTION
(200 TV LINES)
INDICATES
10 SHADES
OF GRAY SCALE
IN PICTURE
INDICATES
HORIZONTAL
RESOLUTION
AT EDGE
OF PICTURE
(200 TV LINES)
INDICATES HORIZONTAL
AND VERTICAL
RESOLUTION
AT CORNER
OF PICTURE
INDICATES HORIZONTAL
RESOLUTION
AT CENTER
OF PICTURE
(200 TV LINES)
NOTE: THE MINIMUM SPACED DISCERNIBLE BLACK AND WHITE TRANSITION BOUNDARIES
IN THE TWO WEDGE AREAS ARE THE VERTICAL (HORIZONTAL WEDGE) AND
HORIZONTAL (VERTICAL WEDGE) LIMITING RESOLUTION VALUES.
FIGURE 5-18
EIA resolution target
One cycle is equivalent to one black-and-white transition and represents the minimum sampling information
needed to resolve the elemental areas of the scene image.
A figure of merit for solid-state CCTV cameras is the total
number of pixels reproduced in a picture area. A typical
value is 380,000 pixels for a good 525-line CCTV system.
A parameter deserving mention that is used in lens,
camera, and image-intensifier literature is the modulation
transfer function (MTF). This concept was introduced to
assist in predicting the overall system performance when
cascading several devices such as the lens, camera, transmission medium, and monitor or recorder in one system.
The MTF provides a figure of merit for a part of the system
(such as the camera or monitor) acting alone or when
the parts are combined with other elements of the system.
It is used particularly in the evaluation of LLL devices
(Chapter 19).
The resolution for a good monochrome security camera
is 550−600 TV lines and for a color camera is 450−480 TV
lines. The data sheets from manufacturers of solid-state
cameras (and monitors) often quote the number of pixels
instead of TV line resolution. However, unless the number of pixels is converted into equivalent TV lines, it is
hard to compare picture resolution. Table 5-3 summarizes
the state of the art in solid-state sensors and gives information on the horizontal and vertical pixels available for
representative 1/6, 1/4-, 1/3-, and 1/2-inch format types.
When monochrome solid-state sensor cameras were first
introduced, the sensors had a maximum horizontal resolution of approximately 200 TV lines per picture height.
These early low-resolution sensors had 288 horizontal by
394 vertical pixels. Present-day sensors have horizontal resolutions of 400–600 TV lines per picture height. Mediumresolution camera sensors have 510H × 492(V) pixels,
and high-resolution cameras have 739H × 484(V) pixels.
Improvements in the resolution of solid-state sensors
to match the best tube sensors have resulted from various approaches with the most successful increase coming
from increased pixel density. These strides in decreasing
the pixel size have resulted from the techniques used to
136
CCTV Surveillance
TYPE
DESCRIPTION
HORIZONTAL VERTICAL
TOTAL
RESOLUTION
(TV LINES)
COMMENTS
1/6 CCD
COLOR (NTSC)
811
508
412,000
480
7 mm DIAMETER SENSOR HEAD
1/6 CCD
COLOR (NTSC)
736
480
340,000
470
ULTRA-FAST IP SPEED DOME
1/6 CCD
COLOR (PAL)
736
544
400,000
460
ULTRA-FAST IP SPEED DOME
1/4 CCD
COLOR (NTSC)
768
494
380,000
480
SURVEILLANCE
1/4 CCD
COLOR (NTSC), B/W
768
494
380,000
470
NETWORK IP SPEED DOME
1/4 CCD
COLOR (PAL), B/W
752
582
438,000
470
NETWORK IP SPEED DOME
1/4 CMOS
COLOR (NTSC)
640
480
307,200
480
NETWORK IP
1/3 CCD
MONOCHROME (NTSC)
510
492
251,000
380
SURVEILLANCE
1/3 CCD
MONOCHROME (CCIR)
512
582
297,000
380
1/3 CCD
COLOR (NTSC), B/W
771
492
380,000
480/570 (B/W)
SURVEILLANCE
DAY/NIGHT SURVEILLANCE
1/3 CCD
COLOR (NTSC)
768
494
380,000
480
SURVEILLANCE
1/3 CCD
COLOR (PAL)
811
508
412,000
480
SURVEILLANCE
1/3 CMOS
COLOR (NTSC)
640
480
307,200
340
COVERT SURVEILLANCE
1/2 CCD
COLOR (NTSC), B/W
768
494
380,000
480
DAY/NIGHT SURVEILLANCE
1/2 CCD
COLOR (PAL), B/W
752
582
440,000
480
DAY/NIGHT SURVEILLANCE
1/2 CCD
MONOCHROME (NTSC)
811
508
412,000
570
HIGH RESOLUTION B/W
*RESOLUTION IS THE ABILITY TO JUST DISCERN TWO ADJACENT BLACK LINES SEPARATED BY A WHITE SPACE. THE SYSTEM SHOULD HAVE A GRAY SCALE
WITH A MINIMUM OF 10 LINES FROM BLACK TO WHITE.
FOR DIGITAL VIDEO SYSTEMS THE HORIZONTAL AND VERTICAL RESOLUTIONS ARE APPROXIMATELY 0.75 × NUMBER OF PIXELS.
FOR LEGACY NTSC AND PAL SYSTEMS, VERTICAL RESOLUTION IS LIMITED BY THE 525 AND 625 HORIZONTAL LINE SCAN RATE AND THE HORIZONTAL
RESOLUTION BY THE SYSTEM BANDWIDTH.
B/W = BLACK/WHITE (MONOCHROME)
Table 5-3
Resolution of Representative Solid-State CCD and CMOS Cameras
manufacture very large scale integrated (VLSI) devices for
computers. Image sensors are VLSI devices. The majority
of solid-state sensors in use today have a 1/4-, 1/3- or 1/2inch image format. There are some available with 1/5- and
1/6-inch image formats, and larger ones with 2/3-inch
formats.
Several other techniques are used to improve resolution.
In one camera configuration, image-shift enhancement
results in a doubling of the ILT CCD imager horizontal
resolution by shifting the visual image in front of the CCD
sensor by one-half pixel. This technique simultaneously
reduces aliasing, which causes a fold-back of the highfrequency signal components, resulting in "herringbone"
or jagged edges in the image. This artifact is often seen
when viewing plaid patterns on clothing and screens, with
medium to low resolution solid-state cameras. Aliasing
reduces resolution and causes considerable loss in picture
intelligence.
Another technique used to improve the horizontal resolution without increasing the pixel count is offsetting
each row of pixels by one-half pixel, generating a zigzag
of the pixel rows. This arrangement, in conjunction with
corresponding clocking, allows simultaneous readout of
two horizontal rows and nearly doubles the horizontal
resolution compared with conventional detectors with
identical pixel counts.
5.6.3 Static vs. Dynamic Resolution
The previous section described static resolution. This represents resolution achieved when a camera views a stationary
scene. When a camera views a moving target—a person
walking through the scene, a car passing by—or the camera scans a scene, a new parameter called dynamic resolution
is defined. Under either the moving-target or scanning
condition, extracting intelligence from the scene depends
on resolving, detecting, and identifying fine detail. The
solid-state camera has the ability to resolve rapid movement without degradation in resolution under almost all
suitable lighting conditions.
When high resolution is required while viewing very
fast moving targets, solid-state cameras with an electronic
shutter are used to capture the action. Many solid-state
cameras have a variable-shutter-speed function, with common shutter speeds of 1/60, 1/1000, and 1/2000. This
shuttering technique is equivalent to using a fast shutter
Cameras—Analog, Digital, and Internet
speed on a film camera. The ability to shutter solid-state
cameras results in advantages similar to those obtained in
photography: the moving object or fast-scan panning that
would normally produce a blurred image can now produce
a sharp one. The only disadvantage this technique has is
that since a decreased amount of light enters the camera,
the scene lighting must be adequate for the system to work
successfully.
5.6.4 Sensitivity
Sensitivity of a camera is measured in foot candles
(FtCd) or lux (1 FtCd = 9.3 lux) and usually refers to
the minimum light level required to get an acceptable
video picture. There is a great deal of confusion in the
video industry over camera specifications with respect to
what an acceptable video picture is. Manufacturers use
two definitions for camera sensitivity: (1) sensitivity at
the camera sensor faceplate and (2) minimum scene
illumination.
Sensitivity at the faceplate indicates the minimum light
required at the sensor chip to get an acceptable video picture. Minimum scene illumination indicates the minimum
light required at the scene to get an acceptable video picture. When sensitivity is defined as the minimum scene
illumination, parameters such as the scene reflectance,
the lens optical speed (f/#), usable video, automatic
gain control (on, off), and shutter speed should be
defined.
With regard to reflectance, most camera manufactures
use 89% or 75% (white surface) reflectance surface to
define the minimum scene illumination. If the actual
scene being viewed has the same reflectance as the data
sheet then this is a correct measurement. This is usually
not the case. Typical light reflectivities of different materials range from snow 90%, grass 40%, brick 25%, to blacktop 5%. It is apparent that if the camera is viewing a black
car, only about 5% of the light is reflected back to the
camera and therefore at least fifteen times more light is
required at the scene to give the same amount of light
that would come from a white surface.
One camera technology that significantly increases the
sensitivity of the CCD sensor over existing devices by a factor of two uses an on-chip lens (OCL) technique. By manufacturing the sensor with microscopic lenses on each pixel,
the incoming light is concentrated on the photo-sensor
areas thereby increasing the sensitivity of the camera.
An improvement particularly important in CMOS sensors
incorporates microscopic lenses that cover the active area
of each pixel as well as the inactive area between pixels, thereby eliminating the ineffective areas between the
microlenses. This increases sensitivity by over a factor of
two and reduces the smear level significantly compared to
that of the original technology.
137
5.7 SENSOR FORMATS
The development of the superior solid-state CCD sensor
color camera for the VCR home consumer market accelerated the use of color cameras in the security industry.
There are three popular image format sizes for solid-state
security cameras: 1/4-, 1/3-, and 1/2-inch. All security sensor formats have a horizontal-by-vertical geometry of 4 × 3
as defined in the EIA and NTSC standards. For a given
lens, the 1/4-inch format sensor sees the smallest scene
image and the 1/2-inch sees the largest, with the 1/3-inch
format camera seeing proportionally in between.
The ISIT tube cameras using the 1-inch tube to provide
LLL capabilities have by all intents and purposes been
replaced by their solid-state counterpart, the ICCD. As
a basis for comparison with other formats, Figure 5-19
shows the solid state CCD, CMOS, and tube image formats
compared to photographic film formats.
For reference the 16 mm semiprofessional film camera,
and the 35 mm film camera used for bank holdup and
forensic applications is shown. Table 5-4 lists the three
popular video image format sizes: 1/4-, 1/3-, and 1/2-inch,
and four less used sizes: 1-, 2/3-, 1/6-, and 1/5-inch.
For reference, the physical target area in tube cameras
is circular and usually corresponds to the diagonal of the
lens image circle. The tube active target is the inscribed
4 × 3 rectangular aspect ratio area scanned by the electron
beam in the tube. Since each pixel is used in the solid-state
camera image the target area in the solid-state sensor is the
full sensor 4 × 3 format array. The camera sensor format
is important since it determines the lens format size with
which it must operate and, along with the lens focal length
(FL), sets the video system field of view (FOV).
As a general rule, the larger the sensor size, the larger
the diameter of the lens glass size required which translates into increased lens size, weight, and cost. Any lens
designed for a larger format can be used on a smaller
format camera. The opposite is not true, for example a
lens designed for a 1/3-inch format will not work properly
on a 1/2-inch format camera and will produce vignetting
(dark area surrounding the image).
5.7.1 Solid-State
Most solid-state cameras using CCD or CMOS sensor
technology have 1/4-, 1/3-, and 1/2-inch formats. The
sensor arrays are rectangular in shape and have the active
area sizes as listed in Table 5-4 and shown in Figure 519. Significant progress has been made in producing
exceptionally high-quality 1/4-, 1/3-, and 1/2-inch format
sensors that rival the sensitivity of some of earlier larger
2/3- and 1-inch solid-state or tube sensors. Most color
cameras used in security applications have single-chip
sensors with three-color stripe filters integral with the
image sensor. Typical sensitivities for these color cameras
138
CCTV Surveillance
1/3"
TUBE (REFERENCE)
• VIDICON
• SILICON
• SIT, ISIT
NOMINAL
1" TUBE
DIAMETER
1"
2/3"
1/2"
N/A
4.8 × 6.4
(8 DIAG.)
SOLID STATE
1/6"
6.6 × 8.8
(11 DIAG.)
1/4"
1/3"
ACTIVE
SENSOR
AREA (mm)
9.6 × 12.8
(16 DIAG.)
1/2"
2/3"
• CCD
• CMOS
• CID
2.4 × 3.2
(4.0 DIAG.)
1.8 × 2.4
(3.0 DIAG.)
3.6 × 4.8
(6.0 DIAG.)
4.8 × 6.4
(8 DIAG.)
6.6 × 8.8
(11 DIAG.)
35 mm FILM
16 mm
SUPER 8 mm
FILM (REFERENCE)
• SUPER 8 mm
• 16 mm
• 35 mm
4.1 × 5.8
7.4 × 10.3
24 × 36
NOTE: ALL DIMENSIONS IN mm
FIGURE 5-19
Tube, solid state and film image formats
IMAGE SENSOR SIZE
DIAGONAL (d )
FORMAT
HORIZONTAL (h )
VERTICAL (v )
mm
inches
mm
inches
mm
inches
16
0.63
12.8
0.50
9.6
0.38
2/3"
11
0.43
8.8
0.35
6.6
0.26
* 1/2"
8
0.31
6.4
0.25
4.8
0.19
* 1/3"
6
0.24
4.8
0.19
3.6
0.14
* 1/4"
4
0.16
3.2
0.13
2.4
0.1
1/6"
3
0.12
2.4
0.09
1.8
0.07
1"
(REFERENCE)
* MOST COMMON CCTV SENSOR FORMATS
Table 5-4
CCTV Camera Sensor Formats
range from 0.5 to 2.0 FtCd (4.6 to 18.6 lux) for full
video, which is less sensitive than their monochrome
counterpart by a factor of about 10. Low resolution color
cameras have a horizontal resolution of about 330 TV
lines. High-resolution color cameras have a horizontal
resolution of about 480 TV lines.
5.7.2 Image Intensifier
The most common image intensifier is the ICCD and
uses standard monochrome resolution CCD image formats. Typical values for the format resolution are 500–600
for a 1/2-inch sensor.
Cameras—Analog, Digital, and Internet
5.7.3 Thermal IR
The thermal IR camera uses a long-wave IR array fabricated using completely different manufacturing techniques as compared with CCD or ICCD manufacture.
These sensors are far more difficult to manufacture and
have far lower yields than do other solid-state sensors. As
a result the number of pixels in the sensor is significantly
less. Typical sensor arrays have 280–320 horizontal TV line
resolution. Future generations of these thermal IR cameras will have near equivalent resolution to those of CCD
and CMOS cameras.
5.8 CAMERA LENS MOUNTS
Several lens-to-camera mounts are standard in the CCTV
industry. Some are mechanically interchangeable and
others are not. Care must be taken so that the lens mount
matches the camera mount. The two widely used cameralens mounts are the C and CS mount. Small surveillance cameras use the 10, 12, and 13 mm thread diameter
minilens mounts. The 10 and 12 mm diameter mounts
have a 0.5 mm pitch and the 13 mm diameter mount has a
1.0 mm pitch. Large bayonet mounts are used with specialized cameras and lenses on some occasions. These lens-tocamera mounts are described in the following sections.
139
of the CS mount system is that the lens can be smaller,
lighter, and less expensive than its C mount counterpart.
The CS mount camera is completely compatible with the
common C mount lens when a 5 mm spacer ring is inserted
between the C mount lens and the CS mount camera. The
opposite is not true: a CS mount lens will not work on a
C mount camera. Table 5-5 summarizes the present lens
mount parameters.
5.8.2 Mini-Lens Mounts
The proliferation of small minilenses (see Chapter 4) and
small CCD and CMOS cameras has led to widespread
use of smaller lens/camera mounts. Manufacturers supply
these mini-lenses and cameras with mounts having metric
thread sizes of 10, 12, or 13 mm diameter and thread
pitches of 0.5 and 1.0 mm. The two widely used sizes are
the 10 and 12 mm diameter with 0.5 mm pitch.
5.8.3 Bayonet Mount
The large 2.25-inch-diameter bayonet mount is used
primarily in custom security, industrial, broadcast, and
military applications with three-sensor color cameras, LLL
cameras, and long FL large lenses. It is only in limited use
in the security field.
5.8.1 C and CS Mounts
5.8.4 Lens–Mount Interferences
For many years, all 1-, 2/3-, and 1/2-inch cameras used an
industry-standard mount called the C mount to mechanically couple the lens to the camera. Figure 5-20 shows the
mechanical details of the C and CS mounts.
The C mount camera has a 1-inch-diameter hole with
32 threads per inch (TPI) and the C mount lens has a
matching thread (1–32 TPI) that screws into the camera
thread. The distance between the lens rear mounting surface and the image sensor for the C mount is 0.69 inches
(17.526 mm).
With the introduction of the smaller 1/4- and 1/3inch (and 1/2-inch) format cameras and lenses, it became
possible and desirable to reduce the size of the lens and
the distance between the lens and the sensor. A mount
adopted by the industry for 1/4-, 1/3, and 1/2-inch-sensorformat cameras became the CS mount. The CS mount
matches the C mount in diameter and thread but the
distance between the lens rear mounting surface and the
image sensor for the CS mount is 0.492 inches (12.5 mm).
The CS mount is 0.2 inches (5 mm) shorter than the C
mount. Since the lens is 5 mm closer to the sensor, the
lens can be made smaller in diameter. A C mount lens
can be used on a CS mount camera if a 5 mm spacer is
interposed between the lens and the camera and if the
lens format covers the camera format size. The advantage
Figure 5-21 illustrates a potential problem with some lenses
when used with CCD or solid-state cameras. Some of the
shorter-FL lenses (2.2, 2.6, 3.5, and 4.8 mm) have a protrusion that extends behind the C or CS mount or minimount and can interfere with the filter or window used
with the solid-state sensor. This mechanical interference
prevents the lens from fully seating in the mount, thereby
causing the image to be out of focus. Most lens and
camera manufacturers are aware of the problem and for
the most part have designed lenses and cameras that are
compatible. However, since lenses are often interchanged,
the potential problem exists and the security designer
should be aware of the condition.
5.9 ZOOM LENS–CAMERA MODULE
The requirement for a compact zoom lens and camera
combination has been satisfied with a zoom lens–camera
module. This module evolved out of a requirement for a
lightweight, low inertia camera-lens for use in high-speed
pan/tilt dome installations in casinos, airports, malls, retail
stores, etc. The camera–lens module has a mechanical cube
configuration so that it can easily be incorporated into a
140
CCTV Surveillance
1" DIAMETER
32 THREADS PER INCH
CAMERA
1" DIAMETER
32 THREADS PER INCH
C MOUNT
LENS
CAMERA
CS MOUNT
LENS
SENSOR
SENSOR
0.492"
(12.5 mm)
0.69"
(17.526 mm)
5 mm SPACER
C MOUNT
LENS
CS MOUNT
LENS
=
+
NOTE: DIFFERENCE BETWEEN C MOUNT AND CS MOUNT: 17.526 mm – 12.5 mm = 5 mm (SPACER)
FIGURE 5-20
Mechanical details of the C mount and CS mount
CAMERA
MOUNT TYPE
MOUNTING SURFACE
TO SENSOR DISTANCE (d )
MOUNT TYPE
inch
mm
THREAD:
DIAMETER (D)
C
0.069
17.526
1-inch DIA.
32 TPI
CS
0.492
12.5
1-inch DIA.
32 TPI
10 mm DIA.
0.50 mm PITCH
MINI: 10 mm
MINI: 12 mm
VARIES FROM *
3.5 mm (0.14")
TO 9 mm (0.35")
MINI: 12 mm
LENS
D
d
12 mm DIA.
0.50 mm PITCH
13 mm DIA.
0.50 mm PITCH
TPI—THREADS PER inch
TPM—THREADS PER mm
* VARIES WITH MANUFACTURER
TO CONVERT A C MOUNT LENS TO A CS MOUNT, ADD A 5 mm SPACER
Table 5-5
Standard Camera/Lens Mount Parameters
pan/tilt dome housing and be pointed in any direction at
high speeds (Figure 5-22).
The module assembly includes the following components and features: (1) rugged, compact mechanical structure suitable for high-speed pan/tilt platforms; (2) large
optical zoom ratio, typically 16 or 20 to 1; (3) large elec-
tronic zoom ratio, typically 8 or 10 to 1; and (4) a 1/4-inch
solid-state color camera with excellent sensitivity and resolution. Options include: (1) automatic focus and (2) image
stabilization capability (see Section 4.5.11).
The automatic-focusing option is useful providing the
lens is zooming slowly and the module is not panning or
Cameras—Analog, Digital, and Internet
141
FIGURE 5-21 Lens-mount
interference
CROSSHATCHED
AREA REPRESENTS
MECHANICAL
INTERFERENCE
INFRARED
FILTER
C MOUNT
LENS
CAMERA
SENSOR
MECHANICAL
INTERFERENCE
BETWEEN LENS
AND FILTER
at the 3.6 mm FL setting). At the wide-angle setting the
lens and camera covers a 54 horizontal angular FOV. At
the telephoto setting, it covers a 25 horizontal angular
FOV. The lens–camera module is also available, packaged
for mounting on standard pan/tilt platforms.
5.10 PANORAMIC 360 CAMERA
FIGURE 5-22
Zoom lens–camera module
tilting rapidly. When a person walks into the lens FOV
the automatic-focus lens changes focus from the surrounding scene to the moving person, keeping the person in
focus. The auto-focus system keeps the person in focus
even though they move toward or away from the lens.
Auto-focus is ineffective while the lens is zooming and
should not be used if the module is panning and/or tilting
rapidly. In this situation the system becomes “confused”
and does not know what object to focus on, causing the
person to be out of focus in the picture. The zoom lens
in a typical module has an FL range of 36−80 mm (f/1.6
There has always been a need to see “all around” an
entire room or area, seeing 360 horizontally and 90 vertically with one panoramic camera and lens. Early versions
of such a 360 FOV camera systems were achieved using
multiple cameras and lenses and combining the scenes
as a split screen on the monitor. Panoramic lenses have
been available for many years but have only recently been
combined with high resolution digital cameras and DSP
electronics using sophisticated mathematical transforms
to take advantage of their very wide-angle capabilities. The
availability of high resolution solid-state cameras has made
it possible to map a 360 by 90 hemispherical FOV onto
a rectangular monitor with good resolution. Figure 4-31
shows a panoramic camera and operational diagram having a 360 horizontal and a 90 vertical FOV.
In operation, the lens collects light from the 360
panoramic scene and focuses it onto the camera sensor as
a donut-shaped image (Figures 4-31 and 4-32). The electronics and mathematical algorithm convert this donutshaped panoramic image into the rectangular (horizontal
and vertical) format for normal monitor viewing. In operation, a joystick or computer mouse is used to electronically
142
CCTV Surveillance
FIGURE 5-23 High
definition television
(HDTV) formats
16
4
HDTV
16:9
9
FORMAT
HORIZONTAL
(PIXELS)
4:3
3
VERTICAL
(PIXELS)
VERTICAL
TV LINES
ASPECT RATIO
16 × 9
4×3
ARRAY SIZE:
PIXELS
HDTV 720i
1280
720
2,073,000
HDTV 720p
1280
720
2,073,000
HDTV 1080i
1920
1080
2,073,000
HDTV 1080p
1920
1080
2,073,000
NTSC *
525
921,600
PAL /SECAM *
625
921,600
* ANALOG REFERENCE: STANDARD TELEVISION (SDTV)
pan and tilt the camera so that at any given time a segment
of the 360 horizontal by 90 vertical image is displayed
on the monitor.
5.11 HIGH DEFINITION TELEVISION (HDTV)
High definition television (HDTV) provides a new video
display format having a 16×9 horizontal by vertical format,
thereby providing a significantly increased resolution over
that of standard NTSC 4 × 3 format (Figure 5-23).
The reason for defining this new format is to provide:
(1) a higher resolution or definition video display, (2) one
that has a format that better matches the view seen by
the human eye (wider horizontal view), and (3) a format more closely matching the many images that the eye
sees, i.e. landscapes, parking lots, etc. This new format was
originally developed for the consumer market; however, it
will find its way into the video security market because of
the superior monitor display format and resolution it provides. The new HDTV format and size has many variations
and has not yet been standardized in the security industry.
Not all HDTV images have the same number of horizontal lines or the same resolutions. The way the different
picture formats are painted on the screen is also different. HDTV formats available include: 720p, 1080i, and
1080p/24. The first number in the type designation is the
vertical resolution or how many scan lines there are from
the top to the bottom of the picture. This first designation
is usually followed by a letter. The letter is either an “i”
or “p.” These are the abbreviations for interlaced (i) or
progressive (p) scans respectively. Progressive means that
the whole picture is painted from the top of the screen to
the bottom and then a new frame is painted over again.
Interlaced means only half the image is painted first (oddnumbered lines) and then the other half of the image is
painted (even-numbered lines). There seems to be a general consensus that the progressive scan is better than the
interlaced. All present NTSC video security video systems
using the 4 × 3 format use 2:1 interlaced lines and every
computer monitor uses progressive. The last number in
the designation 24, 30, or 60 refers to the frame rate.
At present, the best HDTV system is 1080i, and interlaced 30 frame/60 fields per second, system similar to
NTSC, but with the 16 × 9 picture format of HDTV. HDTV
video improves the intelligence provided in many security displays since it presents a wider horizontal aspect
ratio, has higher resolution, and can support a larger
screen size. The increased resolution produces crisper,
sharper images.
5.12 SUMMARY
There have been many important improvements and innovations in the development of the video camera and its use
in the security field. The single most significant advances
in CCTV camera technology have been the development
of the CCD and CMOS solid-state camera image sensor, IR thermal cameras, IP camera, and the use of DSP.
These sensors and camera electronics offer a compelling
advantage over original vacuum-tube technology because
of solid-state reliability, inherent long life, low cost,
low-voltage operation, low power dissipation, geometric
Cameras—Analog, Digital, and Internet
reproducibility, absence of image lag, DSP, and visible
and/or IR response. These solid-state cameras have provided the increased performance, reliability, and stability needed in monochrome, color, and IR video security
systems.
The availability of solid-state color cameras has made
a significant impact on the security video industry. Color
cameras provide enhanced video surveillance because of
their increased ability to display and recognize objects
and persons. The choices available for lighting in most
security applications is sufficient for most color cameras
143
to have satisfactory sensitivity and resolution. Solid-state
color cameras have excellent color rendition, maintain
color balance, and need no color rebalancing when light
level or lighting color temperature varies.
Intensified charge coupled device cameras coupled to
tube or micro-channel plate intensifiers provide the low
light sensitivity required in dawn to dusk applications
and some nighttime applications. Room temperature,
thermal IR cameras have provided the “eyes” when no
visible or near-IR light is available and visible sensors are
inoperable.
This page intentionally left blank
Chapter 6
Analog Video, Voice, and Control
Signal Transmission
CONTENTS
6.1
6.2
6.3
Overview
Base-band Signal Analysis
6.2.1 Video Picture Signal
6.2.2 Video Synchronization Signal
6.2.3 Voice Signal
6.2.4 Control Data Signals
6.2.5 Modulation and Demodulation
6.2.6 Signal Bandwidth
Wired Video Transmission
6.3.1 Coaxial Cable
6.3.1.1 Unbalanced Single-Conductor
Cable
6.3.1.2 Connectors
6.3.1.3 Amplifiers
6.3.2 Balanced Two-Conductor Twin-axial Cable
Transmission
6.3.2.1 Indoor Cable
6.3.2.2 Outdoor Cable
6.3.2.3 Electrical Interference
6.3.2.4 Grounding Problems
6.3.2.5 Aluminum Cable
6.3.2.6 Plenum Cable
6.3.3 Two-Wire Cable Unshielded Twisted Pair
(UTP) Transmission
6.3.3.1 Balanced 2-Wire Attributes
6.3.3.2 The UTP Technology
6.3.3.3 UTP Implementation with Video,
Audio, and Control Signals
6.3.3.4 Slow-Scan Transmission
6.3.4 Fiber-Optic Transmission
6.3.4.1 Background
6.3.4.2 Simplified Theory
6.3.4.3 Cable Types
6.3.4.3.1 Multimode Step-Index
Fiber
6.3.4.3.2
6.4
6.5
Multimode Graded-Index
Fiber
6.3.4.3.3 Cable Construction and
Sizes
6.3.4.3.4 Indoor and Outdoor
Cables
6.3.4.4 Connectors and Fiber Termination
6.3.4.4.1 Coupling Efficiency
6.3.4.4.2 Cylindrical and Cone
Ferrule Connector
6.3.4.4.3 Fiber Termination Kits
6.3.4.4.4 Splicing Fibers
6.3.4.5 Fiber-Optic Transmitter
6.3.4.5.1 Generic Types
6.3.4.5.2 Modulation Techniques
6.3.4.5.3 Operational Wavelengths
6.3.4.6 Fiber-Optic Receiver
6.3.4.6.1 Demodulation
techniques
6.3.4.7 Multi-Signal, Single-Fiber
Transmission
6.3.4.8 Fiber Optic—Advantages/
Disadvantages
6.3.4.8.1 Pro
6.3.4.8.2 Con
6.3.4.9 Fiber-Optic Transmission: Checklist
Wired Control Signal Transmission
6.4.1 Camera/Lens Functions
6.4.2 Pan/Tilt Functions
6.4.3 Control Protocols
Wireless Video Transmission
6.5.1 Transmission Types
6.5.2 Frequency and Transmission Path
Considerations
145
146
CCTV Surveillance
6.5.3
6.6
6.7
6.8
6.9
6.10
6.11
Microwave Transmission
6.5.3.1 Terrestrial Equipment
6.5.3.2 Satellite Equipment
6.5.3.3 Interference Sources
6.5.4 Radio Frequency Transmission
6.5.4.1 Transmission Path Considerations
6.5.4.2 Radio Frequency Equipment
6.5.5 Infrared Atmospheric Transmission
6.5.5.1 Transmission Path Considerations
6.5.5.2 Infrared Equipment
Wireless Control Signal Transmission
Signal Multiplexing/De-multiplexing
6.7.1 Wideband Video Signal
6.7.2 Audio and Control Signal
Secure Video Transmission
6.8.1 Scrambling
6.8.2 Encryption
Cable Television
Analog Transmission Checklist
6.10.1 Wired Transmission
6.10.1.1 Coaxial Cable
6.10.1.2 Two-Wire UTP
6.10.1.3 Fiber-Optic Cable
6.10.2 Wireless Transmission
6.10.2.1 Radio Frequency (RF)
6.10.2.2 Microwave
6.10.2.3 Infrared
Summary
6.1 OVERVIEW
Closed circuit television (CCTV) and open circuit television (OCTV) video signals are transmitted from the
camera to a variety of remote monitors via some form
of wired or wireless transmission channel. Control,
communications, and audio signals are also transmitted
depending on the system. This chapter covers most of the
analog techniques for transmitting these signals. Chapter 7
describes the techniques for transmission of digital signals.
Analog transmission is still critically important because of
the immense installed base of analog equipment in the
security field. These video systems are in operation, and
will remain so for many years to come.
In its most common form, the video signal is transmitted
at base-band frequencies over a coaxial cable. This chapter
identifies techniques and problems associated with transmitting video and other signals from the camera site to the
remote monitoring location using wired copper-wire and
fiber optics, and through-the-air wireless transmission.
Electrical-wire techniques include coaxial cable and twowire unshielded twisted-pair (UTP). Coaxial cable is suitable for all video frequencies with minimum distortion
or attenuation. Two-wire UTP systems using standard conductors (intercom wire, etc.) use special transmitters and
receivers that preferentially boost the high video frequencies to compensate for their loss over the wire length.
Faithful video signal transmission is one of the most
important aspects of a video system. Each color video
channel requires approximately a 6 MHz bandwidth.
Monochrome picture transmission needs only a 4.2 MHz
bandwidth. Figure 6-1 shows the single-channel video
bandwidth requirements for monochrome and color systems.
Using information from other chapters, it is not difficult to specify a good lens, camera, monitor, and video
recorder to produce a high-quality picture. However, if
means of transmission does not deliver an adequate signal
from the camera to the monitor, recorder, or printer, an
unsatisfactory picture will result. The final picture is only
as good as the weakest link in the system and it is often
the transmission means. Good signal transmission requires
that the system designer and installer choose the best transmission type, and use high-quality materials, and practices
professional installation techniques. A poor transmission
system will degrade the specifications for camera, lens,
monitoring, and recording equipment.
Fiber optics offers a technology for transmitting highbandwidth, high-quality, multiplexed video pictures, and
audio and control signals over a single fiber. Fiber-optic
technology has been an important addition to video signal transmission means. The use of fiber-optic cable has
significantly improved the picture quality of the transmitted video signal and provided a more secure, reliable, and
cost-effective transmission link. Some advantages of fiber
optics over electrical coaxial-cable or two-wire UTP systems
include:
• high bandwidth providing higher resolution or simultaneous transmission of multiple video signals;
• no electrical interference to or from other electrical
equipments or sources;
• strong resistance to tapping (eavesdropping), thereby
providing a secure link; and
• no environmental degradation: unharmed by corrosion,
moisture, and electrical storms.
Wireless transmission techniques use radio frequencies
(RF) in the very high frequency (VHF) and ultra high
frequency (UHF) bands, as well as microwave frequencies at 900 MHz, 1.2 GHz, and 2.4 GHz and 5.8 GHz in the
S and X bands (2–50 GHz). Low-power microwave and RF
systems can transmit up to several miles with excellent picture quality, but the higher power systems require an FCC
(Federal Communications Commission) license for operation. Wireless systems permit independent placement of
the CCTV camera in locations that might be inaccessible
for coaxial or other cables.
Cable-less video transmission using IR atmospheric
propagation is discussed. Infrared laser transmission
requires no FCC approval but is limited in range depending on visibility. Transmission ranges from a few hundred
Analog Video, Voice, and Control Signal Transmission
RELATIVE
POWER
VIDEO INFORMATION
(AMPLITUDE MODULATION)
PICTURE
CARRIER
147
SOUND
CENTER
FREQUENCY
COLOR
SUBCARRIER
1.0
AUDIO
INFORMATION
(FREQUENCY
MODULATION)
0
1
1.25
2
3
4
3.58
5
6
.25
FREQUENCY
(MHz)
4.5
6.00
FIGURE 6-1
Single channel CCTV bandwidth requirements
feet in poor visibility to several thousand feet or even many
miles in good visibility.
Infrared is capable of bidirectional communication;
control signals are sent in the opposite direction to the
video signal and audio is sent in both directions.
The wired and wireless transmission techniques outlined
above account for the majority of transmission means from
the remote camera site to the monitoring site. There are,
however, many instances when the video picture must be
transmitted over very long distances—tens, hundreds, or
thousands of miles, or across continents. These are accomplished using digital techniques (Chapter 7). Two-wire,
coaxial, or fiber-optic cables for real-time transmission are
often not practical in metropolitan areas where a video
picture must be transmitted from one building to another
building through congested city streets not in sight of
each other.
A technique developed in the 1980s for transmitting
a video picture anywhere in the world over telephone
lines or any two-wire network is called “slow-scan video
transmission.” This technique uses a non-real-time twowire technology that permits the transmission of a video
picture from one location to any other location in the
world, providing that a two-wire or wireless voice-grade
link (telephone line) is available. This system was the
forerunner to the present Internet, intranet, and World
Wide Web (WWW).
The slow-scan system took a real-time camera video
signal and converted into a non-real-time signal and
transmitted it at a slower frame rate over any audio communications channel (3000 Hz bandwidth). Unlike conventional video transmission, in which a real-time signal
changed every 1/30th of a second, the slow-scan transmission method sent a single snapshot of a scene over a time
period of 1–72 seconds depending on the resolution specified. This effect is similar to that of opening your eyes
once every second or once every minute or somewhere in
between. When used with an alarm intrusion or VMD the
slow-scan equipment began sending pictures, once every
few seconds at low resolution (200 TV lines), or every
32 seconds at high resolution (500 TV lines).
A quantum change and advancement has occurred
in the video surveillance industry in the past five years.
Computer based systems now use digital techniques and
equipments from the camera, transmission means, switching and multiplexing equipment, to the DVRs, solid-state
LCD, and plasma monitors. The most dramatic change,
however, has been in the use of digital transmission
(Chapter 7). Now with the Internet and WWW, and digital
signal compression a similar function but much improved
transmission is accomplished over any wired or wireless
network.
A basic understanding of the capabilities of the aforementioned techniques, as well as the advantages and
148
CCTV Surveillance
disadvantages of different transmission means, is essential to optimize the final video picture and avoid costly
retrofits. Understanding the transmission requirements
when choosing the transmission means and hardware is
important because it constitutes the most labor-intensive
portion of the video installation. Specifying, installing,
and testing the video signal and communication cables
for intra-building and inter-building wiring represents the
major labor cost in the video installation. If the incorrect
cable is specified and installed, and must be removed and
replaced with another type, serious cost increases result.
In the worst situation, where cables are routed in underground outdoor conduits, it is imperative to use the correct size and type of cable so as to avoid retrenching or
replacing cables.
The horizontal line timing pulses occur at 63.5 microsecond intervals. The CCIR/PAL vertical standard timing is
1/50 second, 1/25 second, and 64 microseconds. The magnitude of these timing pulses is shown in Chapter 5.
6.2.3 Voice Signal
In the NTSC standard, the voice and sound information
is contained in a sub-carrier centered at 4.5 MHz and
at 4.5, 5.5, 6.0 and 6.5 MHz in the CCIR/PAL systems.
The signal is frequency modulated (FM) for high fidelity
reproduction.
6.2.4 Control Data Signals
6.2 BASE-BAND SIGNAL ANALYSIS
The video signal generated by the analog camera is called
the “base-band video signal.” It is called base-band because
it contains low frequencies, from 30 Hz for NTSC (25 hertz
for CCIR) to 6 MHz. To accomplish fiber-optic transmission and wireless RF, microwave, and IR transmission the
base-band signal is modulated with a carrier frequency.
The monochrome or color video signal is a complex
analog waveform consisting of the picture information
(intensity and color) and synchronizing timing pulses. The
waveform was defined in specified by the SMPTE. The full
specifications are contained in standards RS-170, RS-170A,
and RS-170RGB.
6.2.1 Video Picture Signal
For a monochrome camera the picture information
is contained in the single amplitude modulated (AM)
intensity waveform. The video signal amplitude for full
monochrome and color is 1 volt peak to peak. For a color
camera the information is contained in three color waveforms containing the red, green, and blue color contents
of the scene. The three colors can faithfully reproduce
the color picture. The color signal from the camera sensor can be modified in two different forms: (1) composite
video, (2) Y (intensity), C (color), and (3) red (R), green
(G), blue (B). The monochrome and color video signals
are described in Chapter 5.
6.2.2 Video Synchronization Signal
The video synchronization signals consist of vertical field
and frame timing pulses, and horizontal line timing pulses.
The NTSC standard field and frame timing pulses occur
at 1/60 second. and 1/30 second intervals respectively.
While not generating part of the standard NTSC signal,
command and control data can be added to the signal.
The bits and bytes of digital information are handed during the vertical retrace times between frames and fields.
Camera control (on/off, etc.), lens control (focus, zoom,
iris control), and camera platform control (pan, tilt, presets, etc.) signals are digitally controlled to perform these
functions.
6.2.5 Modulation and Demodulation
To accomplish fiber-optic transmission, the base-band
video signal is converted to an FM signal. For RF transmission the base-band video signal is frequency modulated
with the RF of the carrier and 928 MHz (also 435, 1200,
1700 MHz and others). For microwave transmission the
base-band is modulated with a camera frequency of 2.4
and 5.8 GHz.
6.2.6 Signal Bandwidth
The base-band color video signal for NTSC is 30 Hz–6 MHz
(4 MHz for monochrome), and 25 Hz–7MHz for
CCIR/PAL.
6.3 WIRED VIDEO TRANSMISSION
6.3.1 Coaxial Cable
Coaxial cable is used widely for short to medium distances (several hundred to several thousand feet) because
its electrical characteristics best match those required to
transmit the full-signal bandwidth from the camera to the
monitor. The video signal is composed of slowly varying (low-frequency) and rapidly varying (high-frequency)
components. Most wires of any type can transmit the low
Analog Video, Voice, and Control Signal Transmission
frequencies (20 Hz to a few thousand Hz); practically any
wire can carry a telephone conversation. It takes the special
coaxial-cable configuration to transmit the full spectrum
of frequencies from 20 Hz to 6 MHz without attenuation,
as required for high-quality video pictures and audio.
There are basically two types of coaxial and two types of
twin-axial cable for use in video transmission systems:
1.
2.
3.
4.
75-ohm unbalanced indoor coaxial cable
75-ohm unbalanced outdoor coaxial cable
124-ohm balanced indoor twin-axial cable
124-ohm balanced outdoor twin-axial cable.
The cable construction for the coaxial and twin-axial types
are shown in Figure 6-2. The choice of a particular coaxial cable depends on the environment in which it will be
used and the electrical characteristics required. By far the
most common coaxial cables are the RG59/U and the
RG11/U, having a 75-ohm impedance. For short camerato-monitor distances (a few hundred feet), preassembled
or field-terminated lengths of RG59/U coaxial cable with
BNC connectors at each end are used. The BNC connector is a rugged video and RF connector in common
use for many decades and the connector of choice for all
base-band video connections. Short preassembled lengths
of 5, 10, 25, 50, and 100 feet, with BNC-type connectors
attached, are available. Long cable runs (several hundred
feet and longer) are assembled in the field, made up of
a single length of coaxial cable with a connector at each
end. For most interior video installations, RG59/U (0.25
inch diameter), or RG11/U (0.5 inch diameter), 75-ohm
unbalanced coaxial cable is used. When using the larger
diameter RG11/U cable, a larger UHF-type connector is
used. When a long cable run of several thousand feet or
more is required, particularly between several buildings, or
if electrical interference is present, the balanced 124 - ohm
coaxial cable or fiber-optic cable is used. When the camera
and monitoring equipments are in two different buildings,
and likely at different ground potentials, an unwanted signal may be impressed on the video signal which shows up
as an interference (wide bars on the video screen) and
makes the picture unacceptable. A two-wire balanced or
fiber-optic cable eliminates this condition.
Television camera manufacturers generally specify the
maximum distance between camera and monitor over
which their equipment will operate when interconnected
with a specific type of cable. Table 6-1 summarizes the
transmission properties of coaxial and twin-axial cables
when used to transmit the video signal.
In applications with cameras and monitors separated by several thousand feet, video amplifiers are
required. Located at the camera output and/or somewhere along the coaxial-cable run, they permit increasing
TWINAXIAL
(BALANCED)
COAXIAL
(UNBALANCED)
COPPER SHIELDING
GROUND LEAD
FLEXIBLE
OUTER
JACKET
FOAM
DIELECTRIC
POLYPROPYLENE
COPPER
CENTER
CONDUCTOR
INTERCONNECTING
SCHEMATIC
CAMERA
CABLE IMPEDANCE: 75 ohms
TYPES: RG59/U, RG11/U, RG8/U
FIGURE 6-2
Coaxial-twin-axial cable construction
149
MONITOR
150
CCTV Surveillance
MAXIMUM RECOMMENDED CABLE LENGTH (D )
COAXIAL
TYPE
CABLE WITH AMPLIFIER
CABLE ONLY
FEETS
CONDUCTOR
(GAUGE)
METER
FEETS
METER
NOMINAL
DC RESISTANCE
(ohms/1000 ft)
RG59/U
750
230
3,400
1,035
22 SOLID COPPER
10.5
RG59 MINI
200
61
800
250
20 SOLID COPPER
41.0
RG6/U
1,500
455
4,800
1,465
18 SOLID COPPER
RG11/U
1,800
550
6,500
1,980
14 SOLID COPPER
CAMERA
D
6.5
1.24
MONITOR
NOTE: IMPEDANCE FOR ALL CABLES = 75 ohms
Table 6-1
Coaxial Cable Run Capabilities
the camera-to-monitor distance to 3400 feet for RG59/U
cable and to 6500 feet for RG11/U cable.
The increased use of color television in security applications requires the accurate transmission of the video
signal with minimum distortion by the transmitting cable.
High-quality coaxial-cable, UTP, and fiber-optic installations satisfy these requirements.
While a coaxial cable is the most suitable hard-wire cable
to transmit the video signal, video information transmitted
through coaxial cable over long distances is attenuated
differently depending on its signal frequencies. Figure 6-3
illustrates the attenuation as a function of distance and
frequency as exhibited by standard coaxial cables.
The attenuation of a 10 MHz signal is approximately
three times greater than that of a 1 MHz signal when using
a high-quality RG11/U cable. In video transmission, a
3000-foot cable run would attenuate the 5 MHz part of the
video signal (representing the high-resolution part of the
picture) to approximately one-fourth of its original level
at the camera; a 1 MHz signal would be attenuated to only
half of its original level. At frequencies below 500 kHz, the
attenuation is generally negligible for these distances. This
variation in attenuation as a function of frequency has
an adverse effect on picture resolution and color quality.
The signal deterioration appears on monitors in the form
of less definition and contrast and poor color rendition.
For example, severe high-frequency attenuation of a signal
depicting a white picket fence against a dark background
would cause the pickets to merge into a solid, smearing
mass, resulting in less intelligence in the picture.
The most commonly used standard coaxial is RG59/U,
which also has the highest signal attenuation. For a 6 MHz
bandwidth, the attenuation is approximately 1 dB per 100
feet, representing a signal loss of 11%. A 1000-foot run
would have a 10 dB loss—that is, only 31.6% of the video
signal would reach the monitor end.
In a process called “vidi-plexing,” special CCTV cameras
transmit both the camera power and the video signal over
a single coaxial cable (RG59/U or RG11/U). This singlecable camera reduces installation costs, eliminates power
wiring, and is ideal for hard-to-reach locations, temporary
installations, or camera sites where power is unavailable.
6.3.1.1 Unbalanced Single-Conductor Cable
The most widely used coaxial cable for video security transmission and distribution systems is the unbalanced coaxial
cable, represented by the RG59/U or RG11/U configurations. This cable has a single conductor with a characteristic impedance of 75 ohms, and the video signal is applied
between the center conductor and a coaxial braided or
foil shield (Figure 6-2).
Single-conductor coaxial cables are manufactured with
different impedances, but video transmission uses only the
75-ohm impedance, as specified in EIA standards. Other
cables that may look like the 75-ohm cable have a different
electrical impedance and will not produce an acceptable
Analog Video, Voice, and Control Signal Transmission
151
ATTENUATION
(dB/100 ft)
RG6/U
1.50
RG11/U
RG59/U
16 GAUGE
BALANCED
VIDEO PAIR
1.25
1.00
FOAM RG11/U *
0.75
0.50
0.25
FIBER OPTIC CABLE
0
0
10
20
30
40
50
FREQUENCY
(MHz)
* PREFERRED DIELECTRIC: CELLULAR (FOAM) POLYETHYLENE INDOORS,
SOLID POLYETHYLENE OUTDOORS
FIGURE 6-3
Coaxial cable signal attenuation vs. frequency
television picture when used at a distance of 25 or 50 feet
or more.
The RG59/U and RG11/U cables are available from
many manufacturers in a variety of configurations. The
primary difference in construction is the amount and type
of shielding and the insulator (dielectric) used to isolate
the center conductor from the outer shield. The most
common shields are standard single copper braid, double
braid, or aluminum foil. Aluminum foil–type should not
be used for any CCTV application. It is used only for cable
television. Common dielectrics are foam, solid plastic, and
air, the latter having a spiral insulator to keep the center
conductor from touching the outer braid. The cable is
called unbalanced, because the signal current path travels
in the forward direction from the camera to the monitor
on the center conductor and from the monitor back to
the camera again on the shield, which produces a voltage
difference (potential) across the outer shield. This current
(and voltage) has the effect of unbalancing the electrical
circuit.
For short cable runs (a few hundred feet), the deleterious effects of the coaxial cable—such as signal attenuation, hum bars on the picture, deterioration of image
resolution, and contrast—are not observed. However, as
the distance between the camera and monitor increases
to 1000–3000 feet, all these effects come into play. In
particular, high-frequency attenuation sometimes requires
equalizing equipment in order to restore resolution and
contrast.
Video coaxial cables are designed to transmit maximum signal power from the camera output impedance
(75 ohms) with a minimum signal loss. If the cable characteristic impedance is not 75 ohms, excessive signal loss
and signal reflection from the receiving end will occur and
cause a deteriorated picture.
The cable impedance is determined by the conductor
and shield resistance of the core dielectric material, shield
construction, conductor diameter, and distance between
the conductor and the shield. As a guide, resistance of
the center conductor for an RG59/U cable should be
approximately 15 ohms per 1000 feet, and for an RG11/U
cable, approximately 2.6 ohms per 1000 feet. Table 6-2
summarizes some of the characteristics of the RG59/U
and RG11/U coaxial cables.
6.3.1.2 Connectors
Coaxial cables are terminated with several types of connectors: the PL-259, used with the RG11/U cable, and
the BNC, used with the RG59/U cable. The F-type is an
RF connector used in cable television systems. Figure 6-4
illustrates these connectors.
The BNC has become the connector of choice in the
video industry because it provides a reliable connection
152
CCTV Surveillance
CABLE
TYPE
ATTENUATION (dB) @ 5–10 MHZ
100 ft
200 ft
300 ft
400 ft
500 ft
1000 ft
1500 ft
2000 ft
RG59/U
1.0
2.0
3.0
4.0
5.0
10.0
15.0
20.0
RG59 MINI
1.3
2.6
3.9
5.2
6.5
13.0
19.5
26.0
RG6/U
.8
1.6
2.4
3.2
4.0
8.0
12.0
16.0
RG11/U
.51
1.02
1.53
2.04
2.55
5.1
2422/UL1384 *
3.96
7.9
11.9
18.8
19.8
39.6
59.4
79.2
2546 *
1.82
3.6
5.5
7.3
9.1
18.2
27.3
36.4
RG179B/U
2.0
4.0
6.0
8.0
10.0
20.0
30.0
40.0
SIAMESE: RG59
(2) #22AWG
1.0
2.0
3.0
4.0
5.0
10.0
15.0
20.0
7.66
10.2
* MOGAMI
NOTE: IMPEDANCE FOR ALL CABLES = 75 ohms
dB LOSS
% SIGNAL REMAINING
Table 6-2
RCA
1
2
3
4.5
6
8
10.5
90
80
70
60
50
40
30
14
20
20
10
Coaxial Cable Attenuation vs. Length
BNC
UHF
with minimum signal loss, has a fast and positive twist-on
action, and has a small size, so that many connectors can
be installed on a chassis when required. There are essentially three types of BNC connectors available: (1) solder,
(2) crimp-on, and (3) screw-on.
The most durable and reliable connectors are the solder and crimp-on. They are used when the connector
is installed at the point of manufacture or in a suitably
equipped electrical shop. The crimp-on and screw-on types
are the most commonly used in the field, during installation and repair of a system. Either type can be successfully
assembled with few tools in most locations. The crimp-on
type uses a sleeve, which is attached to the cable end after
the braid and insulation have been properly cut back; it is
crimped onto the outer braid and the center conductor
with a special crimping plier. When properly installed, this
cable termination provides a reliable connection.
To assemble the screw-on type, the braid and insulation
are cut back and the connector slid over the end of the
cable and then screwed on. This too is a fairly reliable
type of connection, but it is not as durable as the crimp-on
type, since it can be inadvertently unscrewed from the end
of the cable. The screw-on termination is less reliable if
the cable must be taken on or off many times.
6.3.1.3 Amplifiers
SMA
F
SIAMESE
POWER/BNC
FIGURE 6-4 RCA, BNC, F, SMA, UHF and siamese cable
connectors
When the distance between the camera and the monitor
exceeds the recommended length for the RG59/U and
RG11/U cables, it is necessary to insert a video amplifier to
boost the video signal level. The video amplifier is inserted
at the camera location or somewhere along the coaxial
Analog Video, Voice, and Control Signal Transmission
cable run between the camera and the monitor location
(Figure 6-5).
The disadvantage of locating the video amplifier somewhere along the coaxial cable is that since the amplifier
requires a source of AC (or DC) power, the power source
must be available at its location. Table 6-1 compares the
cable-length runs with and without a video amplifier. Note
that the distance transmitted can be increased more than
fourfold with one of these amplifiers.
When the output from the camera must be distributed
to various monitors or separate buildings and locations,
a distribution amplifier is used (see Figure 6-5). This
amplifier transmits and distributes monochrome and color
video signals to multiple locations. In a quad unit, a
single video input to the amplifier results in four identical, isolated video outputs capable of driving four 75-ohm
RG59/U or RG11/U cables. The distribution amplifier is
in effect a power amplifier, boosting the power from the
single camera output so that multiple 75-ohm loads can be
driven. A potential problem with an unbalanced coaxial
cable is that the video signal is applied across the single
inner conductor and the outer shield, thereby impressing
a small voltage (hum voltage) on the signal. This hum
voltage can be eliminated by using an isolation amplifier,
a balanced coaxial cable, or fiber optics.
6.3.2 Balanced Two-Conductor Twin-axial
Cable Transmission
Balanced twin-axial cables are less familiar to the CCTV
industry than the unbalanced cables. They have a pair of
inner conductors surrounded by insulation, a coaxial-type
shield, and an outer insulating protective sheath has a
characteristic impedance of 124 ohms. They have been
used for many years by telephone industry for transmitting
video information and other high-frequency data. These
cables have an outside diameter (typically 0.5 inch) and
their cost, weight, and volume are higher than those of an
unbalanced cable. Since the polarity on balanced cables
must be maintained, the connector types are usually polarized (keyed). Figure 6-6 shows the construction and configuration of a balanced twin-axial cable system.
The primary purposes for using balanced cable are
to increase transmission range and to eliminate the picture degradation found in some unbalanced applications.
Unwanted hum bars (dark bars on the television picture)
SCENE
EXTENDED
RANGE
CAMERA
COAX
VIDEO
AMPLIFIER
MONITOR
COAX
DVR / VCR
*
DISTRIBUTION
TO MULTIPLE
RECEIVING
EQUIPMENT
SCENE
MONITOR
DVR / VCR
*
CAMERA
COAX
MONITOR
DISTRIBUTION
AMPLIFIER
SWITCHER
COAX
MONITOR
DVR / VCR
*
MONITOR
* EQUIPMENT IN THE
SAME OR MULTIPLE
LOCATIONS
MONITOR
DVR / VCR
*
MONITOR
DVR / VCR
PRINTER
FIGURE 6-5
153
Video amplifier to extend range and/or distribute signal
154
CCTV Surveillance
METALLIC
BRAID
DUAL
COPPER
CONDUCTORS
OUTER
INSULATED
JACKET
IMPEDANCE: 124 ohms
FOAM
INSULATION
INTERCONNECTING
SCHEMATIC
SCENE
CAMERA
FIGURE 6-6
BALANCED
TRANSMITTING
TRANSFORMER
124 ohm
BALANCED
CABLE
BALANCED
RECEIVING
TRANSFORMER
MONITOR
Balanced twin-axial cable construction and interconnection
are introduced in unbalanced coaxial transmission systems
when there is a difference in voltage between the two ends
of the coaxial cable (see Section 6.3.1.1). This can often
occur when two ends of a long cable run are terminated
in different buildings, or when electrical power is derived
from different power sources—in different buildings or
even within the same building.
Since the signal path and the hum current path through
the shield of an unbalanced cable are common and result
in the hum problem, a logical solution is to provide a
separate path for each. This is accomplished by applying
the signal between each center conductor of two parallel unbalanced cables (Figure 6-6). The shields of the
two cables carry the ground currents while the two conductors carry the transmitted signal. This technique has
been used for many years in the communications industry to reduce or eliminate hum. Since the transmitted
video signal travels on the inner conductors, any noise or
induced AC hum is applied equally to each conductor. At
the termination of the run the disturbances are cancelled
while the signal is directed to the load unattenuated. This
technique in effect removes the unwanted hum and noise
signals.
While the balanced transmission line offers many advantages over the unbalanced line, it has not been in
widespread use in the video security industry. The primary reason is the need for transformers at the camerasending and monitor-receiving ends, as well as the need
for two-conductor twin-axial cable. All three hardware
items require additional cost as compared with the unbal-
anced single-conductor coaxial cable. The use of UTP
transmission has become a popular replacement for the
coaxial cable (Section 6.3.3), or fiber optics, described in
Section 6.3.4.
6.3.2.1 Indoor Cable
Indoor coaxial cable is small in diameter (0.25 inch), uses
a braided shield, and is much more flexible than outdoor cable. To maintain the correct electrical impedance
this smaller outside diameter cable requires proportionally smaller inner conductors. This decrease in diameter
of the cable conductor causes a corresponding increase
in the cable signal attenuation and therefore the RG59/U
indoor cable cannot be used over long distances. The
impedance of any coaxial cable is directly related to the
spacing between the inner conductor and the shield;
any change in this spacing caused by tight bends, kinking, indentations, or other factors will change the cable
impedance resulting in picture degradation. Since indoor
cabling and connectors need no protection from water,
solder, crimp-on, or screw-on connectors can be used.
6.3.2.2 Outdoor Cable
Outdoor video transmission applications put additional
physical requirements on the coaxial cable. Environmental
factors such as precipitation, temperature changes, humidity, and corrosion are present for both above-ground
Analog Video, Voice, and Control Signal Transmission
and buried installations. Other above-ground considerations include: wind loading, rodent damage, and electrical
storm interference. For direct burial applications, ground
shifts, damage due to water, and rodent damage are potential problems. Outdoor coaxial cabling is 1/2 inch in diameter or larger, since the insulation qualities in outside
protective sheathing must be superior to those of indoor
cables and their electrical qualities are better than indoor
RG59/U cables. Outdoor cables have approximately 16
gauge inner-conductor diameters resulting in much less
signal loss than the smaller, approximately 18 gauge center conductor indoor RG59/U cables. Outdoor cables are
designed and constructed to take much more physical
abuse than the indoor RG59/U cable. Outdoor cables are
not very flexible and care must be taken with extremely
sharp bends. As a rule of thumb, outdoor cabling should
always be used for cable runs of more than 1000 feet,
regardless of the environment.
Outdoor video cable may be buried, run along the
ground, or suspended on utility poles. The exact method
should be determined by the length of the cable run, the
economics of the installation, and the particular environment. Environment is an important consideration.
In locations with severe weather, electrical storms or
high winds, it is prudent to locate the coaxial cable underground, either direct-buried or enclosed in a conduit. This
method isolates the cable from the severe environment,
improving the life of the cable and reducing signal loss. In
locations having rodent or ground-shift problems, enclosing the cable in a separate conduit will protect it. For short
cable runs between buildings (less than 600–700 feet) and
where the conduit is waterproof, indoor RG59/U cable is
suitable.
There are about 25 different types of RG59/U and about
10 different types of RG11/U cable but only a few are suitable for video systems. For optimum performance, choose
a cable that has 95% copper shield or more and a copper
or copper-clad center conductor. The copper-clad center
conductor has a core of steel and copper cladding, has
higher tensile strength, and is more suitable for pulling
through conduit over long cable runs. While cable with
65% copper shield is available, 95% shielding or more
should be used to reduce and prevent outside electromagnetic interference (EMI) signals from penetrating the
shield, causing spurious noise on the video signal. A coaxial cable with 95% shield and a copper center conductor
will have a loop resistance of approximately 16–57 ohms
per 1000 feet.
155
near other electrical power distribution equipment or
machinery producing high electromagnetic fields.
In outdoor applications, in addition to the above the
adverse environmental conditions caused by lightning
storms or other high-voltage noise generators, such as
transformers on power lines, electrical substations, automobile/truck electrical noise, or other EMI must be
considered.
In the case of EMI, a facility site survey should be made
of the electromagnetic radiation present in any electrically
noisy power distribution equipment. The cables should
then be routed away from such equipment so that there
is no interference with the television signal.
When a site survey indicates that the coaxial cable must
run through an area containing large electrical interfering signals (EMI) caused by large machinery, highvoltage power lines, refrigeration units, microwaves, truck
ignition, radio or television stations, fluorescent lamps,
two-way radios, motor-generator sets, or other sources, a
better shielded cable, such as a twin-axial, tri-axial, UTP,
or fiber optic cable may be the answer. The tri-axial
cable has a center conductor, an insulator, a shield, a
second insulator, a second shield, and the normal outer
polyethylene or other covering to protect it from the environment. The double shielding significantly reduces the
amount of outside EMI radiation that gets to the center
conductor.
The number of horizontal bars on the monitor can
indicate where the source of the problem is. If the monitor
has six dark bars, multiplying 6 by 60 equals 360, which
is close to a 400-cycle frequency. This interference could
be caused by an auxiliary motor-generator set often found
in large factory machines operating at this frequency. To
correct the problem, the cable could be rerouted away
from the noise source, replaced with a balanced twin-axial
or tri-axial cable, UTP, or for 100% elimination of the
problem, upgraded to fiber-optic cable.
If lighting and electrical storms are anticipated and signal loss is unacceptable, outdoor cables must be buried
underground and proper high voltage–suppression circuitry must be installed at each end of the cable run and
on the input power to the television equipment.
In new installations with long cable runs (several thousand feet to several miles) or where different ground voltages exist, a fiber-optic link is the better solution, although
balanced systems and isolation amplifiers can often solve
the problem.
6.3.2.4 Grounding Problems
6.3.2.3 Electrical Interference
For indoor applications, interference and noise can result
from the following problems: (1) different ground potentials at the ends of the coaxial cable at different video
equipment locations in a building, and (2) coaxial cable
Ground loops are by far the most troublesome and noticeable video cabling problem (Figure 6-7). Ground loops
are most easily detected before connecting the cables, by
measuring the electrical voltage between the coaxial-cable
shield and the chassis to which it is being connected. If
the voltage difference is a few volts or more, there is a
156
CCTV Surveillance
HUM BAR
PICTURE
TEARING
HUM BAR
HUM BAR
HUM BAR
FIGURE 6-7
Hum bars caused by ground loops
potential for a hum problem. As a precaution, it is good
practice to measure the voltage difference before connecting the cable and chassis for systems with a long run or
between any two electrical supplies to prevent any damage
to the equipment.
Many large multiple-camera systems have some distortion in the video picture caused by random or periodic
noise or if more severe, by hum bars. The hum bar appears
as a horizontal distortion across the monitor at two locations: one-third and two-thirds of the way down the picture. If the camera is synchronized or power-line-locked,
the bar will be stationary on the screen. If the camera is
not line-locked, the distortion or bar will continuously roll
slowly through the picture. Sometimes the hum bars are
accompanied by sharp tearing regions across the monitor or erratic horizontal pulling at the edge of the screen
(Figure 6-7). This is caused by the effect of the high voltages on the horizontal synchronization signal. Other symptoms include uncontrolled vertical rolling of the scene on
the screen when there are very high voltages present in
the ground loop.
Interference caused by external sources or voltage differences can often be predicted prior to installation. The
hum bar and potential difference between two electrical systems usually cannot be determined until the actual
installation. The system designer should try to anticipate
the problem and, along with the user, be prepared to
devote additional equipment and time to solve it. The
problem is not related to equipment at the camera or
monitor end or to the cable installed; it is strictly an effect
of the particular environment encountered, be it EMI
interference or difference in potential between the main
power sources at each location. The grounding problem
can occur at any remote location, and it can be eliminated
inexpensively with the installation of an isolation amplifier. Another solution, described in Section 6.3.4, is the
use of fiber-optic transmission means, which eliminates
electrical connections entirely.
One totally unacceptable solution is the removal of the
third wire on a three-pronged electrical plug, which is used
to ground the equipment chassis to earth ground. Not
only is such removal a violation of local electrical codes
and Underwriters Laboratory (UL) recommendations, it
is a hazardous procedure. If the earth ground is removed
from the chassis, a voltage can appear on the camera,
monitor, or other equipment chassis, producing a “hot”
chassis that, if touched, can shock any person with 60–70
volts.
When video cables bridge two power distribution
systems, ground loops occur. Consider the situation
(Figure 6-8) in which the CCTV camera receives AC power
from power source A, while some distance away or in a
different building the CCTV monitor receives power from
distribution system B.
The camera chassis is at 0 volts (connected to electrical ground) with reference to its AC power input A. The
monitor chassis is also at 0 volts with respect to its AC
distribution system B. However, the level of the electrical ground in one distribution system may be higher (or
lower) than that of the ground in the other system; hence
a voltage potential can exist between the two chassis. When
a video cable is connected between the two distribution
system grounds, the cable shield connects the two chassis and an alternating current flows in the shield between
the units. This extraneous voltage (causing a ground-loop
current to flow) produces the unwanted hum bars in the
video image on the monitor.
The second way in which hum bars can be produced
on a television monitor is when two equipment chassis
are mechanically connected, such as when a camera is
mounted on a pan/tilt unit. If the camera receives power
from one distribution system and the chassis of the pan/tilt
unit is grounded to another system with a different level,
a ground loop and hum bars may result. The size and
extent of the horizontal bars depends on the severity of
the ground potential difference.
6.3.2.5 Aluminum Cable
Although coaxial cable with aluminum shielding provides
100% shielding, it should only be used for RF cable television (CATV) and master television (MATV) signals used
for home video cable reception. This aluminum-shield
type should never be used for CCTV for two reasons: (1) it
Analog Video, Voice, and Control Signal Transmission
157
LOCATION B
LOCATION A
CAMERA
(PAN/ TILT, ETC.)
POWER SOURCE A
MONITOR (OR SWITCHER,
VCR, PRINTER, ETC.)
POWER SOURCE B
COAXIAL CABLE
CONTROL
WIRE
GROUND
COAXIAL
SHIELD
GROUND
117 VAC POWER
FROM SYSTEM A
LOCATION B
GROUND
0 VOLTS
LOCATION A
GROUND
117 VAC POWER
FROM SYSTEM B
COAXIAL
SHIELD
GROUND
*
VOLTAGE DIFFERENCE
*
NOTE: THE VOLTAGE DIFFERENCE BETWEEN GROUND A AND B
CAN BE 5–30 VOLTS, CAUSING CURRENT TO FLOW IN THE
CABLE SHIELD, HUM BARS AND FAULTY OPERATION
FIGURE 6-8
Two source AC power distribution system
has higher resistance, and (2) it distorts horizontal synchronization pulses.
The added resistance—approximately seven times more
than that of a 95% copper or copper-clad shield—
increases the video cable loop resistance, causing a reduction in the video signal transmitted along the cable.
The higher loop resistance means a smaller video signal
reaches the monitoring site, producing less contrast and
an inferior picture. Always use a good-grade 95% copper
braid RG59/U cable to transmit the video signal up to
1000 feet and an RG11/U to transmit up to 2000 feet.
Distortion of the horizontal synchronization pulse causes
picture tearing on the monitor, depicting straight-edged
objects with ragged edges.
6.3.2.6 Plenum Cable
Another category of coaxial cable is designed to be used
in the plenum space in large buildings. This plenum cable
has a flame-resistant exterior covering and very low smoke
emission. The cable can be used in air-duct air-conditioning
returns and does not require a metal conduit for added
protection. The cable, designated as “plenum rated,” is
approved by the National Electrical Code and UL.
6.3.3 Two-Wire Cable Unshielded Twisted Pair
(UTP) Transmission
It is convenient, inexpensive, and simple to transmit the
video signal over an existing two-wire system. A standard, twisted pair, two-wire telephone, intercom, or other
electrical system with an appropriate UTP transmitter and
receiver has the capability to transmit all of the highfrequency information required for an excellent resolution monochrome or color picture. The UTP is a CAT-5,
CAT-5e, or CAT-3 cable. The higher the level of CAT
(5e) cable the greater the distance. Either a passive (no
power required) or an active (12VDC, longer distanced)
transmitter/receiver pair can be used. The passive system
uses a small transmitter and receiver-one at each end of
the pair of wires—and transmits the picture at distances
of a few hundred feet to 3000 feet. The active powered
system (12VDC) can transmit the video image 8000 feet
for monochrome and 5000 feet for color. Picture resolution can be equivalent to that obtained with a coaxial cable
158
CCTV Surveillance
system. The two-wire pair must have a continuous conductive path from the camera to the monitor location. Highfrequency emphasis in the transmitter and receiver compensate for any attenuation of the high frequencies. The
balanced UTP configuration makes the cable immune to
most external electrical interference and in many environments the UTP cable can be located in the same conduit
with other data cables.
The UTP system must have a conductive (resistive copper) path for the two wires. The signal path cannot have
electrical switching circuits between the camera and the
monitor location; however, mechanical splices and connectors are permissible.
The components for the two-wire system can cost more
than equivalent coaxial cable since an additional transmitter and receiver are required. However, this cost may be
small compared with the cost of installing a new coaxial
cable from the camera to the monitor location. Figure
6-9 illustrates the block diagram and connections for the
UTP, and active transmitter and receiver pair.
6.3.3.1 Balanced 2-Wire Attributes
The UTP provides a technology that can significantly
reduce external electrical radiation from causing noise in
the video signal. It also eliminates ground loops present in
unbalanced coaxial transmission since isolation designed
into the UTP transmitters and receivers.
6.3.3.2 UTP Technology
The UTP technology is based on the concept that any
external electrical interference affects each of the two
conductors identically so that the external disturbance is
canceled and has no effect on the video signal. The transmitter unit converts the camera signal 75 ohm impedance
to match the UTP 100 ohm CAT-5e impedance and provides the frequency compensation required. The receiver
unit amplifies and reconstructs the signal and transmits it
over a short distance to the television monitor via 75 ohm
coaxial cable. Most active transmitters and receivers have
3 to 5 position dip switches which are set depending
on the cable length to optimize the video signal waveform. Both the transmitter and the receiver are powered
by either 12VDC or self-powered from the camera or
monitor.
The UTP system can be operated with CAT-3, 5, 5e, and
6 as defined in the TIA/EIA 568b.2 standard. CAT-5e is
now used for most new video installations and supercedes
the extensively installed CAT-5 cable.
UNSHIELDED
TWISTED PAIR
(UTP)
CAT-3, 5, 5e
COAX
75 ohm
VIDEO
RECEIVER
VIDEO
TRANSMITTER
COAX
75 ohm
100 ohm
CAMERA
MONITOR
(A) ACTIVE TRANSMITTER
FIGURE 6-9
Two wire UTP video transmission system
(B) ACTIVE RECEIVER
Analog Video, Voice, and Control Signal Transmission
6.3.3.3 UTP Implementation with Video, Audio,
and Control Signals
The UTP transmitter is located between the camera
and the CAT-5e UTP cable input to transmit video,
audio, alarms, and control signals. The receiver is located
between the monitor end of the CAT-5e UTP cable and
the monitor, recorder, or control console. UTP transmitters are small enough to be part of the camera electronics
or can be powered by the camera, audio, and/or control
electronics (Figure 6-10).
6.3.3.4 Slow-Scan Transmission
The wireless transmission systems described in the previous sections all result in real-time video transmission. A
scheme for transmitting the television picture over large
distances, even anywhere in the world, uses slow-scan television transmission (Figure 6-11).
This non-real-time technique involves storing one television picture frame (snapshot) and sending it slowly over a
telephone or other audio-grade network anywhere within
a country or to another country. The received picture is
reconstructed at the remote receiver to produce a continuously displayed television snapshot. Each snapshot takes
anywhere from several to 72 seconds to transmit, with
a resulting picture having from low to high resolution,
depending on the speed of transmission. A TL effect is
achieved, and every scene frame is transmitted spaced
from several to 72 seconds apart.
Through this operation, specific frames are serially captured, sent down the telephone line, and reconstructed by
159
the slow-scan system. Once the receiver has stored the digital picture information, if the transmitter is turned off or
the video scene image does not change, the receiver continues to display the last video frame continuously (30 fps)
as a still image. The image stored in the receiver or transmitter changes when the system is commanded, manually
or automatically, to take a new snapshot. Figure 6-12 illustrates the salient difference between real-time and nonreal-time television transmission.
Implementation. Figure 6-12 shows the relationship of
non-real-time or slow-scan television transmission. At
the camera site the first frame starts at time zero
(Figure 6-12a), the second frame at 1/30th of a second,
and the third frame at 2/30th of a second (the same as for
real-time). Before these frames are transmitted over the
audio-grade transmission link, the signal is processed at
the camera site in a transmitter processor. The processor
captures Frame 1 from the camera, that is, it memorizes
(digitizes) the CCTV picture. The processor then slowly
(at 2 seconds per frame, as shown in Figure 6-12b) transmits the video frame, element by element, line by line,
until the receiver processor located at the monitor site has
accepted all 525 lines in that frame.
The significant difference between real-time and slowscan transmission is the time it takes to transmit the
picture. In the real-time case, it is 1/30th of a second,
the real-time of the frame. In the case of the slow-scan
(Figure 6-12b), it may take 2, 4, 8, 32, up to 72 seconds to
transmit that single frame to the monitor site. Figure 6-13
is a block diagram of a simplex (one-way) slow-scan
system.
CAMERA
COAX
TRANSMITTER:
• VIDEO
• AUDIO
• ALARM
RECEIVER:
• AUDIO
• CONTROL
DEDICATED TWO
WIRE SYSTEM
•
•
•
SHIELDED 2 WIRE
TWISTED PAIR (UTP)
TELEPHONE
RECEIVER:
• VIDEO
• AUDIO
• ALARM
VIDEO
COAX
TRANSMITTER:
• AUDIO
• CONTROL
MICROPHONE
MICROPHONE
SPEAKER
SPEAKER
COMMAND
FUNCTIONS
ALARM
INPUTS
FIGURE 6-10
Real-time transmission system with video, audio, and controls
ALARM
OUTPUT
DEVICES
160
CCTV Surveillance
LOCATION 1
LOCATION 2
SLOW-SCAN
TRANSCEIVER
SLOW-SCAN
TRANSCEIVER
CAMERA
CAMERA
DUPLEX (TWO-WAY) NETWORK
3000 Hz BANDWIDTH
MONITOR
MONITOR
(A) PICTURE RESOLUTION: 128 × 64 (H × V)
FULL PICTURE TRANSMIT TIME: 2.6 SEC
(B) PICTURE RESOLUTION: 256 × 128 (H × V)
FULL PICTURE TRANSMIT TIME: 8.0 SEC
(C) PICTURE RESOLUTION: 512 × 256 (H × V)
FULL PICTURE TRANSMIT TIME: 31 SEC
NOTE: PICTURE TRANSMIT UPDATE TIME DEPENDS ON MOTION IN PICTURE. MONOCHROME PICTURE.
FIGURE 6-11
Slow-scan video transmission and transmitted pictures over telephone lines
(A) REAL-TIME (30 FPS)
(B) SLOW-SCAN
FRAME 1
T=0
FRAME 1
T=0
FRAME 2
T = 1/30 SEC
FRAME 2
T = 2 SEC
G
FRAME 3
T = 2/30 SEC
FRAME 3
T = 4 SEC
G
CONTINUOUS VIEWING
OF MOVING VEHICLE
FIGURE 6-12
Real-time video transmission vs. non-real-time (slow-scan)
SNAP SHOTS OF
PERSON WALKING
Analog Video, Voice, and Control Signal Transmission
NUMBER OF
GRAY SCALE
LEVELS
SCAN
RATE
161
DIGITAL
TELEPHONE
DIALER
LENS
CAMERA
ANALOG/
DIGITAL
CONVERTER
VIDEO
FRAME
GRABBER
AUDIO
DRIVE
AMPLIFIER
DIGITAL
VIDEO
COMPRESSION
DIGITAL
TO ANALOG
CONVERTER
ALARM
INPUT
ANY DUPLEX AUDIO BANDWITH
COMMUNICATIONS LINK
AUDIO
FREQUENCY
DEMODULATOR
FIGURE 6-13
ANALOG
TO DIGITAL
CONVERTER
DIGITAL
VIDEO
DECOMPRESSION
TEMPORARY
MEMORY
DIGITAL
TO ANALOG
CONVERTER
MONITOR
Slow-scan system block diagram—simplex (one-way)
To increase transmission speed, complex compression
and modulation algorithms are used so that only the
changing parts of a scene (i.e. the movements) are transmitted. Another technique first transmits areas of high
scene activity with high resolution and then areas of lower
priority with lower resolution. These techniques increase
the transmission of the intelligence in the scene. By
increasing transmission time of the frame from 1/30th
of a second to several seconds, the choice of cable or
transmission path changes significantly. For slow-scan it
is possible to send the full video image on a twisted-pair
or telephone line or any communications channel having a bandwidth equivalent to audio frequencies, that
is, up to only 3000 Hz (instead of a bandwidth up to
4.2 MHz, as needed in real-time transmission). So all existing satellite links, mobile telephones, and other connections can be used. A variation of this equipment for an
alarm application can store multiple frames at the transmitting site, so if the information to be transmitted is an
alarm, this alarm video image can be stored for a few seconds (every 1/30 second) and then slowly transmitted to
the remote monitoring site frame by frame, thereby transmitting all of the alarm frames. Figure 6-14 shows the interconnecting diagram and controls for a typical slow-scan
system.
Resolution, Scene Activity vs. Transmit Time. Transmit
time per frame is determined by video picture resolution
and activity (motion) in the scene. The larger the number
of gray-scale levels and number of colors transmitted, the
longer the transmit time per frame of video. If only a
few gray-scale levels are transmitted (photocopy quality),
or a limited number of colors and a small amount of
motion in the scene are present, then there is less picture
information to transmit and short (1–8 seconds) transmit
times result. High gray-scale levels (256 levels), full color,
and motion require more information and longer transmission times. Slow-scan transmission is a compromise
between resolution and scene activity and required scene
update time.
6.3.4 Fiber-Optic Transmission
6.3.4.1 Background
One of the most significant advances in communications
and signal transmission has been the innovation of fiber
optics. However, the concept of transmitting video signals
over fiber optics is not new. The transmission of optical
signals in fibers was investigated in the 1920s and 1930s
but it was not until the 1950s that Kapany invented the
162
CCTV Surveillance
AC OR DC
POWER
AC OR DC
POWER
1
CAMERA
MONITOR
TRANSMISSION
PATH:
ANY AUDIO
GRADE
CHANNEL
2
CAMERA
3
CAMERA
1
CAMERA
2
4
CAMERA
CAMERA
MONITOR *
TALK /
VIEW
AUTO
ANSWER /
TRANSMIT
TALK /
VIEW
AUTO
ANSWER /
TRANSMIT
3
CAMERA
4
CAMERA
KEYPAD CONTROLS: • TRANSMIT TIME (1, 2, 4, 8, 16, 32, 64)
• SHADES OF GRAY (16, 32, 64, 128)
• PIXEL RESOLUTION (32, 64, 128, 256, 512)
• QUAD OR FULL SCREEN
• CAMERA SELECTOR (4 CAMERAS)
• EXTERNAL DEVICES (VCR, HORN, LIGHTS, ETC.)
* DUPLEX SYSTEM (2 WAY) USE MONITOR AND CAMERA AT EACH END
FIGURE 6-14
Slow-scan interconnecting diagram and controls
practical glass-coated (clad) glass fiber and coined the
term fiber optics.
Clad fiber was actively investigated in the 1960s by
K.C. Kao and G.A. Hockham, researchers at Standard
Telecommunications Laboratories in England, who proposed that this type of waveguide could form the basis of a
new transmission system. In 1967 attenuations through the
fiber was more than 1000 dB per kilometer (0.001% transmission/km) which were impractical for transmission purposes, and researchers focused on reducing these losses.
Figure 6-16 shows a comparison of fiber-optic transmission
vs. other electrical transmission means. In 1970, investigators Kapron, Keck, and Maurer at Corning Glass Works
announced a reduction of losses to less than 20 dB per
kilometer in fibers hundreds of meters long. In 1972 Corning announced a reduction of 4 dB per one kilometer of
cable, and in 1973 Corning broke this record with a 2 dB
per kilometer cable. This low-loss achievement made a revolution in transmission of wide-bandwidth, long-distance
communications inevitable. In the early 1970s, manufacturers began making glass fibers that were sufficiently lowloss to transmit light signals over practical distances of
hundreds or a few thousand feet.
Broadband fiber-optic components are much more
expensive than cable. They should be used when there
is a definite need for them. Note also that video signals
must be digitized to avoid nonlinear transmitter/receiver
effects.
Why use fiber-optic transmission when coaxial cables
can provide adequate video signal transmission? Today’s
high-performance video systems require greater reliability
and more “throughput,” that is, getting more signals from
the camera end to the monitor end, over greater distances,
and in harsher environments. The fiber-optic transmission system preserves the quality of the video signal and
provides a high level of security.
The information-carrying capacity of a transmission
line, whether electrical or optical, increases as the carrier
frequency increases. The carrier for fiber-optic signals
is light, which has frequencies several orders of magnitude (1000 times) greater than radio frequencies, and
the higher the carrier frequency the larger the bandwidth
that can be modulated onto the cable. Some transmitters
and receivers permit multiplexing multiple television signals, control signals, and duplex audio onto the same fiber
optic because of its wide bandwidth.
The clarity of the picture transmitted using fiber optics
is now limited only by the camera, environment, and
Analog Video, Voice, and Control Signal Transmission
OPTICAL CONNECTORS
(SMA, SFR, LFR)
TRANSMITTER
COAXIAL
CABLE
LENS
CAMERA
BNC
ELECTRICAL TO
OPTICAL
SIGNAL
BNC
CONVERTER
GND
163
OPTICAL CONNECTORS
(SMA, SFR, LFR)
RECEIVER
FIBER OPTIC
CABLE
12 VDC
POWER
CONVERTER
117 VAC TO
12 VDC
117 VAC
POWER
COAXIAL
OPTICAL TO
CABLE
ELECTRICAL
SIGNAL
BNC
CONVERTER
GND
12 VDC
POWER
CONVERTER
117 VAC TO
12 VDC
BNC
MONITOR
117 VAC
POWER
FIBER OPTIC SYSTEM COMPONENTS:
• TRANSMITTER
• RECEIVER
• FIBER OPTIC CABLE
• POWER CONVERTERS (2)
FIGURE 6-15
Fiber optic transmission system
monitoring equipment. Fiber-optic systems can transmit
signals from a camera to a monitor over great distances—
typically several miles—with virtually no distortion or loss
in picture resolution or detail. Figure 6-15 shows the block
diagram of the hardware required for a fiber-optic system.
The system uses an electrical-to-optical signal converter/transmitter, a fiber cable for sending the light
signal from the camera to the monitor, and a light-toelectrical signal receiver/converter to transform the signal
back to a base-band video signal required by the monitor. At both camera and monitor ends standard RG59/U
coaxial cable or UTP wire is used to connect the camera
and monitor to the system.
A glass fiber optic–based video link offers distinct
advantages over copper-wire or coaxial-cable transmission
means:
• The system transmits information with greater fidelity
and clarity over longer distances.
• The fiber is totally immune to all types of electrical
interference—EMI or lightning—and will not conduct
electricity. It can touch high-voltage electrical equipment or power lines without a problem.
• The fiber being nonconductive does not create any
ground loops.
• The fiber can be serviced while the transmitting or
receiving equipment is still energized since no electrical
power is involved.
• The fiber can be used where electrical codes and common sense prohibit the use of copper wires.
• The cable will not corrode and the glass fiber is unaffected by salt and most chemicals. The direct-burial type
of cable can be laid in most kinds of soil or exposed to
most corrosive atmospheres inside chemical plants or
outdoors.
• Since there is no electrical connection of any type, the
fiber poses no fire hazard to any equipment or facility
in even the most flammable atmosphere.
• The fiber is virtually unaffected by atmospheric conditions, so the cable can be mounted aboveground and
on telephone poles. When properly applied, the cable
is stronger than standard electrical wire or coaxial cable
and will therefore withstand far more stress from wind
and ice loading.
• Single or multiple fiber-optic cables are much smaller
and lighter than a coaxial cable. It is easier to handle
and install, and uses less conduit or duct space. A single
optical cable weighs 8 pounds per 3300 feet and has an
overall diameter of 0.156 inches. A single coaxial cable
weighs 330 pounds per 3300 feet and is approximately
0.25 inches in diameter.
164
CCTV Surveillance
• It transmits the video signal more efficiently (i.e. with
lower attenuation) and since over distances of less than
50 miles it needs no repeater (amplifier), it is more
reliable and easier to maintain.
• It is a more secure transmission medium, since not only
is it hard to tap but an attempted tap is easily detected.
The economics of using a fiber-optic system is complex. Users evaluating fiber optics should consider the
costs beyond those for the components themselves. The
small size, lightweight, and flexibility of fiber optics often
present offsetting cost advantages. The prevention of
unanticipated problems such as those just listed can easily
offset any increased hardware costs of fiber-optic systems.
With such rapid advances, the security system designer
should consider fiber optics the optimum means to transmit high-quality television signals from high-resolution
monochrome or color cameras to a receiver (monitor,
switcher, recorder, printer, and so on) without degradation. This section reviews the attributes of fiber-optic systems, their design requirements, and their applications.
6.3.4.2 Simplified Theory
The fiber-optic system uses a transmitter at the camera and
a receiver at the monitor and the fiber cable in between
(Figure 6-15). The following sections describe these three
components. By far the most critical is the fiber-optic
cable, since it must transmit the video light signal over
a long distance without attenuation distortion (changing
its shape or attenuation at high frequencies). As shown
in Figure 6-15, the signal from the camera is sent to the
transmitter via standard coaxial cable. At the receiver end,
the output from the receiver is likewise sent via standard
wire cable to the monitor or recording system.
The optical transmitter at the camera end converts
(modulates) the electrical video analog signal into a corresponding optical signal. The output from the transmitter
is an optical signal generated by either an LED or an
ILD, emitting IR light. When more than one video signal
is to be transmitted another option is to transmit multiple signals over one fiber using wavelength multiplexing
(Section 6.3.4.7).
The multi-fiber-optic cable consists of multiple glass
fibers, each acting as a waveguide or conduit for one
video optical signal. The glass fibers are enclosed in a protective outer jacket whose construction depends on the
application.
The fiber-optic receiver collects the light from the end
of the fiber-optic cable and converts (demodulates) the
optical signal back into an electrical signal having the same
waveform and characteristics as the original video signal at
the camera and then sends it to the monitor or recorder.
The only variation in this block diagram for a single
camera is the inclusion of a connection, splice, or repeater
that may be required if the cable run is very long (many
miles). The connector physically joins the output end of
one cable to the input end of another cable. The splice
reconnects two fiber ends so as to make them continuous.
The repeater amplifies the light signal to provide a good
signal at the receiver end.
How does the fiber-optic transmission system differ from
the electrical cable systems described in the previous sections? From the block diagram (Figure 6-15) it is apparent that two new hardware components are required: a
transmitter and a receiver. The transmitter provides an
amplitude- or frequency-modulated representation of the
video signal at near-IR wavelengths which the fiber optic
transmits, and at a level sufficient to produce a high-quality
picture at the receiver end. The receiver collects whatever
light energy is available at the output of the fiber-optic
cable and converts it efficiently, with all the information
from the video signal retained, into an electrical signal that
is identical in shape and amplitude to the camera output
signal. As with any of the transmission means, the fiberoptic cable attenuates the video signal. Figure 6-16 shows
the attenuation frequency for current fiber-optic cable as
compared with telephone cable, special high-frequency
cable, coaxial cable, and early fiber-optic cable.
The fiber-optic cable efficiently transmits the modulated
light signal from the camera end over a long distance
to the monitor, while maintaining the signal’s shape and
amplitude. Characteristics of fiber-optic cable are totally
different from those of coaxial cable or two-wire transmission systems.
Before discussing the construction of the fiber-optic
cable, we will briefly describe the transmitting light. In any
optical material, light travels at a velocity (Vm ) characteristic of the material, which is lower than the velocity of light
(C ) in free space of air (Figure 6-17a).
The ratio (fraction) of the velocity in the material compared with that in free space defines the refractive index
(n) of the material:
n=
C
Vm
When light traveling in a medium of a particular refractive
index strikes another material of a lower refractive index,
the light is bent toward the interface of the two materials (Figure 6-17b). If the angle of incidence is increased,
a point is reached where the bent light will travel along
the interface of the two materials. This is known as the
“critical angle” (C ). Light at any angle greater than the
critical angle is totally reflected from the interface and
follows a zigzag transmission path (Figure 6-17b,c). This
zigzag transmission path is exactly what occurs in a fiberoptic cable: the light entering one end of the cable zigzags
through the medium and eventually exits at the far end at
approximately the same angle. As shown in Figure 6-17c,
some incoming light is reflected from the fiber-optic end
and never enters the fiber.
Analog Video, Voice, and Control Signal Transmission
ATTENUATION
(dB/km)
30
SPECIAL HIGH
TELEPHONE,
FREQUENCY
UTP 3, 5e CABLE
CABLE
165
COAXIAL
CABLE
25
EARLY FIBER
OPTIC CABLE:
20 dB/km (1970)
20
(100 TO 1 LOSS)
15
10
IMPROVED CABLE:
4 dB/km (1972)
5
CURRENT CABLE:
2– 4 dB/km
0
0.1
1.0
10
5 MHz
MAX. VIDEO
BANDWIDTH
FIGURE 6-16
100
1000
FREQUENCY
(MHz)
NOTE: 1 KILOMETER (km) = .67 mile
Attenuation vs. frequency for copper and fiber optic cable
In practice, an optical fiber consists of a core, a cladding,
and a protective coating. The core material has a higher
index of refraction than the cladding material and therefore the light, as just described, is confined to the core.
This core material can be plastic or glass, but glass provides a far superior performance (lower attenuation and
greater bandwidth) and therefore is more widespread for
long-distance applications.
One parameter often encountered in the literature is
the numerical aperture (NA) of a fiber optic, a parameter that indicates the angle of acceptance of light into
a fiber—or simply the ease with which the fiber accepts
light. The NA is an important fiber parameter that must
be considered when determining the signal-loss budget of
a fiber-optic system. To visualize the concept, picture a
bottle with a funnel (Figure 6-18). The larger the funnel
angle, the easier it is to pour liquid into the bottle. The
same concept holds for the fiber. The wider the acceptance angle, the higher the NA, the larger the amount of
light that can be funneled into the fiber from the transmitter. The larger or higher an optical fiber NA, the easier
it is to launch light into the fiber, which correlates to
higher coupling efficiency. Since fiber-optic systems are
often coupled to LEDs, which are the light generators at
the transmitter, and since LEDs have a less-concentrated,
diffuse output beam than ILDs, fiber optics with high NAs
allow more collection of the LED output power.
In order for the light from the transmitter to follow
the zigzag path of internally reflected rays, the angles of
reflection must exceed the critical angle. These reflection
angles are associated with “waveguide modes.” Depending
on the size (diameter) of the fiber-optic core, one or more
modes are transmitted down the fiber. The characteristics
and properties of these different cables carrying singlemode and multimode fibers are discussed in the next
section.
Like radio waves, light is electromagnetic energy. The
frequencies of light used in fiber-optic video, voice, and
data transmission are approximately 36 × 1014 , which is
several orders of magnitude higher than the highest radio
waves. Wavelength (the reciprocal of frequency) is a more
common way of describing light waves. Visible light with
wavelengths from about 400 nm for deep violet to 750 nm
for deep red covers only a small portion of the electromagnetic spectrum (see Chapter 3). Fiber-optic video
transmission uses the near-IR region, extending from
approximately 750 to 1500 nm, since glass fibers propagate
light at these wavelengths most efficiently, and efficient
detectors (silicon and germanium) are available to detect
such light.
166
CCTV Surveillance
(C)
SMALL AMOUNT
REFLECTED
(A) REFRACTION OF LIGHT
REFRACTED
LIGHT
CORE
CLADDING
θ1
GLASS
θ2
nCORE > nCLADDING
INCOMING
LIGHT
VELOCITY = VM
(B)
θ1
LIGHT LOST TO CLADDING
θC = CRITICAL ANGLE
FREE SPACE
(AIR)
θ1
θC
VELOCITY = C
NA ACCEPTED
θ1
NA = SIN θ
LIGHT TRANSMITTED
THROUGH CABLE CORE
NA = NUMERICAL APERTURE
FIGURE 6-17
Light reflection/transmission in fiber optics
6.3.4.3 Cable Types
The most significant part of the fiber-optic signal transmission system is the glass fiber itself, a thin strand of very
pure glass approximately the diameter of a human hair.
The fiber transmits visible and near-IR frequencies with
extremely high efficiency. Most fiber-optic system operate
at IR wavelengths of 850, 1300, or 1550 nm. Figure 6-19
shows where these near-IR light frequencies are located
with respect to the visible light spectrum.
Most short (several miles long) fiber-optic security systems operate at a wavelength of 850 nm rather than 1300
or 1550 nm because 850 nm LED emitters are more readily available and less expensive than their 1300 nm or
1550 nm counterparts. Likewise, IR detectors are more
sensitive at 850 nm. LED and ILD radiation at the 1300
and 1550 nm wavelengths is transmitted along the fiberoptic cables more efficiently than at the 850 nm frequency;
they are used for much longer run cables (hundreds of
miles).
Two types of fibers are used in security systems: (1) multimode step-index (rarely), and (2) graded-index. These
two types are defined by the index of refraction (n) profile of the fiber and the cross section of the fiber core.
The two types have different properties and are used in
different applications.
6.3.4.3.1 Multimode Step-Index Fiber
Figure 6-20a illustrates the physical characteristics of the
multimode step-index fiber.
The fiber consists of a center core of index n = 147
and outer cladding of index n = 2. Light rays enter the
core and are reflected a multiple number of times down
the core and exit at the far end. Since this fiber propagates many modes, it is called “multimode step-index.”
The multimode step-index is usually 50, 100, or even 200
microns (0.002, 0.004, or 0.008 inches) in diameter. The
fiber core itself is clad with a thin layer of glass having a
sharply different index of refraction. Light travels down
the fiber, constantly being reflected back and forth from
the interface between the two layers of glass. Light that
enters the fiber at a sharp angle is reflected at a sharp
angle from the interface and is reflected back and forth
many more times, thus traveling more slowly through the
fiber than light that enters at a shallow angle. The difference in the arrival time at the end of the fiber limits the
bandwidth of the step-index fiber, so that most such fibers
provide good signal transmission up to a 20 MHz signal for
Analog Video, Voice, and Control Signal Transmission
NEAR
FIELD
FIBER
CLADDING
C
A
θ
FIBER CORE
NUMERICAL APERTURE: NA = SIN θ = A /C
NA
0.1
0.2
0.3
0.4
0.5
FAR
FIELD
TYPICAL NA VALUES IN GLASS
f/#
θ (DEGREES)
5.7
5.00
11.5
2.45
17.5
1.58
23.4
1.14
30.0
0.87
S = SENDING
R = RECEIVING
NAS
NAR
SIGNAL
LOSS (dB)
1.0
LIGHT ENERGY
MISSED BY FIBER
(CROSSHATCH)
LIGHT FROM
FIBER 1
0.1
.01
LIGHT FROM
FIBER 2
.80
FIBER 1
.85
.90
.95
1.00
1.05
FIBER 2
NUMERICAL APERTURE MISMATCH RATIO =
FIGURE 6-18
1.10
NAR
NAS
Fiber optic numerical aperture
NEAR-IR
RELATIVE/
OUTPUT
RESPONSE
VISIBLE
850
1300
1550
10
9
HUMAN EYE
VIDICON
(RFEFERENCE)
RESPONSE
8
CCD,CMOS
7
6
5
4
3
2
1
0
FIGURE 6-19
400
500
600
700
800
Fiber optic transmission wavelengths
900
1000
1100
1200
1300 1400 1500 1600
WAVELENGTH (NANOMETERS)
167
168
CCTV Surveillance
(A) MULTI-MODE STEP-INDEX FIBER
N2
TYPICAL
ATTENUATION:
N1
50 mm
7–15 dB/km
@ 850 nm
NA = 0.30
125 mm DIA.
(B) MULTI-MODE GRADED-INDEX FIBER
VARIABLE N
TYPICAL
ATTENUATION:
N2
2.5–5.0 dB/km
@ 850 nm
50 mm
0.7–2.5 dB/km
@ 1300 nm
NA = .20
125 mm DIA.
FIGURE 6-20
Multimode fiber optic cable
about 1 kilometer. This limitation is more than adequate
for many video applications.
6.3.4.3.2 Multimode Graded-Index Fiber
The multimode graded-index fiber is the workhorse of
the video security industry (Figure 6-20b). Its low power
attenuation—less than 3 dB (50% loss) per kilometer at
850 nm—makes it well suited for short and long cable
runs. Most fibers are available in 50-micron-diameter core
with 125-micron total fiber diameter (exclusive of outside protective sheathing). Graded-index fiber sizes are
designated by their core/cladding diameter ratio, thus
the 50/125 fiber has 50-micron-diameter core and a
125-micron cladding. The typical graded-index fiber has
50/125
62.5/125 †
100/140
TYPICAL CABLE PARAMETERS
DIAMETER *
(MICRONS)
FIBER
TYPE
SINGLE FIBER
2 FIBER
OD **
4 FIBER
CLADDING
BUFFERING
OD
(mm)
WEIGHT
(kg/km)
(mm)
WEIGHT
(kg/km)
50
125
250
2.6
6.5
3.4 × 6
22
8
55
62.5
125
250
3.0
6.4
3.0 × 6.1
18
9.4
65.5
140
250
2.6
6.5
3.4
22
7.1
50
100
** CABLE OUTSIDE DIAMETER OR CROSS SECTION
† MOST
WIDELY USED IN SECURITY APPLICATIONS
1 mm = 1000 MICRONS
1 kg/km = 0.671 lb/1000 ft
Standard Fiber Optic Cable Sizes
OD
(mm)
WEIGHT
(kg/km)
CORE
* FIBER DIAMETER (1 MICRON = .00004 inch)
Table 6-3
a bandwidth of approximately 1000 MHz and is one of
the least expensive fiber types available. The 50/125 fiber
provides high efficiency when used with a high-quality
LED transmitter or for very long distances or very wide
bandwidths, with an ILD source. Table 6-3 lists some of
the common cable sizes available.
For the graded-index fiber, the index of refraction (n)
of the core is highest at the center and gradually decreases
as the distance from the center increases (Figure 6-20b).
Light in this type of core travels by refraction: the light rays
are continually bent toward the center of the fiber-optic
axis. In this manner the light rays traveling in the center
of the core have a lower velocity due to the high index
of refraction, and the rays at the outer limits travel much
Analog Video, Voice, and Control Signal Transmission
faster. This effect causes all the light to traverse the length
of the fiber in nearly the same time and greatly reduces
the difference in arrival time of light from different
modes, thereby increasing the fiber bandwidth–carrying
capability. The graded-index fiber satisfies long-haul, widebandwidth security system requirements that cannot be
met by the multimode step-index fiber.
6.3.4.3.3 Cable Construction and Sizes
A fiber-optic cable consists of a single optical fiber that
is surrounded by a tube of plastic substantially larger
than the fiber itself. Over this tube is a layer of Kevlar
reinforcement material. The entire assembly is then covered with an outer jacket, typically made of polyvinyl chloride (PVC). This construction is generally accepted for
use indoors or where the cable is easily pulled through
a dry conduit. The two approaches to providing primary
protection to a fiber is the tight buffer and loose tube
(Figure 6-21).
The tight buffer uses a dielectric (insulator) material
such as PVC or polyurethane applied tightly to the fiber.
For medium- and high-loss fibers (step-index type), such
cable-induced attenuation is small compared with overall attenuation. The tight buffer offers the advantages of
smaller bend radii and better crush resistances than loose-
tube cabling. These advantages make tightly buffered
fibers useful in applications of short runs where sharp
bends are encountered or where cables may be laid under
carpeted walking surfaces.
The loose-tube method isolates the fiber from the rest
of the cable, allowing the cabling to be twisted, pulled,
and otherwise stressed with little effect on the fiber. Microbends caused by tight buffers are eliminated by placing the
fiber within a hard plastic tube that has an inside diameter
several times larger than the diameter of the fiber. Fibers
for long-distance applications typically use a loose tube
since decoupling of the fiber from the cable allows the
cable to be pulled long lengths during installation. The
tubes may be filled with jelly to protect against moisture
that could condense and freeze and damage the fiber.
Multimode graded-index fiber is available in several primary core sizes: 50/125, 62.5/125, and 100/140. Table 6-3
summarizes the properties of different fiber-cable types
used in security systems, indicating the sizes and weights.
The first number, in the fiber designation (50 in 50/125),
refers to the core outside diameter size, the second (125)
to the glass fiber outside diameter (the sizes exclude reinforcement or sheathing). The fiber size is expressed in
microns: 1 micron (m) equals one one-thousandth of a
LOOSE TUBE BUFFER
OUTER
PROTECTIVE
JACKET
(3 mm DIA.)
TIGHT BUFFER 50/125 FIBER
LOOSE JACKET
BUFFER
(250 µm DIA.)
CLADDING
(125 µm DIA.)
OUTER
PROTECTIVE
JACKET
(3 mm DIA.)
CORE
(50 µm DIA.)
STRENGTH MEMBER
TIGHT BUFFER
JACKET (940 µm DIA.)
KEVLAR
STRENGTH
MEMBER
CLADDING
(125 µm DIA.)
TRANSMITTING
CORE (50 µm DIA.)
PVC OR
POLYURETHANE
INSULATOR
TYPICAL OPTICAL CHARACTERISTICS
NOTE: 1000 µm (MICRONS) = 1 mm (MILLIMETER)
(125 µm = .125 mm, 50 µm = .05 mm)
FIGURE 6-21
169
MINIMUM BANDWIDTH: 200 MHz
ATTENUATION: @ 850 nm = 4–6 dB/km
@ 1300 nm = 3 dB/km
NUMERICAL APERTURE = NA = .25
Tight-buffer and loose-tube single fiber optic cable construction
170
CCTV Surveillance
TYPICAL LOSS
CABLE LOSS TYPE
COMMENTS
(dB)
(%)
AXIAL-LATERAL DISPLACEMENT (10%)
0.55
12.0
ANGULAR MISALIGNMENT (2 DEGREES)
0.30
6.7
END SEPARATION (AIR GAP)
0.32
7.0
MOST CRITICAL FACTOR
FUNCTION OF NUMERICAL APERTURE
ESSENTIALLY ELIMINATED USING
INDEX MATCHING FLUID
END FINISH: (A)ROUGHNESS (1 MICRON)
(B)NON PERPENDICULAR
0.50
0.25
11.0
CORE SIZE MISMATCH:
1% DIAMETER TOLERANCE
±5% DIAMETER TOLERANCE
0.17
0.83
4.0
18.0
LOSS OCCURS ONLY WHEN LARGER
1.66
31.6
CRITICAL FACTOR WHEN NAS IS
LARGER THAN NAR
NUMERICAL APERTURE (NA)
DIFFERENCE OF ± 0.02 (2%)
NOTE: dB = DECIBELS = 10 LOG
POWERS
POWERR
Table 6-4
5.6
INCLUDES FRESNEL LOSS (.35 dB)
LOSS NOT COMMONLY FOUND
CORE COUPLES INTO SMALLER CORE
S = SENDING FIBER
R = RECEIVING FIBER
Fiber Optic Connector Coupling Losses
millimeter (1/1000 mm). By comparison, the diameter of
a human hair is about 0.002 inches or 50 microns.
Each size has advantages for particular applications,
and all three are EIA standards. The most popular and
least expensive multimode fiber is the 50/125, used extensively in video security. It has the lowest NA of any multimode fiber, which allows the highest bandwidth. Because
50/125 has been used for many years, established installers
are experienced and comfortable working with it. Many
connector types are available for terminating the 50/125
cable, an alternative to the 62.5/125 fiber.
The 50/125 and 62.5/125 were developed for telephone networks and are now used extensively for video.
An 85/125 was developed specifically for computer or
digital local networks where short distances are required.
The slightly larger 85-micron size permits easier connector
specifications and LED source requirements.
The 100/140 multimode fiber was developed in response
to computer manufacturers, who wanted an LEDcompatible, short-wavelength, optical-fiber data link that
could handle higher data rates than coaxial cable. While this
fiber was developed for the computer market it is excellent
for short-haul CCTV security applications. It is least sensitive
to fiber-optic geometry variations and connector tolerances
which generally means lower losses at joint connections.
This is particularly important in industrial environments
where the cable may be disconnected and connected
many times. The only disadvantage of 140 outsidediameter is that it is nonstandard, so available connectors
are fewer and more expensive than those for the 125 size.
6.3.4.3.4 Indoor and Outdoor Cables
Indoor and outdoor fiber-optic cables differ in the jacket
surrounding the fiber and the protective sheath that gives
it sufficient tensile strength to be pulled through a conduit or overhead duct or strung on poles. Single indoor
cables (Figure 6-22) consist of the clad fiber-optic cable
surrounded by a Kevlar reinforcement sheath, wrapped in
a polyurethane jacket for protection from abrasion and
the environment. The outdoor cable has additional protective sheathing for additional environmental protection.
Plenum fiber-optic cables are available for indoor applications that require specific smoke- and flame-retardant
characteristics and do not require the use of a metal
conduit. When higher tensile strength is needed, additional strands of Kevlar are added outside the polyethylene
jacket and another polyethylene jacket provided over these
Kevlar reinforcement elements. Some indoor cables utilize
a stranded-steel central-strength member or nonmetallic
Kevlar. Kevlar is preferred in installations located in explosive areas or areas of high electromagnetic interference,
where nonconducting strength members are desirable.
Analog Video, Voice, and Control Signal Transmission
(A) INDOOR CABLE
(B) OUTDOOR CABLE
SINGLE
DUPLEX
KEVLAR REINFORCEMENT
TIGHT BUFFER
LOW SMOKE AND
FLAME SPREAD
FLUOROPOLYMER
JACKET
PVC
OUTER
JACKET
OPTICAL FIBER
BRAIDED
KEVLAR STRENGTH
MEMBER
FIGURE 6-22
171
LOOSE
TUBE
BUFFERED
OPTICAL
FIBER
Indoor and outdoor fiber optic cable construction
The mechanical properties of cables typically found on
data sheets include crush resistance, impact resistance,
bend radius, and strength.
An outdoor cable or one that will be subjected to
heavy stress—in long-cable-run pulls in a conduit or aerial
application—uses dual Kevlar/polyethylene layers as just
described. The polyethylene coating also retards the deleterious effects of sunlight and weather.
When two fibers are required, two single cable structures
may be paired in siamese fashion (side by side) with a
jacket surrounding around them.
If additional fiber-optic runs are required, multi-fiber
cables (having four, six, eight, or ten fibers) with similar properties are used (Figure 6-23). The fibers are
enclosed in a single or multiple buffer tube around a
tensile-strength member composed of Kevlar and then surrounded with an outer jacket of Kevlar.
6.3.4.4 Connectors and Fiber Termination
This section describes fiber-optic connectors, techniques
for finishing the fiber ends when terminated with connectors, and potential connector problems. For very long
cable runs, joining and fusing the actual glass-fiber core
and cladding is done by a technique called “splicing.”
Splicing joins two lengths of cable by fusing the two fibers
(locally melting the glass) and physically joining them in
a permanent connection (Section 6.3.4.4.4).
Fiber-optic cables require connectors to couple the optical signal from the transmitter at the camera into the fiberoptic cable, and at the monitoring end to couple the light
output from the fiber into the receiver. If the fiber-optic
cable run is very long or must go through various barriers
(e.g. walls), the total run is often fabricated from sections
of fiber-optic cable and each end joined with connectors.
This is equivalent to an inter-line coaxial connector.
A large variety of optical connectors is available for
terminating fiber-optic cables. Most are based on butt coupling of cut and polished fibers to allow direct transmission
of optical power from one fiber core to the other. Such a
connection is made using two mating connectors, precisely
centering the two fibers into the connector ferrules and
fixing them in place with epoxy. The ferrule and fiber surfaces at the ends of both cables are ground and polished
to produce a clean optical surface. The two most common
types are the cylindrical and cone ferrule connectors.
6.3.4.4.1 Coupling Efficiency
The efficiency of light transfer from the end of one fiberoptic cable to the following cable or device is a function
of six different parameters:
1.
2.
3.
4.
5.
6.
Fiber-core lateral or axial misalignment
Angular core misalignment
Fiber end separation
Fiber distortion
Fiber end finish
Fresnel reflections.
Of these loss mechanisms, distortion loss and the effects
of fiber end finish can be minimized by using proper
techniques when the fibers are prepared for termination.
172
CCTV Surveillance
(A) INDOOR MULTI-FIBER
(B) OUTDOOR AERIAL-SELF SUPPORTING
KEVLAR REINFORCEMENT
CORE TAPE WRAP
CORE TAPE WRAP
TIGHT
BUFFER
TIGHT BUFFER
OPTICAL
FIBER
LOW SMOKE AND
FLAME SPREAD
FLUOROPOLYMER
JACKET
KEVLAR
REINFORCEMENT
OPTICAL FIBER
OUTER POLYETHYLENE
JACKET
INNER JACKET
FILLING COMPOUND
EPOXY GLASS
CENTRAL STRENGTH
MEMBER
(C) INDOOR
FIGURE 6-23
(D) OUTDOOR
(E) ARMORED
Multi-conductor fiber optic cable
A chipped or scratched fiber end will scatter much of
the light signal power, but proper grinding and polishing
minimize these effects in epoxy/polish-type connectors.
Lateral misalignment of fiber cores causes the largest
amount of light loss, as shown in Figure 6-24a.
An evaluation of the overlap area of laterally misaligned
step-index fibers indicates that a total misalignment of
10% of a core diameter yields a loss of greater than 0.5 dB.
This means that a fiber core of 0.002 inches (50 microns)
must be placed within 0.0001 inches of the center of its
connector for a worst-case lateral misalignment loss of
0.5 dB. While this dimension is small, the connection is
readily accomplished in the field.
Present connector designs maintain angular alignment
well below one degree (Figure 6-24b), which adds only
another 0.1 dB (2.3%) of loss for most fibers.
Fiber end–separation loss depends on the NA of the
fiber. Since the optical light power emanating from a
transmitting fiber is in the form of a cone, the amount
of light coupled into the receiving fiber or device will
decrease as the fibers are moved apart from each other
(Figure 6-24c). A separation distance of 10% of the core
diameter using a fiber with an NA of 0.2 can add another
0.1 dB of loss.
Fresnel losses usually add another 0.3 to 0.4 dB when
the connection does not use an index-matching fluid
(Figure 6-24d).
The summation of all of these different losses often
adds up to 0.5–1.0 dB for ST-type (higher for SMA 1906)
terminations and connections (Table 6-4).
6.3.4.4.2 Cylindrical and Cone Ferrule Connector
In the cylindrical ferrule design, the two connectors are
joined and the two ferrules are brought into contact inside
precisely guiding cylindrical sleeves. Figure 6-25 shows the
geometry of this type of connection.
Lateral offset in cylindrical ferrule connectors is usually the largest loss contributor. In a 50-micron gradedindex fiber, 0.5 dB (12%) loss results from a 5-micron
offset. A loss of 0.5 dB can also result from a 35-micron
gap between the ends of the fibers, or from a 2.5 tilted
fiber surface. Commercial connectors of this type reach
0.5–1 dB (12–26%) optical loss for ST type and higher
for SMA 1906. Optical-index-matching fluids in the gap
further reduce the loss.
The cone ferrule termination technique centers the
fiber in one connector and insures concentricity with the
mating fiber in the other connector using a cone-shaped
Analog Video, Voice, and Control Signal Transmission
(A) AXIAL (LATERAL) DISPLACEMENT (D )
LATERAL
MISALIGNMENT
LOSS (dB)
173
(C) END SEPARATION (S )
END
SEPARATION
LOSS (dB)
D
4
D
S
L
3
.5 NA
3
2
2
1
.2 NA
.15 NA
1
.1
.2
.3
.4
.5
.1
.2
LATERAL MISALIGNMENT RATIO L / D
(B) ANGULAR MISALIGNMENT θ
TILT
ANGULAR
LOSS (dB)
.3
.4
.5
END SEPARATION RATIO S/D
(D) END FINISH
SURFACE
ROUGHNESS
LOSS (dB)
θ
D
1.0
1.5
.15 NA
.2 NA
1.0
.5
.5 NA
.5
FRESNEL LOSS
(FOR PERFECT END FINISH)
0
1
2
3
4
1
5
MISALIGNMENT ANGLE θ IN DEGREES
8
15
FIBER FACE ROUGHNESS (µm)
TOTAL CONNECTOR LOSS = D + S + 0 + E (dB)
LOSS RANGE = 0.3 + 0.2 + 0.1 = 0.6 (GOOD) TO 0.7 + 0.5 + 0.2 + = 1.4 (POOR)
FIGURE 6-24
Factors affecting fiber optic coupling efficiency
CYLINDRICAL
CONICAL
FIBER
SLEEVE
FIGURE 6-25
PLUG
FIBER
SLEEVE
PLUG
Cylindrical and conical butt-coupled fiber optic ferrule connectors
plug instead of a cylindrical ferrule. The key to the cone
connector design is the placement of the fiber-centering
hole (in the cone) in relationship to the true center, which
exists when the connector is mated to its other half. A fiber
(within acceptable tolerances) is inserted into the ferrule,
adhesive is added, and the ferrule is compressed to fit
the fiber size while the adhesive sets. The fiber faces and
ferrule are polished to an optical finish and the ferrule
(with fiber) is placed into the final alignment housing.
Most low-loss fiber-optic connections are made utilizing
the cone-shaped plug technique. The two most popular
cone-shaped designs are the small-fiber resilient (SFR)
bonded connector and the SMA, a redesigned coaxial connector style (Figure 6-26).
174
CCTV Surveillance
TYPE: SMA
THREADED—SCREW ON
TYPE: ST, SFR (SMALL FIBER
RESILIENT) POLARIZED AND
SPRING LOADED QUARTER
TURN BAYONET LOCK
FIGURE 6-26
SMA and SFR connectors
Both use the cone-shaped ferrule, which provides a reliable, low-cost, easily assembled termination in the field.
Both connectors can terminate fibers with diameters of
125-micron cladding.
The technique eliminates almost all fiber and connector
error tolerance buildup that normally causes light losses. It
makes use of a resilient material for the ferrule, metal for
the construction of the retainer assembly, and a rugged
metallic connector for termination. The fiber alignment
is repeatable after many “connects” and “disconnects” due
to the tight interference fit of the cone-shaped ferrule into
the cone-shaped mating half. This cone design also forms a
sealed interface for a fiber-to-fiber or fiber-to-active-device
junction, such as fiber cable to transmitter or fiber cable
to receiver. Tolerances in the fiber diameter are absorbed
by the resiliency of the plastic ferrule. This connector
offers a maximum signal loss of 1.0 dB (26%) and provides
repeatable coupling and uncoupling with little increase in
light loss.
The popular SMA-style connector is compatible with
many other SMA-manufacturer-type connectors and terminates the 125-micron fibers. Internal ferrules insure axial
fiber alignment to within 0.1 . The SMA connector has a
corrosion-resistant metal body and is available in an environmentally sealed version.
6.3.4.4.3 Fiber Termination Kits
An efficient fiber-optic-cable transmission system relies
on a high-quality termination of the cable core and
cladding. This step requires the use of perhaps unfamiliar but easy techniques with which the installer must
be acquainted. Fiber-terminating kits are available from
most fiber-cable, connector, and accessory manufacturers.
Figure 6-27 shows a complete kit, including all grinding and polishing compounds, alignment jigs, tools, and
instructions.
Manufacturers can provide descriptions of the various techniques for terminating the ends of fiber-optic
cables, including cable preparation, grinding, polishing,
testing, etc.
6.3.4.4.4 Splicing Fibers
Splicing of multimode fibers is sometimes necessary in
systems having long fiber-optic-cable runs (longer than
2 km). In these applications it is advantageous to splice
cable sections together rather than connect them by using
connectors. A splice made between two fiber-optic cables
can provide a connection with only one-tenth the optical
loss of that obtained when a connector is inserted between
fibers. Good fusion splices made with an electric arc produce losses as low as 0.05–0.1 dB (1.2–2.3% loss). Making a
splice via a fusing technique is more difficult and requires
more equipment and skill than terminating the end of a
fiber with a connector. It is worth the effort if it eliminates the use of an in-line amplifier. The splice can also be
used to repair a damaged cable, eliminating the need to
add connector terminations that would decrease the light
signal level.
6.3.4.5 Fiber-Optic Transmitter
The fiber-optic transmitter is the electro-optic transducer
between the camera video electrical signal output and the
light signal input to the fiber-optic cable (Figure 6-15).
The function of the transmitter is to efficiently and accurately convert the electrical video signal into an optical
signal and couple it into the fiber optic. The transmitter
electronics convert the amplitude-modulated video signal
through LED or ILD into an AM or FM light signal, which
faithfully represents the video signal. The transmitter consists of an LED for normal security applications or an ILD
when a long range transmission is required. The former
is used for most CCTV security applications. Figure 6-28
illustrates the block diagram for the transmitter unit.
6.3.4.5.1 Generic Types
The LED light source is a semiconductor device made
of gallium arsenide (GaAs) or a related semiconductor
compound which converts an electrical video signal to an
optical signal. The LED is a diode junction that spontaneously emits nearly monochromatic (single wavelength
or color) radiation into a narrow light beam when current
is passed through it. While the ILD has a very narrow beam
Analog Video, Voice, and Control Signal Transmission
175
FIBER END GRINDING AND POLISHING
CABLE TERMINATING KIT
SMA
POLISH
TOOL
WET WITH
WATER
600 GRIT
3 MICRON
0.3 MICRON
GRIND AND POLISH IN FIGURE 8 PATTERN
FIGURE 6-27
Fiber optic termination kit
LENS
SCENE
CAMERA
COAX
CABLE
FIBER OPTIC TRANSMITTER
IMPEDANCE
MATCHING
INPUT
ATTENUATOR
STAGE
FIGURE 6-28
LINEARIZING
MODULATOR
STAGE
HIGH
LINEARITY
DRIVER
AMPLIFIER
LOW LOSS
CONNECTOR
LIGHT EMITTING
DIODE (LED)
WITH OPTICS
FIBER
OPTICS
Block diagram of LED fiber optic transmitter
width and is more powerful, the LED is more reliable, less
expensive, and easier to use. The ILD is used in very long
distance, wide-bandwidth fiber-optic applications.
The LED’s main requirements as a light source are:
(1) to have a fast operating speed to meet the bandwidth
requirements of the video signal, (2) to provide enough
optical power to provide the receiver with a signal-to-noise
(S/N) ratio suitable for a good television picture, and
(3) to produce a wavelength that takes advantage of the
low-loss propagation characteristics of the fiber.
The parameters that constitute a good light source
for injecting light into a fiber-optic cable are those that
produce as intense a light output into as small a cone
diameter as possible. Another factor affecting the lighttransmission efficiency is the cone angle of the LED output that can be accepted by and launched down the
fiber-optic cable. Figure 6-29 illustrates the LED-coupling
problem.
The entire output beam from the LED (illustrated by
the cone of light) is not intercepted or collected by the
fiber-optic core. This unintercepted illumination loss can
be a problem when the light-emitting surface is separated
from the end of the fiber core. Most LEDs have a lens
at the surface of the LED package to collect the light
from the emitting source and concentrate it onto the core
of the fiber.
176
CCTV Surveillance
LED OUTPUT
BEAM
LIGHT EMITTING
DIODE (LED)
JUNCTION
LENS AND
WINDOW
LENS
FIBER OPTIC
CLADDING
CORE
LIGHT
BEAM
LED
FIGURE 6-29
LED light beam output cone angle
6.3.4.5.2 Modulation Techniques
For video security applications, the electrical signal from
the camera is AM or FM and converted to light output
variations in the LED or ILD. The optical output power
varies directly to the electrical input signal for AM and
is constant for FM. LEDs with an 850 nm IR wavelength
emission are best suited since they can be amplitude modulated: the electrical video signal can be converted to a
light output signal that is a near-linear function of the LED
drive current. This produces a very faithful transformation
of the electrical video information to the light information
that is transmitted along the fiber-optic cable.
of aluminum. In most transmitters today, the emitting
wavelength is 850 nm, which matches the maximum transmission capability of the glass fiber.
Alternative transmitting wavelengths are 1060, 1300, and
1550 nm, which are regions where glass fibers exhibit a
lower attenuation and dispersion than at 850 nm. These
wavelengths are produced by combining the element
indium with gallium arsenide (to get InGaAs) and are
used in some long-distance transmission applications.
6.3.4.5.3 Operational Wavelengths
An important characteristic of the transmitter output is the
wavelength of the emitted light. This should be compatible
with the fiber’s minimum-attenuation wavelength, which is
850 nm (in the IR region) for most CCTV fiber-optic cable.
The wavelength of light emitted by an LED depends on the
semiconductor material composition. Pure GaAs diodes
emit maximum power at a wavelength of 940 nm (nearIR), which is undesirable because most glass fibers have
a high attenuation at that wavelength. Adding aluminum
to GaAs to produce a GaAlAs diode yields a maximum
power output at a wavelength between 800 and 900 nm,
with the exact wavelength determined by the percentage
The term receiver at the output end of the fiber-optic
cable refers to a light-detecting transducer and its related
electronics that provides any necessary signal conditioning to restore the signal to its original shape at the input
and additional signal amplification. The most common
fiber-optic receiver uses a photodiode to convert the incident light from the fiber into electrical energy. To interface the receiver with the optical fiber, the proper match
between light source, fiber-optic cable, and light detector
is required. In the AM transmission system, the optical
power input at the fiber is modulated so the photodetector
operating in the photocurrent mode must provide good
linearity, speed, and stability. The photodiode produces
6.3.4.6 Fiber-Optic Receiver
Analog Video, Voice, and Control Signal Transmission
177
FIBER OPTIC RECEIVER
FIBER OPTIC
CONNECTOR
FIBER
OPTIC
RECEIVER
OPTICS
OPTICAL
DETECTOR
(PIN DIODE)
HIGH
GAIN
VIDEO
AMPLIFIER
VARIABLE
GAIN
POST
AMPLIFIER
OUTPUT
DRIVER
COAXIAL
CABLE
VIDEO
OUTPUT
SIGNAL
GAIN
MONITOR
FIGURE 6-30
Fiber optic receiver block diagram
no electrical gain and is therefore followed by circuits that
amplify electrical voltage and power to drive the coaxial
cable. Figure 6-30 illustrates the block diagram for the
receiver unit.
The light exiting from the receiver end of an optical
fiber spreads out with a divergence approximately equal
to the acceptance cone angle at the transmitter end of
the fiber. Photodiodes are packaged with lenses on their
housings so that the lens collects this output energy and
focuses it down onto the photodiode-sensitive area.
After the light energy is converted into an electrical
signal by the photodiode, it is linearly amplified and conditioned to be suitable for transmission over standard coaxial
cable or two-wire UTP to a monitor or recorder.
6.3.4.6.1 Demodulation Techniques
The receiver demodulates the video light signal to its
original base-band video form. This takes the form of
either AM or FM demodulation. Since FM modulation–
demodulation is less sensitive to external electrical influences it is the technique of choice in most systems.
6.3.4.7 Multi-Signal, Single-Fiber Transmission
The primary attribute of fiber-optic transmission is the
cable’s wide signal bandwidth capability. Transmitting a
single video signal on a single fiber easily fits within
the bandwidth capability of all fiber-optic cables. Modulators and demodulators in transmitters, receivers, and
transceivers permit transmission of bidirectional video,
audio, and control signals over a single optical-fiber
cable. Using the full-duplex capabilities of the system, the
transceiver at the camera transmits video and audio signals
from the camera location to the monitor location while
simultaneously receiving audio, control, or camera genlock signals from the transceiver at the monitor location.
All transmissions occur via the same single optical-fiber
cable. The transmitter and receiver contain all circuitry
for the bidirectional transmission of pan/tilt, zoom, focus,
iris, contact-closure, and video in the opposite direction.
When more than one video signal is to be transmitted by
optical fiber between two points, either multiple fibers may
be used or the signals may be combined using wavelength
division multiplexing (WDM MUX), thus saving the cost
of additional fibers and expensive multi-fiber connectors.
Using this technique the outputs from each optical transmitter operating at different wavelengths (1060, 1300, and
1550 m) are modulated by separate video signals and
combined on a single optical fiber and combined using
WDM MUX. These video signals may then be separated
at the other end of the fiber by a WDM de-multiplexer
(WDM DE-MUX) (Figure 6-31).
Typically two or more wavelengths are provided by two
LED transmitters operating at wavelengths between 850
and 1550 nm. The WDM MUX and DE-MUX may be fabricated using an optical coupler and splitter assembly, using
lens and grating components. The lens focuses each of the
channels onto the grating, which then separates the channels according to wavelength and according to the grating
spacing. Data sheets for a typical WDM MUX/DE-MUX
devices include the following specifications:
1. Number of Channels: The number of video signals that
can be multiplexed and de-multiplexed over the optical
fiber to which the WDM MUX/DEMUX are connected.
178
CCTV Surveillance
WAVELENGTH DIVISION MULTIPLEXING (WDM)-TRANSMITTER
ELECTRICAL SIGNAL
TO LIGHT PULSE
MODULATORS
MODULATED
LASER/LED
1060 nm
INPUT: * VIDEO 1
LIGHT
COUPLER
1300 nm
VIDEO 2
1550 nm
VIDEO 3
* UP TO 32 CHANNELS
OF VIDEO, VOICE, DATA
SINGLE FIBER OPTIC
DWDM-RECEIVER
GRATING **
COLLIMATED
BEAM
** GRATING FUNCTION:
DISPERSE MULTI-WAVELENGTH
LIGHT INTO CONSTITUANT
COMPONENTS (λ1 λ 2 λ 3 …)
λ1
λ2
λ3
LIGHT
DETECTORS
OUTPUT:
VIDEO 1
LIGHT TO ELECTRICAL
SIGNAL DEMODULATORS
VIDEO 2
VIDEO 3
FF631
FIGURE 6-31
Wavelength division multiplexing and de-multiplexing video signal
2. Center Wavelengths: Center wavelength of the channels
over which the video signals are multiplexed.
3. Channel Spacing: The minimum distance (wavelength or frequency) between channels in a WDM
MUX/DEMUX system. In the illustration the channel spacing is approximately 0.8 nm, approximately
100 GHz.
4. Bandwidth (also referred to as Passband Width): The linewidth of a specific wavelength channel. A manufacturer
generally specifies the line-width at 1 dB, 3 dB and 20 dB
insertion loss as shown in Figure 6-32.
5. Maximum Insertion Loss: Loss sustained by the video signals when the WDM MUX/DE-MUX is applied in a
system. Typical values range from 1.5 dB for a 4-channel
device to 6 dB for a high number of channels, WDM
MUX/DE-MUX.
6. Isolation: The loss or attenuation between video signal
channels, usually more than 30 dB.
6.3.4.8 Fiber Optic—Advantages/Disadvantages
Figure 6-33 illustrates two typical channels over which
the video signals 1540.56 m and 154135 m are
multiplexed.
6.3.4.8.1 Pro
Widest Bandwidth. In general the bandwidth capacity
is directly proportional to the carrier frequency. Light
Why go through all the complexity and extra expense of
converting the electrical video signal to a light signal and
then back again? Fiber optics offers several very important
features that no electrical cabling system offers, including:
• ultra-wide bandwidth supporting multiple video, audio,
control, and communications signals on one fiber
• complete electrical isolation
• complete noise immunity to RFI, EMI, and electromagnetic pulse (EMP)
• transmission security (fiber-optic cable is hard to tap)
• no spark or fire hazard or short-circuit possibility
• absence of crosstalk
• no RFI/EMI radiation.
Table 6-5 compares the features of coaxial and fiber-optic
transmission.
Analog Video, Voice, and Control Signal Transmission
INSERTION
LOSS (dB)
CHANNEL
SPACING
179
PASS
BAND
0
INSERTION
LOSS
–5
–10
–15
–20
SIGNAL OVERLAP
REGION— CROSSTALK
–25
–30
–35
–40
1542.0
1542.5
1543.0
1543.5
1544.0
1544.5
1545.0
1545.5
WAVELENGTH (nm)
FIGURE 6-32
Line-width vs. insertion loss of a specific wavelength channel
(and near IR) frequencies are approximately 1014 Hz. Typical video microwave transmitters operate at 1010 GHz or
1010 Hz. Fiber optics has a 104 or 10,000 times higher
bandwidth capability than microwave.
Electrical Isolation. The complete electrical isolation of
the transmitting section (i.e. the camera, lens controller,
pan/tilt, and related equipment) from the receiving
section (i.e. the monitor, recorder, printer, switching network, and so on) is very important for inter-building and
intra-building locations when a different electrical power
source is used for each location. Using fiber-optic transmission prevents all possibility of ground loops and ground
voltage differences that could require the redesign of a
coaxial cable–based system.
RFI, EMI, and EMP Immunity. When a transmission path
runs through a building or outdoors past other electrical
equipment, the site survey usually cannot uncover all possible contingencies of existing RFI/EMI noise. This is also
true of EMP and lightning strikes. Therefore, using fiber
optics in the initial design prevents any problems caused
by such noise sources.
Transmission Security. Since the fiber optic has no
electrical noise to leak and no visible light, it exhibits
excellent inherent transmission security and it is hard to
intercept. Methods for compromising the fiber-optic cable
are difficult, and the intrusion is usually detected. To tap
a fiber-optic cable, the bare fiber in the cable must be isolated from its sheath without breaking it. This will probably
end the tapping attempt. If the bare fiber is successfully
isolated, an optical tap must be made, the simplest of
which is achieved by bending the fiber into a small radius
and extracting some of the light. If a measurable amount
of power is tapped (which is necessary for a useful tap),
the tap can be detected by monitoring the power at the
system receiver. In contrast, tapping a coaxial cable is easy
to do and hard to detect.
No Fire Hazards. Since no electricity is involved in any
part of the fiber-optic cable, there is no chance or opportunity for sparks or electrical short circuits, and hence
no fire hazards. Short circuits and other hazards encountered in electrical wiring systems can start fires or cause
explosions. When a light-carrying fiber is severed, there is
no spark, and a fiber cannot short-circuit in the electrical
meaning of the term.
Absence of Crosstalk. Because the transmission medium
is light, there is no crosstalk between any of the fiber-optic
cables. Therefore there is no degradation due to the close
proximity of cables in the same bundle, as there can be
when multiple channels are encased in the same electrical
cable.
180
CCTV Surveillance
BI-DIRECTIONAL WAVELENGTH DIVISION MULTIPLEXING (WDM)
CAMERA 1
VIDEO
TRANSMITTER
λ1
PAN/ TILT/ZOOM
PLATFORM
4 CHANNEL *
CAMERA 3
MUX /DMUX
VIDEO
CAMERA 2
TRANSMITTER
DOME
VIDEO
TRANSMITTER
PAN/ TILT
CAMERA
LENS
CONTROLS
RECEIVER
λ2
λ3
SINGLE FIBER OPTIC
IN 1550 nm BAND
λ4
λ1
4-CHANNEL *
MUX / DMUX
λ2
λ3
* MANY WDM/DMDM ARE BI-DIRECTIONAL
DISPLAY
RECEIVER
RECEIVER
SWITCHER
RECEIVER
λ4
PAN / TILT
CONTROLLER
TRANSMITTER
LENS
CONTROLLER
FIGURE 6-33 Video signals and controls multiplexed over four channels
Analog Video, Voice, and Control Signal Transmission
dB/1000 ft
dB/km
OUTSIDE
DIAMETER
(inches)
ATTENUATION
DESIGNATION
@ 5–10 MHz
CABLE TYPE
dB/100 ft
RG59
COAXIAL
1.0
10.0
32.8
.242
3.5–4.0
RG59 MINI
COAXIAL
1.3
13.0
42.6
.135
1.5
RG6
COAXIAL
.8
8.0
26.2
.272
7.9
RG11
COAXIAL
.51
5.1
16.7
.405
9 –11
2422/UL1384
MINI–COAX
3.96
39.6
129.9
.079
0.9
2546 †
MINI–COAX
1.82
18.2
59.7
.13
1.4
†
MINI–COAX
2.1
21.0
69.0
.118
1.0
MINI–COAX
2.0
20.0
65.6
.089
1.0
.036
—
.12
.50 –1.0
.244
—
2895
RG179B/U
10/125 *
FIBER
OPTIC
850 NM **
1300 NM
50/125
FIBER
OPTIC
140/200
FIBER
OPTIC
—
—
—
.01–.02
.1–.2
.4–.8
850 NM
1300 NM
.12–.21
1.2–2.1
4–7.0
.09–.18
.9–1.8
850 NM
1300 NM
.08–.18
.02–.14
.8–1.8
.2–1.4
3–6.0
2.5–6.0
.8–4.5
* USED ONLY IN VERY LONG DISTANCE, WIDE BANDWIDTH APPLICATIONS
** TRANSMISSION WAVELENGTH
†
WEIGHT (lb)
PER 100 ft
(NANOMETER–NM)
MOGAMI
Table 6-5
181
1 KILOMETER (km) = 3280 FEET (ft)
1 MILE (Mi) = 1.609 KILOMETERS (km)
1 POUND (lb) = .454 KILOGRAMS (kg)
Comparison of Fiber Optic and Coaxial Cable Transmission
No RFI or EMI Radiation. Fibers do not radiate energy.
They generate no interference to other systems. Therefore, the fiber-optic cable will not emit any measurable
EMI/RFI radiation, and other cabling in the vicinity will
suffer no noise degradation. There are no FCC requirements for fiber-optic transmission.
6.3.4.8.2 Con
Higher Cost. Coaxial cable and connectors are inexpensive and no transmitter/receiver pairs are required. For
short distances if new cable must be run, coaxial is the
most cost-effective.
Connector Termination More Difficult. Fiber-optic cable
and connectors cost more than coaxial. Terminating fiber
cables takes longer than coaxial cables and requires more
technical skill.
6.3.4.9 Fiber-Optic Transmission: Checklist
The following are some questions that should be asked
when considering the use or design of a new fiber-optic
transmission system.
1. What are the lengths of cable runs? If over 500 feet,
then fiber optic should be considered. In screen rooms
or tempest areas, runs as short as 10 feet sometimes
require fiber-optic cable.
2. What size core/clad-diameter fiber should be used?
The most common diameter is 50/125 microns.
3. What wavelength should be used—850, 1060, 1300, or
1550 nm? The most common wavelengths are 850 and
1300 nm.
4. How many fibers are necessary for transmitting video,
audio, and controls? Should single- or multi-fiber cable
be used?
5. Are the cable runs going to be indoors or outdoors?
Separate the indoor and outdoor requirements and
determine if outdoor fiber cables are required. What
will the outdoor environment be (lighting, etc.)?
6. If outdoors, will the fiber be strung on poles, surfacemounted on the ground, undergo direct burial, or
pass through a conduit? Choose cable according to
manufacturers’ recommendations.
7. If indoors, will it be in a conduit, cable tray or trough,
plenum, or ceiling? Choose cable according to manufacturers’ recommendations.
8. What temperature range will the fiber-optic cable
experience? Most cable types will be suitable for most
indoor environments. For outdoor use, the cable chosen must operate over the full range of hot and cold
temperatures expected and must withstand ice and
wind loading if mounted above ground level.
9. Are there any special considerations such as water,
ice, chemicals? See manufacturers’ specifications for
extreme environmental hazards.
10. Are there special safety codes? Fiber-optic cable is available with plenum-grade or special abrasion-resistant
construction.
182
CCTV Surveillance
11. Should spare cables be included? Each design is different, but it is prudent to include one or more spare
fiber-optic cables to account for cable failure or future
system growth. The number of spares also depends on
how easy or difficult it is to replace a cable or add to
existing cables.
6.4 WIRED CONTROL SIGNAL TRANSMISSION
Fixed cameras do not require any control functions. Moving cameras require pan, tilt, focus and sometimes iris
control signals for proper operation.
Microwave transmission also is controlled by the FCC.
Rules are set forth and licenses are required for certain frequencies and applications, which limits usage to
specific purposes. The US government currently exercises strict control over transmission of wireless video via
RF and microwave. Up until recently, RF and microwave
atmospheric video transmission links were limited to governmental agencies (federal, state, and local) that could
obtain the necessary licenses. Now some low-power RF and
microwave transmitters and receivers suitable for short
links (less than a mile) are available for use without an
FCC license. High power and RF microwave links are
licensable by private users after a frequency check is made
with the FCC.
6.4.1 Camera/Lens Functions
6.5.1 Transmission Types
Lenses can require zoom, focus, and iris-control signals.
6.4.2 Pan/Tilt Functions
Moving cameras require pan/tilt functions to scan the
camera horizontally, tilt it vertically, and set preset pointing directions to specific locations in the scene.
6.4.3 Control Protocols
The simplest controls can take the form of on/off or proportional voltage control wiring for each function. These
direct controls require the largest number of wires. The
most standard two- and four-wire control protocol for
cameras, lens, and pan/tilt platforms are the RS422 and
RS485.
6.5 WIRELESS VIDEO TRANSMISSION
Most video security systems transmit video, audio, and control signals via coaxial cable, two-wire, or fiber-optic transmission means. These techniques are cost-effective and
reliable, and provide an excellent solution for transmission. However, there are applications and circumstances
that require wireless transmission of video and other
signals.
The video signal can be transmitted from the camera to
the monitor through the atmosphere, without having to
connect the two with hard wire or fiber. The most familiar technique is the transmission of commercial television
signals from some distant transmitter tower to consumer
television sets, broadcast through the atmosphere on VHF
and UHF radio frequency channels.
Commercial broadcasting is of course rigidly controlled
by the FCC, whose regulations dictate its precise usage.
Some examples of wireless TV transmission described
in the following sections include microwave (groundto-ground station, satellite), RF over VHF or UHF, and
light-wave transmission using IR beams. The hardware
cost of RF, microwave, and lightwave systems is considerably higher than any of the copper-wire or fiber-optic
systems, and such systems should be used only when absolutely necessary as when their use avoids expensive cable
installations (such as across roadways), or in temporary
or covert applications, wireless transmission becomes
cost-effective.
The results obtainable with hard-wired copper-wire
or fiber-optic video transmission are usually predictable,
with the exception of interference that might occur due
to the copper-wire cables running near electromagnetic
radiating equipment or electrical storms. The results
obtained with wireless transmission are generally not as
predictable because of the variable nature of the atmospheric path and materials through which RF, microwave,
or light signals must travel, as well as the specific transmitting and propagating characteristics of the particular wavelength or frequency of transmission. Each of
the three wireless transmitting regimes acts differently
because of the wide diversity in frequencies at which they
transmit.
6.5.2 Frequency and Transmission Path
Considerations
The RF link constitutes the lowest carrier frequency
(Figure 6-34). It penetrates many visually opaque
materials, goes around corners, and does not require a
line-of-sight path (i.e. a receiver in sight of the transmitter) when transmitting from one location to another. The
radio frequencies are, however, susceptible to attenuation
and reflection by metallic objects, ground terrain, or large
Analog Video, Voice, and Control Signal Transmission
183
POWER
OUTPUT
RF
SPECTRUM
100
MICROWAVE
SPECTRUM
UHF
VHF
10.525 GHz
920–930 MHz
5–5.8 GHz
21–24 GHz
2.4–2.5 GHz
150
980
8
13 GHz
0
100 MHz
FIGURE 6-34
200
500
1000 MHz
1 GHz
2 GHz
5 GHz 10 GHz
20 GHz
FREQUENCY
Wireless video transmission frequencies
buildings or structures, and therefore they sometimes produce unpredictable results.
The microwave link requires an unobstructed line of
sight; any metallic or wet objects in the transmission path
cause severe attenuation and reflection, often rendering a
system useless. However, metallic poles or flat surfaces can
sometimes be used to reflect the microwave energy, allowing the beam to turn a corner. Reflection of this type does
reduce the energy reaching the receiver and the effective
range of the system. Some microwave frequencies penetrate dry nonmetallic structures such as wood or drywall
walls and floors, so that non-line-of-sight transmission is
possible.
The frequency range most severely attenuated by the
atmosphere and blocked completely by any opaque object
is a light-wave signal in the near-IR region. The IR beam
can be strongly attenuated by heavy fog or precipitation, severely reducing its effective range as compared
with clear-line-of-sight, clear-weather conditions. As would
be expected, the IR-wavelength system requires a clear
line of sight with no opaque obstructions whatsoever
between the transmitter and the receiver. The IR beam
can be reflected off one or more mirrors to go around
corners. The advantages of the IR system over RF and
microwave links are: (1) security (since it is hard to tap
a narrow light beam), (2) high bandwidth (able to carry
multiple channels of information), and (3) bidirectional
operation.
6.5.3 Microwave Transmission
Microwave systems applicable in television transmission
have been allocated frequencies in bands from 1 to 75 GHz
(see Table 6-6).
Microwave frequencies, which approach light-wave frequencies, are usually transmitted and received by parabolically shaped reflector antennas or metallic horns. Even
when a line of sight exists, there can be signal fading,
caused primarily by changes in atmospheric conditions
between the transmitter and the receiver, a problem that
must be taken into account in the design. This fading can
result at any frequency, but in general is more severe at
the higher microwave frequencies.
6.5.3.1 Terrestrial Equipment
For terrestrial use, several manufacturers provide reliable
microwave transmission equipment suitable for transmitting video, audio, and control signals over distances of
from several hundred feet to 10–20 miles in line-of-sight
conditions.
One system transmits a single NTSC video channel and
two 20 kHz audio channels over a distance of 1 mile. A
high-gain directional antenna is available to extend the
system operating range to several miles. Figure 6-35a shows
the transmitter and receiver units. This system operates at
184
CCTV Surveillance
BAND/USE *
FREQUENCY
BAND (GHz)
SECURITY
FREQUENCY
RANGE (GHz)
USAGE
RESTRICTIONS
L
1.2–1.7
1.2–1.7
VIDEO TRANSMITTER
S
2.4–2.5
2.4–2.5
VIDEO TRANSMITTER
LOW POWER /HIGH POWER
FCC PART 15
S
2.6–3.95
G
3.95–5.85
—
C
4.9–7.05
—
C
5–5.8
5–5.8
VIDEO TRANSMITTER
LOW POWER/HIGH POWER
FCC PART 15
J
5.85–8.2
8.4–8.6
VIDEO TRANSMITTER
H
7.05–10.0
—
X
8.2–12.4
M
10.0–15.0
10.525
P
12.4–18.0
10.35–10.8
VIDEO TRANSMITTER
N
15.0–22.0
21.2–23.2
VIDEO UP TO 3 Mi RANGE
K
2.45
LOW POWER—NO RESTRICTIONS
NO FCC LICENSE REQUIRED
HIGH POWER—LICENSE REQURIED
CONSUMER MICROWAVE OVEN
LOW POWER—NO RESTRICTIONS
NO FCC LICENSE REQUIRED
HIGH POWER—LICENSE REQURIED
10.4–10.6
18.0–26.5
24.125
R
26.5–40.0
—
V
40–75
—
VIDEO, AUDIO, INTRUSION
VIDEO, VOICE, INTRUSION
* ALSO SEE TABLE 6-7
FCC PART 15.249 TRANSMITTER, ANY TYPE OF MODULATION
902–928 MHz–50 mV/M MAXIMUM AT 3 METERS
2.4–2.F835 GHz–50 mV/M MAXIMUM AT 3 METERS
5.735–5.875 GHz–250 mV/M MAXIMUM AT 3 METERS
Table 6-6
GOVERNMENT SECURITY
LAW ENFORCEMENT ONLY
1 GIGAHERTZ (GHz) = 1000 MHz
FCC PART 15.247 TRANSMITTER USING SPREAD SPECTRUM
902–928 MHz–1 WATT MAXIMUM
2.4000–2.4835–1 WATT MAXIMUM
5.725–5.675 GHz–1 WATT MAXIMUM
Microwave Video Transmission Frequencies
a carrier frequency of 2450–2483.5 MHz with a power output of 1 watt. The transmitter and receiver operate from 11
to 16 volts DC derived from batteries, an AC-to-DC power
supply, or 12 volts DC vehicle power. The microwave transmitter utilizes an omnidirectional antenna. A high-gain,
low-noise receiver collects the microwave transmitter signal with an omnidirectional or directional antenna. The
system has a selectable video bandwidth from 4.2 MHz
for enhanced sensitivity or 8 MHz for high resolution
and has a single or dual audio sub-carrier channel for
audio communications between the two sites. It transmits
monochrome or color video with excellent quality.
The 2450–2483.5 MHz band is available for a variety of
industrial applications and requires an FCC license for
operation. The system operates indoors or outdoors, uses
FM, and provides immunity from vehicles, power lines, and
other AM-type noise sources. The microwave frequency
utilized has the ability to penetrate dry walls and ceilings
and reflect off metal surfaces.
Figures 6-35b, c show examples of small short-range
microwave transmitters operating at 2.4 GHz and 5.8 GHz
designed for outdoor use. These systems use directional
patch antennas pointed toward each other to provide the
necessary signal at the receiver from the transmitter.
The systems are weatherproof, pedestal mounted,
and designed for permanent installation. They transmit
excellent full-color or monochrome pictures over an FM
carrier in a frequency range of 2.4 GHz and 5.8 GHz with
a video bandwidth of 10 MHz. In addition to the video
channel, the system is capable of providing up to three
voice or data (control) channels. The data channels may
be used to control pan/tilt, zoom, focus, and iris at the
camera location. Low power systems do not require FCC
licensing. FCC licensing is required for high power systems
and can be obtained for government and industrial users,
providing an authorized interference survey is made to
verify that no interference will result in other equipment.
Other variations and functions, the microwave transmitter/receiver systems can perform include:
1. Operation in any frequency band from 8.5 to 12.4 GHz
with output powers up to 100 milliwatts.
Analog Video, Voice, and Control Signal Transmission
185
(A) 2.4 GHz TRANSMITTER/RECEIVER
(B) 2.4 GHz OUTDOOR
FIGURE 6-35
(C) LONG RANGE 5.8 GHz
Monochrome/color microwave video transmission systems
2. Operation as a command-and-control unit providing
a multi-channel system for transmitting control signal information. The commands are encoded at the
transmitter and decoded at the receiver to control
power on/off, lens focus, zoom and iris, and camera
motion (pan/tilt).
3. An audio channel to provide simplex (one-way) or
duplex (two-way) communications (IR system).
4. The ability to sequence through and transmit the
video outputs from multiple surveillance cameras. The
receiver and control units are located at the monitor
site and the transmitter and sequencer units are located
with the CCTV cameras. The camera outputs are fed
to the sequencer unit. The operator at the receiver
end controls the sequencing of the eight cameras and
has the option to: (1) manually advance through the
186
CCTV Surveillance
cameras, (2) have the cameras sequence automatically,
or (3) change the camera dwell time.
6.5.3.2 Satellite Equipment
Microwave transmission of video signals can be accomplished via satellite. Such systems are in extensive use
for earth-to-satellite-to-earth communications, in which
one ground-based antenna transmits to an orbiting synchronous satellite repeater, which relays the microwave
signal at a shifted frequency to one or more receivers on
earth (Figure 6-36).
While this type of communication and transmission for
video security applications was not put into widespread
use for analog video systems, it now is enjoying wide special use for digital video Internet (WWW) systems. The
satellites used for transmission are in a synchronous orbit
at an altitude of 22,300 miles and appear stationary with
respect to the earth. Satellites are placed in a synchronous
or stationary orbit to permit communications from any two
points in the continental USA by a single “up” and a single
“down” transmission link. Consequently, a characteristic of
domestic satellite video communications is that the trans-
mission cost is independent of terrestrial distance. It takes
0.15 seconds for a microwave signal traveling at the speed
of light to make a one-way journey to or from the satellite.
Therefore, there is a 0.3 second delay between transmission and reception of the video carrier, independent of
ground distance. This delay is not usually a problem for
transmission of video security signals; however, this must
be kept in mind when synchronization of different incoming video signals is required. The signal level reaching the
feed horn depends on the size and shape of the antenna
(Figure 6-37).
The quality of an antenna is determined by how well
it concentrates the radiation intercepted from a target
satellite to a single point and by how well it ignores
noise and unwanted signals coming from sources other
than the target satellite. Three interrelated concepts—
gain, beam width, and noise temperature—describe how
well an antenna performs. Antenna gain is a measure of
how many thousands of times a satellite signal is concentrated by the time it reaches the focus of the antenna.
For example, a typical well-built 10-foot-diameter primefocused antenna dish can have a gain of 40 dB, which is a
factor of 10,000 power gain, which means that the signal is
ORBITTING
SATELLITE
.15 sec
TRANSIT TIME
.15 sec
TRANSIT TIME
MICROWAVE
RECEIVER
FIGURE 6-36
Satellite video transmission systems
MICROWAVE
TRANSMITTER
Analog Video, Voice, and Control Signal Transmission
187
EARTH
ORBITING
SATELLITE
SATELLITE
DISH
ANTENNA
FEED
HORN
LOW
NOISE
AMPLIFIER
ANTENNA
RECEIVING DISH
MONITORING
ROOM
FEED HORN
LOW
LOSS
COAX
CABLE
VIDEO
MONITOR
DISPLAY
LOW-NOISE
AMPLIFIER
(LNA)
AMPLIFIER
DOWN
CONVERTER
TUNER:
UHF
VHF
ANTENNA LNA
FINE POINTING
MECHANISM
FIGURE 6-37
Satellite video receiver system
concentrated 10,000 times higher at the focal point than
anywhere on the antenna. This gain is primarily dependent on the following three factors.
Dish Size. As the size of a dish increases, more radiation from space is intercepted. Thus if the diameter of an
antenna is doubled, the gain is increased fourfold (four
times the area).
Frequency. Gain increases with increasing frequency
because higher-frequency microwaves, being closer to the
frequency of light, behave a little more like light. Thus
they do not spread out like waves in water but can be
focused more easily into straight lines like beams of light.
Since the gain of a microwave antenna is proportional to
the square of the frequency, a signal with twice the frequency is concentrated by an antenna with four times the
gain. As an example, if the gain is 10,000 when a signal
of 5 GHz is received, then it will have a gain of 40,000 at
10 GHz.
Surface Accuracy. Gain is further determined by how
accurately the surface of an antenna is machined or
formed to exactly a parabolic or other selected shape, and
how well the shape is maintained under wind loading,
temperature changes, or other environmental conditions.
A good antenna will see only a narrow beam width and
will be able to pick out a satellite. A poor-quality dish will
see too much extraneous noise and will receive less signal
energy from the satellite of interest and pick up unwanted
energy.
Dish antennas focus on one earth-orbiting satellite at a
time and concentrate the faint signals into a feed horn
(waveguide) that directs the microwave signal into a lownoise amplifier (LNA). The LNA amplifies the weak signal
by 10,000 times and eventually transmits it by cable to the
monitoring location.
Figure 6-37 shows a block diagram of a satellite receiver
system. The LNA is the first active electronic component in
the receiving system that acts on the video signal. The LNA
is analogous to the audio preamplifier in that it provides
the first critical preamplification. Its noise characteristics
generally determine the quality of the final video image
seen on the monitor.
The microwave signal from the LNA is fed via coaxial cable to a down converter which converts the satellite microwave signal to a lower frequency. Since the
signal level is still very low, a special low-loss coaxial
cable must be used and the signal run must be as short
188
CCTV Surveillance
as possible. Increasing cable run decreases signal level,
thereby decreasing the final S/N. The down-converted
microwave signal is eventually converted to VHF or UHF
and displayed on a television receiver or converted to baseband and displayed on a monitor.
Today most satellite receivers generate a base-band signal containing the base-band video and audio and synchronizing information that can be fed directly into a video
monitor or recorder. The receiver also outputs a channel 3 or 4 modulated signal for the input to a standard
television tuner TV set.
6.5.3.3 Interference Sources
Transmission interference occurs when unwanted signals
are received along with the desired satellite signal. Of
the several types of interference, perhaps the most common and irritating is caused by the reception of nearby
microwave signals using the same or adjacent frequency
band. Microwaves reflecting off buildings or even passing
cars are responsible for the interference. Very often, moving the microwave antenna several feet can significantly
reduce the interfering signal levels.
Other interference includes stray signals from adjacent
satellites, or uplink or downlink interference. Finally, a
predictable form of interference is caused by the sun.
Twice a year the sun lines up directly behind each satellite
for periods of approximately ten minutes per day for two
or three days. Since the sun is a source of massive amounts
of radio noise, no transmissions can be received from satellites during these sun outage times. This unavoidable type
of interference can be expected during the normal course
of operation of an earth satellite station.
6.5.4 Radio Frequency Transmission
Radio frequency (RF) is a wireless video transmission
means originally used primarily by government agencies
and amateur radio operators. Government frequencies
include the 1200 MHz (1.2 GHz) and 1700 MHz (1.7 GHz)
bands.
Radio frequency wireless has now found widespread use
in commercial security applications in temporary covert
and permanent surveillance applications. Video transmitters and receivers transmit monochrome or color video
signals over distances of several hundred feet to several
miles using small, portable, battery-operated equipment.
Operating frequencies cover the 150 to 980 MHz, 2.4 GHz,
and 5.8 GHz bands.
While RF transmission provides significant advantages
when a wired system is not possible, there are FCC
restrictions limiting the use of many such transmitters
to government applications. Only low-power transmitters
are available for commercial applications. Any RF systems
used outside the United States require the approval of the
foreign government. Tables 6-7 and 7-5 summarize the
channel frequencies available.
6.5.4.1 Transmission Path Considerations
An RF video signal transmission means can follow either
commercial broadcasting standards in which the visual
signal are amplitude modulated (AM), or noncommercial standards that use an FM signal. In the commercial standard the audio signal is frequency modulated
on the carrier. In both systems the video input signal
ranges from a few hertz to 4.5 MHz. For the low-powered
transmitter/receiver systems used in security applications,
FM modulation has provided far superior performance
(increased range and lack of interference) and is the preferred method. The range obtained with an FM RF transmitter is from three to four times that of the AM type.
Transmitting at standard commercial broadcast video
standards using AM signals and operating on one of the
designated VHF or UHF channels is prohibited by the
FCC since any consumer-type receiver could receive and
display the video picture. This potential is obviously a disadvantage for covert security surveillance. In the case of
FM video transmission, many consumer receivers, though
not designed to receive such signals, do display FM signals
with some degree of picture quality because of nonlinear
and sporadic operation of various receiver circuits. Likewise, the FCC does not permit the commercial use of FM
or other modulation techniques in the commercial VHF
and UHF channels.
Low-power RF transmission in the 902–928 MHz,
2.4 GHz and 5.8 GHz ranges have been approved for general security applications without an FCC license. The 1.2
and 1.7 GHz bands have not been approved for commercial use.
6.5.4.2 Radio Frequency Equipment
Many manufacturers produce wireless video RF and
microwave links operating in the 900 MHz, 2.4 GHz, and
5.8 GHz frequency bands. This equipment operates on
FCC-assigned frequencies with specific maximum transmitter output power levels (a few hundred milliwatts). These
general-purpose RF links operate at output field strengths
50–250 milliwatts per meter at 3 meters (Part 15 of the
FCC specification).
Figure 6-38 illustrates typical RF and microwave video
transmission equipments. Figure 6-38a shows two very
small, four-channel, low power 1.2 GHz (government use
only) and 2.4 GHz transmitters. Any one of four channels
can be selected at a time. Figure 6-38b is a small four
channel 2.4 GHz transmitter and receiver pair using highgain Yaggi antennas for increased range and directionality.
Figure 6-38c shows a long range 2.4 GHz receiver.
Analog Video, Voice, and Control Signal Transmission
COMMERCIAL
TELEVISION
CHANNELS
BAND
FREQUENCY
RANGE (MHz)
USAGE
RESTRICTIONS
VHF—LOWBAND
2–6
54–88
LOW-MEDIUM POWER
SEVERAL MILES RANGE
GOVERNMENT, LAW
ENFORCEMENT ONLY
FM RADIO
—
88–108
COMMERCIAL RADIO
FCC REGULATED
VHF—HIGHBAND
7–13
174–216
LOW-MEDIUM POWER, RANGE
UP TO SEVERAL MILES
GOVERNMENT, LAW
ENFORCEMENT ONLY
SECURITY
—
350–950
SINGLE CHANNEL
TRANSMITTER/RECEIVER
GOVERNMENT, LAW
ENFORCEMENT ONLY
UHF
14–83
470–890
LOW-MEDIUM POWER, RANGE
UP TO SEVERAL MILES
GOVERNMENT, LAW
ENFORCEMENT ONLY
SECURITY
—
902–928
LOW POWER, FCC PART 15
NO RESTRICTIONS, NO FCC
LICENSE REQUIRED
SECURITY
—
1.2–1.7 GHz
SECURITY
—
2.4 GHz
LOW POWER, FCC PART 15 *
SECURITY
—
2.4 GHz
HIGH POWER, FCC PART 90 **
SECURITY
—
5.8 GHz
LOW-MEDIUM POWER
LOW POWER
SEVERAL MILES RANGE
189
GOVERNMENT, LAW
ENFORCEMENT ONLY
NO RESTRICTIONS, NO FCC
LICENSE REQUIRED
GOVERNMENT, LAW
ENFORCEMENT ONLY
NO RESTRICTIONS, NO FCC
LICENSE REQUIRED
ALL SECURITY FREQUENCY BANDS ARE OUTSIDE THE COMMERCIAL TELEVISION BANDS
* INDUSTRIAL, SECURITY, MEDICAL (ISM)
**FCC PART 90, 5 WATT MAXIMUM
Table 6-7
RF and Microwave Video Transmission Frequencies
For indoor applications, most RF and microwave transmitter/receiver systems use omnidirectional dipole antennas for ease of operation. For outdoor operation, dipoles
or whip antennas are used. High-gain Yaggi antennas are
used to increase range and minimize interference from
other radiation sources.
The RF and microwave transmitters and receivers have a
standard 75-ohm input impedance; however, they require
a 50-ohm coaxial cable at the transmitter output and the
receiver input. Using a 75-ohm coaxial cable between the
antenna and the transmitter output or the receiver input
will seriously degrade the performance of the system even
if it is short (1–2 ft). Miniature 50-ohm, RG58U, and RG8
coaxial cables terminated in a small SMA or BNC connector are used.
Figure 6-39 shows the approximate distance between
transmitter and receiver antennas (range) versus transmitted power, for video transmission. The range values are
for smooth and obstacle-free terrain applications using
a dipole antenna at the transmitter and receiver. The
antennas should be located as high above the ground as
possible.
The numbers obtained should be used as a guide only.
Actual installation and experience with specific equipment
on-site will determine the actual quality of the video image
received.
6.5.5 Infrared Atmospheric Transmission
A technique for transmitting a video signal by wireless
means uses propagation of an IR beam of light through
the atmosphere (Figure 6-40).
The light beam is generated by either an LED or an
ILD in the transmitter. The receiver in the optical communication link uses a silicon-diode IR detector, amplifier,
and output circuitry to drive the 75-ohm coaxial cable and
monitor. The transmitter-to-receiver distance and security
requirements of the link determine the type of diode used.
Short-range transmissions of up to several hundred feet
are accomplished using LED. To obtain good results for
longer ranges, up to several miles under clear atmospheric
conditions, ILD must be used.
The LED system costs less and has a wider beam, 10–20
wide, making it relatively simple to align the transmitter
and receiver. The beam width of a typical ILD transmitter
is 01 or 02 , making it more difficult to align and requiring that the mounting structure for both transmitter and
the receiver be very stable in order to maintain alignment.
To insure a good, stable signal strength at the receiver, the
ILD transmitter and receiver must be securely mounted
on the building structure. Additionally, the building structure must not sway, creep, vibrate, or produce appreciable
twist due to uneven thermal heating (sun loading).
190
CCTV Surveillance
(A) MINIATURE TRANSMITTER
WITH DIPOLE ANTENNA
1.2 GHz, 2.4 GHz.
2.4 GHz
4 CHANNEL
4 CHANNEL
POWER OUT: 100 mW
POWER OUT: 0.5 W
FIGURE 6-38
(C) RECEIVER WITH
DIPOLE ANTENNA
(B) SMALL TRANSMITTER AND
RECEIVER WITH YAGGI ANTENNA
2.4 GHz
RF and microwave video transmitters
POWER OUT
(WATTS)
10.0
1.0
0.1
TYPICAL
RANGE
0
0.1
1.0
10.0
TRANSMISSION CONDITIONS: CLEAR AIR, OUTDOOR, NO OBSTRUCTIONS, DIPOLE ANTENNA
FIGURE 6-39
Transmitter RF power out vs. transmission range
(Mi)
Analog Video, Voice, and Control Signal Transmission
DETECTOR RECEIVER
FIELD OF VIEW (FOV)
LED IR
TRANSMITTER
191
IR
RECEIVER
LED BEAM
DIVERGENCE
(10−20°)
BUILDING
BUILDING
CAMERA
MONITOR
ILD IR
TRANSMITTER
ILD
DIVERGENCE
(0.1−0.2°)
DETECTOR RECEIVER
FIELD OF VIEW (FOV)
BUILDING
IR
RECEIVER
BUILDING
FIGURE 6-40
IR atmospheric video transmission system
Both LED and ILD systems can transmit the IR
beam through most transparent window glazing; however,
glazing with high tin content severely decreases signal
transmission, thereby producing poor video quality. The
suitability of the window can be determined only by testing the system. Since many applications require the IR
beam to pass through window panes across a city street
or between two buildings, window IR transmission tests
should be performed prior to designing and installing
such a system.
The primary advantages of the ILD system are longrange (under clear atmospheric conditions) and secure
video, audio, and control signal transmission. ILD atmospheric links are hard to tap because the tapping device—a
laser receiver—must be positioned into the laser beam,
which is hard to accomplish undetected.
6.5.5.1 Transmission Path Considerations
Several transmission parameters must be considered in
any atmospheric transmission link. Both LED and ILD
atmospheric transmission s ystems suffer video signal transmission losses caused by atmosphere path absorption.
Molecular absorption is always present when a light beam
travels through a gas (air). At certain wavelengths of light,
the absorption in the air is so great as to make that wavelength useless for communications purposes. Wavelength
ranges in which the attenuation by absorption is tolerable are called atmospheric windows. These windows have
been extensively tabulated in the literature. All LED and
ILD systems operate in these atmospheric windows.
Another cause of light signal absorption is particles such
as dust and aerosols, which are always present in the atmosphere to some degree. These particles may reach very
high concentrations in a geographical area near a body of
water. In these locations, improved performance can be
achieved by locating the link as high above the ground as
possible.
Fog is a third factor causing severe absorption of the IR
signal. In fog-prone areas, local weather conditions must
be considered when specifying an atmospheric link, since
the presence of fog will greatly influence link downtime.
Figure 6-41 shows the predicted communication range vs.
visibility for a practical LED or ILD atmospheric communications system.
192
CCTV Surveillance
ATMOSPHERIC LOSS FACTORS
ABSORPTION
SCATTERING
SCINTILLATION
AEROSOLS
ATMOSPHERIC
SIGNAL
LOSS
dB
0
–10
–20
–30
–40
–50
–60
–70
.5
FIGURE 6-41
1
2
3
4
5
DISTANCE
(Mi)
Atmospheric absorption factors and visibility
In addition to signal loss, the atmosphere contributes
signal noise, since it exhibits some degree of turbulence.
Turbulence causes a refractive index variation in the signal path (similar to the heat waves seen when there is
solar heating in air—the mirage effect) and its subsequent
wind-aided turbulent mixing. The net effect of this turbulence is to move or bend the IR beam in an unpredictable
direction, so that the transmitter radiation does not reach
the remote receiver. To compensate for this turbulence,
the transmitter beam is made wide enough so that it is
highly unlikely that the beam will miss the receiver. This
wider beam, however, results in lower beam intensity, so
the received signal on average will be less than from a
narrower beam.
6.5.5.2 Infrared Equipment
The transmitter and receiver used in atmospheric IR
transmission systems are very similar to those used in
the fiber-optic-cable transmission system (Section 6.3.4).
The primary differences are in the type of LED (or ILD)
in the transmitter and the optics in both the transmitter
and the receiver (Figure 6-42).
The optics in the transmitter must couple the maximum amount of light from the emitter into the lens and
atmosphere, that is, to produce the specified beam divergence depending on LED or ILD usage. The receiver
optics are made as large as practically possible (several
inches in diameter) to maximize transmitter beam col-
lection, thereby achieving the highest possible S/N. An
example of an atmospheric IR link is shown in Figure 6-43.
The system has a range of approximately 3000 feet and
operates at 12 volts DC. For outdoor applications, the
transmitter is mounted in an environmental housing with
a thermostatically controlled heater and fan, as well as a
window washer and wiper.
6.6 WIRELESS CONTROL SIGNAL
TRANSMISSION
Signal multiplexing has been used to combine audio and
control functions in time-division or frequency-division
multiplexing. One system uses the telephone Touch-Tone
system, which is standard throughout the world. With this
system, an encoder generates a number code corresponding to the given switch (digit) closure. Each switch closure produces a dual Touch-Tone signal which is uniquely
defined and recognized by the remote receiver station. All
that is needed for transmitting the signal is a twisted-pair
or telephone-grade line. With such a system, audio and all
of the conceivable camera functions (pan, tilt, zoom focus,
on/off, sequencing, and others) can be controlled with
a single cable pair or single transmission channel. This
concept offers a powerful means for controlling remote
equipment with an existing transmission path.
It is sometimes advantageous to combine several video
and/or audio and control signals onto one transmission
Analog Video, Voice, and Control Signal Transmission
RECEIVER
TRANSMITTER
LIGHT EMITTING DIODE
(LED)
ATMOSPHERIC
PATH
SILICON DETECTOR
LENS
LENS
VIDEO
AMPLIFIER
LOWNOISE
AMPLIFIER
LED
DRIVER
VIDEO
AMPLIFIER
AND DRIVER
LENS
LED
193
LENS
Si
DETECTOR
ILD
LENS
DIVERGING
BEAM
COLLIMATED
BEAM
FIGURE 6-42
Block diagram of IR video transmitter and receiver
VIDEO SIGNAL
TO NOISE
RATIO (dB)
60
58
56
54
52
50
48
TRANS/REC
DISTANCE
46
(Ff)
1000
2000
3000
4000
5000
6000
7000
VIDEO STANDARD: NTSC, PAL, SECAM (525 TV LINES, 60 Hz OR 625 TV LINES, 50 Hz)
MONOCHROME, COLOR
RANGE: 1 MILE +
LICENSE REQUIREMENT: NONE
TYPE: SIMPLEX (ONE DIRECTION)
SIGNAL BANDWIDTH: 5.5 MHz ±1dB, 7 MHz ±3 dB
TRANSMITTER: LIGHT EMITTING DIODE (860–900 nm)
PEAK POWER OUTPUT: 30 MW
BEAM DIVERGENCE: 3 MILLIRADIANS
POWER: AC, 115/220 V, 50/60 Hz, 25 VA
DC, 12 VDC, 12 WATTS
FIGURE 6-43
IR video transmitter and receiver hardware
RECEIVER: SILICON AVALANCHE DETECTOR
FIELD OF VIEW: 3.75 MILLIRADIANS
POWER: AC, 115/220 V, 50/60 HZ, 25 VA
DC, 12 VDC, 12 WATTS
194
CCTV Surveillance
channel. This is true when a limited number of cables
are available or when transmission is wireless. If cables are
already in place or a wireless system is required, the hardware to multiplex the various functions onto one channel is cost-effective. Multiplexing of video signals is used
in many CATV installations whereby several VHF and/or
UHF video channels are simultaneously transmitted over
a single coaxial cable or microwave link. In CCTV systems,
modulators and demodulators are available to transmit
the video control signals on the same coaxial cable used
to transmit the video signal.
6.7 SIGNAL MULTIPLEXING/DE-MULTIPLEXING
It is sometimes desirable or necessary to combine several video signals onto one communications channel and
transmit them from the camera location to the monitor location. This technique is called multiplexing. Some
systems allow multiplexing video, control, and audio signals onto one channel.
6.7.1 Wideband Video Signal
The camera video signal is an analog base-band signal with frequencies of up to 6 MHz. When more than
one video signal must be transmitted over a single wire
or wireless channel the signals are multiplexed. This is
accomplished by modulating the base-band camera signal
with an RF (VHF or HF) or microwave frequency carrier and combining the multiple video signals onto the
channel.
6.7.2 Audio and Control Signal
The analog and control signals can be multiplexed with
the video signals as sub-carriers on each of the video
signals. In the RF band no more than two channels at
928 MHz are practical. In the microwave band at 2.4 GHz
up to four channels can be used. At 5.8 GHz up to eight
channels can be used.
6.8 SECURE VIDEO TRANSMISSION
When it comes to protecting the integrity of the information on a signal, high-level security applications sometimes require the scrambling of video signals. The video
scrambler is a privacy device that alters a television
camera output signal to reduce the ability to recognize the transmitted signal when displayed on a standard monitor/receiver. The descrambler device restores
the signal to permit retrieval of the original video
information.
6.8.1 Scrambling
Video scrambling refers to an analog technique to hide,
or make covert, the picture intelligence in the picture
signal. Basic types include: (1) negative video, (2) moving
the horizontal lines, (3) cutting and moving sections of
the horizontal lines, and (4) altering or removing the
synchronization pulses. All negative video requires that the
signal modulator at the camera has some synchronization
with the demodulator at the monitoring site.
The key to any analog video scrambling system is to
modify one or more basic video signal parameters to prevent an ordinary television receiver or monitor from being
able to receive a recognizable picture. The challenges in
scrambling-system design are to make the signal secure
without degrading the picture quality when it is reconstructed, to minimize the increase in bandwidth or storage
requirements for the scrambled signal, and to make the
system cost-effective.
There are basically two classes of scrambling techniques.
The first modifies the signal with a fixed algorithm, that
is, some periodic change in the signal. These systems are
comparatively simple and inexpensive to build and are
common in CATV pay television, as well as in some security
applications. The signals can easily be descrambled once
the scrambling code or technique has been discovered. It
is relatively straightforward to devise and manufacture a
descrambling unit to recover the video signal. One of the
earliest techniques for modifying the standard video signal
is called video inversion, in which the polarity of the video
signal is inverted so that a black-on-white picture appears
white-on-black (Figure 6-44).
While this technique destroys some of the intelligence
in the picture, the content is still recognizable. Some
scrambling systems employ a dynamic video-inversion technique: a parameter such as the polarity is inverted every
few lines or fields in a pseudo-random fashion to make the
image even more unintelligible. Another early technique
was to suppress the vertical and/or horizontal synchronization pulses to cause the picture to roll or tear on
the television monitor. Likewise, this technique produced
some intelligence loss, but some television receivers could
still lock on to the picture, or a descrambler could be
built to re-insert the missing pulses and synchronize the
picture, making it intelligible again.
A second class of scrambling systems using much more
sophisticated techniques modifies the signal with an algorithm that continually changes in some unpredictable or
pseudo-random fashion. These more complex dynamic
scrambler systems require some communication channel
between the transmitter and the receiver in order to provide the descrambling information to the receiver unit,
which reconstructs the missing signal. This descrambling
information is communicated either by some signal transmitted along with the television image or by some separate
Analog Video, Voice, and Control Signal Transmission
195
LINE DICING
SIGNAL INVERSION
97
96
STANDARD
VIDEO
VIDEO
SIGNAL
1 2 34
STANDARD
VIDEO
SYNC
t
SYNC
RANDOMLY
CODED
TRANSPOSED
SEGMENTS
SCRAMBLED
SIGNAL
SEGMENTS
INVERTED
VIDEO
96
4
1
97
SYNC
97
96
SYNC
REASSEMBLED
VIDEO
UNSCRAMBLED
SIGNAL
1 2 34
NEGATIVE
VIDEO
SYNC
FIGURE 6-44
Video scrambling techniques
means, such as a different channel in the link. The decoding signal can be sent by telephone or other means.
In a much more secure technique known as “line dicing,” each horizontal line of the video image is cut into segments that are then transmitted in random order, thereby
displacing the different segments horizontally into new
locations (Figure 6-44). A picture so constructed on a standard receiver has no intelligence whatsoever. Related to
line dicing is a technique known as “line shuffling,” in
which the scan lines of the video signal are sent not in
the normal top-to-bottom image format but in a pseudorandom or unpredictable sequence.
It is often necessary to scramble the audio signal in
addition to the video signal, using techniques such as frequency hopping adapted from military technology. Similar to line dicing, this technique breaks up the audio
signal into many different bits coming from four or five
different audio channels and by jumping from one to
another in a pseudo-random fashion scrambles the audio
signal. The descrambler is equipped to tune to the different audio channels in synchronism with the transmitting
signal, thereby recovering the audio information.
In the most sophisticated dynamic scrambling systems,
utilized for direct-broadcast satellites and multi-channel
applications, the video and audio signals are scrambled
in a way that cannot be decoded even by the equipment
manufacturer without the information from the signal
operator. For example, the audio signal can be digitized
and then transmitted in the vertical blanking interval, the
horizontal blanking interval, or on a separate sub-carrier
of the television signal.
6.8.2 Encryption
Video encryption refers to digitizing and coding the video
signal at the camera using a computer and then decoding the digitized signal at the receiver location with the
corresponding digital decoder. Digital encryption results
in a much higher level of security than analog scrambling. Section 7.7.4 analyzes digital encryption techniques
in more detail.
6.9 CABLE TELEVISION
Cable television (CATV) systems distribute multiple channels of video in the VHF or UHF bands using coaxial cable, fiber-optic cable, and RF and microwave
links. Consumer-based CATV employs this modulation–
demodulation scheme using a coaxial or fiber-optic cable.
The multiplexing technique is often used when video
information from a large number of cameras must be
196
CCTV Surveillance
transmitted to a large number of receivers in a network.
Table 6-8 summarizes the VHF and UHF television frequencies used in these CATV RF transmission systems.
In CATV distribution systems, the equipment accepts
base-band (composite video) and audio channels and linearly modulates them to any user-selected RF carrier in
the UHF (470–770 MHz) spectrum. The modulated signal
is then passed through an isolating combiner, where they
are multiplexed with the other signals. The combined signal is then transmitted over a communications channel
and separated at the receiver end into individual video
and audio information channels. At the receiver end the
signal is demodulated and the multiple camera signals are
separated and presented on multiple monitors or switched
one at a time (Figure 6-45).
Cable costs are significantly reduced by modulating multiple channels on a single cable. Since the transmission is
done at radio frequencies, design and installation is far
more critical as compared with base-band CCTV. Highquality CATV systems are now installed with fiber-optic
cable for medium to long distances or distribution within
a building.
6.10 ANALOG TRANSMISSION CHECKLIST
Transmitting the video, audio, and control signals faithfully is all important in any security system. This section
itemizes some of the factors that should be considered in
any design and analysis.
6.10.1 Wired Transmission
The following checklists for coaxial two-wire UTP, and
fiber-optic cable transmission systems show some items
that should be considered when designing and installing
a video security project.
6.10.1.1 Coaxial Cable
1. When using coaxial cable, always terminate all
unused inputs and unused outputs in their respective
impedances.
2. When calculating coaxial-cable attenuation, always
figure the attenuation at the highest frequency to be
used; that is, when working with a 6 MHz bandwidth,
refer to the cable losses at 6 MHz.
3. In long cable runs do not use an excessive number
of connectors since each conductor causes additional
attenuation. Avoid splicing coaxial cables without the
use of proper connectors, since incorrect splices cause
higher attenuation and can cause severe reflection of
the signal and thus distortion.
CHANNEL
DESIGNATION
BAND
FREQUENCY RANGE
PICTURE CARRIER
(MHz)
CATV LOW-BAND
2
CH2
6
CH6
55.25
67.25
CATV MID-BAND
A5
CH95
A1
CH99
91.25
115.25
CATV HIGH-BAND
CH23
CH36
175.25
211.25
CATV MID-BAND
A
CH14
I
CH22
121.25
169.25
217.25
295.25
301.25
547.25
7
CATV SUPER-BAND
CATV HYPER-BAND
J
13
W
CH23
CH26
AA
CH37
PPP
CH78
NOTE:
AIRWAVE VHF TV CHANNELS 2–6
OPERATE FROM 55.25–83.25 MHz
AIRWAVE UHF TV CHANNELS 7–13
OPERATE FROM 175.25–211.25 MHz
AIRWAVE FM STATIONS OPERATE
FROM 88.1–107.9 MHz
Table 6-8
Allocated CATV RF Transmission Frequencies
Analog Video, Voice, and Control Signal Transmission
C1
C2
•
•
•
MULTIPLEXER
COMBINES
CAMERA
AUDIO/ VIDEO
SIGNALS
197
DE-MULTIPLEXER
MICROWAVE
CATV (RF)
FIBER OPTIC
SEPARATES
CAMERA
AUDIO/ VIDEO
SIGNALS
DE-MULTIPLEXER
MULTIPLEXER
SWITCHER
CN
ALL SIGNALS
TRANSMITTED
OVER ONE
WIDEBAND
VIDEO
TRANSMISSION
LINK
MULTIPLE
VIDEO
CAMERAS
M1
OUTPUT DEVICES
• SWITCHER
• MONITOR
• DVR / VCR
• PRINTER
M2
o
o
o
MN
FIGURE 6-45
MULTIPLE
MONITORS
AUDIO/ VIDEO
Multiplexed video transmission system
4. For outdoor applications, be sure that all connectors are
waterproof and weatherproof; many are not, so consult
the manufacturer.
5. Try to anticipate ground loop problems if unbalancedcoaxial-cable video runs between two power sources are
used. Use fiber optics to avoid the problem.
6. Using a balanced coaxial cable (or fiber-optic cable) is
usually worth the increased cost in long transmission
systems. When connecting long cable runs between several buildings or power sources, measure the voltage
before attempting to mate the cable connectors. Be
careful, since the voltage between the cable and the
connected equipment may be of sufficient potential to
harm you.
7. Do not run cable lines adjacent to high-power RF
sources such as power lines, heavy electrical equipment,
other RFI sources, or electromagnetic sources. Good
earth ground is essential when working with long transmission lines. Be sure that there is adequate grounding,
and that the ground wire is eventually connected to a
water pipe ground.
6.10.1.2 Two-Wire UTP
1. Choose a two-wire twisted-pair having approximately 1
twist for 1–2 inches of wire.
2. Choose a wire gauge between 24 AWG (smallest) and
16 AWG.
3. Choose a reputable UTP transmitter/receiver manufacture. Either have the manufacture supply technical
specifications showing performance over the distance
required or test the product first.
4. Will the UTP transmitter/receiver be powered from the
camera or separate 12 VDC power supply?
6.10.1.3 Fiber-Optic Cable
1. Consider the use of fiber optics when the distance
between camera and monitor is more than a few hundred feet (depending on the environment), if it is a color
system, and if the camera and monitor are in different
buildings or powered by different AC power sources.
2. If the cable is outdoors and above ground, use fiber
optics to avoid atmospheric disturbance from lightning.
3. If the cable run is through a hazardous chemical or
electrical area, use fiber optics.
4. Use fiber optics when a high-security link is required.
6.10.2 Wireless Transmission
The following checklists for RF, microwave, and IR transmission systems should be considered when designing and
198
CCTV Surveillance
installing a video security. The wireless video transmission
techniques require more careful scrutiny because of the
many variables that can influence the overall performance
and success of the system.
as other building materials. Microwave energy does
not transmit through metal objects and only partially
through nonmetal.
7. For outdoor transmission, obstructions such as trees,
buildings, etc. must be considered.
6.10.2.1 Radio Frequency (RF)
1. How many channels does the system require? At
928 MHz the maximum number of channels is two to
avoid crosstalk between channels.
2. Are all cameras approximately in the same location? If
in different locations the crosstalk is minimized.
3. RF transmission is susceptible to external electrical interference. Are there probable interferences in
the area?
4. Range is a function of the transmitter power and intervening atmosphere and objects (buildings, trees, etc.).
Is there a line of sight between the transmitter and the
receiver?
5. Is the transmission path indoors or outdoors?
6. For indoor transmission, reflection and absorption of
all objects must be taken into consideration. RF transmission does not penetrate metal objects and is partially
absorbed by other materials.
7. For outdoor transmission, obstructions such as trees,
buildings, etc. must be considered.
6.10.2.2 Microwave
1. How many channels does the system require? At 2.4 GHz
the maximum number of channels is four and for
5.8 GHz is 8 if crosstalk between channels is to be avoided.
2. Are all cameras approximately in the same location? If
they are in different locations the crosstalk is minimized.
3. Microwave transmission is susceptible to external interference from other microwave or noise sources. Are
there probable interferences in the transmission path?
4. Range is a function of the transmitter power and intervening atmosphere and objects (buildings trees, etc.).
Is there a line of sight between the transmitter and
receiver? If not metal panels can be used to redirect
the microwave transmission.
5. Is the transmission path indoor or outdoor?
6. For indoor transmission, reflection and absorption of
metal objects must be taken into consideration, as well
6.10.2.3 Infrared
1. Infrared transmission is very sensitive to the intervening
atmosphere. Dust, fog, and humidity play an important role in the transmission and cause absorption and
scattering of the IR signal.
2. Is there a line of sight between the IR transmitter and
receiver; no obstructions?
3. Can a mirror be used to “see around a corner”?
4. Are the transmitter (most important) and receiver units
mounted on a sturdy nonvibrating mounting. Are the
buildings stable and motionless under high wind, and
over full sun loading conditions?
5. Is secure transmission needed?
6.11 SUMMARY
Video signal transmission is a key component in any CCTV
installation. Success requires a good understanding of
transmission systems.
Most systems use coaxial cable, but fiber-optic cable is
gaining acceptance because of its better picture quality
(particularly with color) and lower risk factor with respect
to ground loops and electrical interference. In special
situations where coaxial or fiber-optic cable is inappropriate, other wired or wireless means are used, such as
RF, microwave, or light-wave transmission. For very long
range applications, non-real-time slow-scan systems are
appropriate.
Many security system designers consider cabling to
be less important than choosing the camera and lens
and other monitoring equipment in a CCTV application.
Often they attempt to cut costs on cabling equipment and
installation time, since they often make up a large fraction of the total system cost. Such equipment is not visible
and can seem like an unimportant accessory. However,
such cost-cutting can drastically weaken the overall system
performance and picture quality.
Chapter 7
Digital Transmission—Video,
Communications, Control
CONTENTS
7.1
7.2
7.3
Overview
7.1.1 Migration from Analog to Digital
7.1.2 Local Area Network (LAN), Wide Area
Network (WAN), Wireless LAN (WiFi)
7.1.3 Internet
7.1.4 Wireless 802.11, Spread Spectrum
Modulation (SSM)
7.1.5 Digital Video Recorder (DVR), Network
DVR (NDVR)
7.1.6 Network Security, Hackers, Viruses,
Reliability
Communication Channels
7.2.1 Wired Channels
7.2.1.1 Local Area Network (LAN)
7.2.1.2 Power over Ethernet (PoE)
7.2.1.3 Wide Area Network (WAN)
7.2.1.4 Internet, World Wide Web (WWW)
7.2.1.5 Leased Land Lines, DSL, Cable
7.2.1.5.1 PSTN-ISDN Link
7.2.1.5.2 DSL Link
7.2.1.5.3 T1 and T3 Links
7.2.1.5.4 Cable
7.2.1.6 Fiber Optic
7.2.2 Wireless Channels
7.2.2.1 Wireless LAN (WLAN, WiFi)
7.2.2.2 Mesh Network
7.2.2.3 Multiple Input/Multiple Output
(MIMO)
7.2.2.4 Environmental Factors:
Indoor–Outdoor
7.2.2.5 Broadband Microwave
7.2.2.6 Infrared (IR)
Video Image Quality
7.3.1 Quality of Service (QoS)
7.3.2 Resolution vs. Frame Rate
7.3.3 Picture Integrity, Dropout
7.4
Video Signal Compression
7.4.1 Lossless Compression
7.4.2 Lossy Compression
7.4.2.1 Direct Cosine Transform (DCT)
7.4.2.2 Discrete Wavelet Transform (DWT)
7.4.3 Video Compression Algorithms
7.4.3.1 Joint Picture Experts Group: JPEG
7.4.3.2 Moving—Joint Picture Experts
Group: M-JPEG
7.4.3.3 Moving Picture Experts Group:
MPEG-2, MPEG-4, MPEG-4 Visual
7.4.3.3.1 MPEG-2 Standard
7.4.3.3.2 MPEG-4 Standard
7.4.3.3.3 MPEG-4 Visual Standard
7.4.3.4 MPEG-4 Advanced Video Coding
(AVC)/H.264
7.4.3.5 JPEG 2000, Wavelet
7.4.3.6 Other Compression Methods:
H.263, SMICT
7.4.3.6.1 H.263 Standard
7.4.3.6.2 SMICT Standard
7.5 Internet-Based Remote Video Monitoring—Network
Configurations
7.5.1 Point to Multi-Point
7.5.2 Point to Point
7.5.3 Multi-Point to Point
7.5.4 Video Unicast and Multicast
7.6 Transmission Technology Protocols: WiFi, Spread
Spectrum Modulation (SSM)
7.6.1 Spread Spectrum Modulation (SSM)
7.6.1.1 Background
7.6.1.2 Frequency Hopping Spread
Spectrum Technology (FHSS)
7.6.1.3 Slow Hoppers
7.6.1.4 Fast Hoppers
199
200
CCTV Surveillance
7.6.1.5
7.7
7.8
7.9
7.10
7.11
7.12
Direct Sequence Spread Spectrum
(DSSS)
7.6.2 WiFi Protocol: 802.11 Standards
7.6.2.1 802.11b Standard
7.6.2.2 802.11a Standard
7.6.2.3 802.11g Standard
7.6.2.4 802.11n Standard
7.6.2.5 802.11i Standard
7.6.3 Asynchronous Transfer Mode (ATM)
Transmission Network Security
7.7.1 Wired Equivalent Privacy (WEP)
7.7.2 Virtual Private Network (VPN)
7.7.3 WiFi Protected Access (WPA)
7.7.4 Advanced Encryption Standard (AES),
Digital Encryption Standard (DES)
7.7.5 Firewalls, Viruses, Hackers
Internet Protocol Network Camera, Address
7.8.1 Internet Protocol Network Camera
7.8.2 Internet Protocol Camera Protocols
7.8.3 Internet Protocol Camera Address
Video Server, Router, Switch
7.9.1 Video Server
7.9.2 Video Router/Access Point
7.9.3 Video Switch
Personal Computer, Laptop, PDA, Cell Phone
7.10.1 Personal Computer, Laptop
7.10.2 Personal Digital Assistant (PDA)
7.10.3 Cell Phone
Internet Protocol Surveillance Systems:
Features, Checklist, Pros, Cons
7.11.1 Features
7.11.2 Checklist
7.11.3 Pros
7.11.4 Cons
Summary
7.1 OVERVIEW
7.1.1 Migration from Analog to Digital
The video security industry is migrating from a technology
of CCTV to open circuit television (OCTV) and Automated Video Surveillance (AVS). The OCTV and the AVS
technologies make use of networked digital surveillance
and digital surveillance systems. There is little doubt that
connecting all video cameras directly to a digital video
network is becoming commonplace and cost effective in
new and existing systems. Classes of video applications
using these networking technologies to advantage are:
(1) remote video surveillance, (2) remote video alarm
verification, (3) rapid deployment video and alarm systems, and (4) remote access to stored digital video images.
The OCTV permits multiple security operators to manage
many remote facilities, and allows almost instantaneous
monitoring of remote sites via these digital networks. Systems using existing analog video cameras can connect to
the Internet via digital servers thereby providing remote
site surveillance and camera control. The AVS is achieved
through the use of smart cameras that can “learn,” and the
use of other “intelligent” algorithms and electronics that
make decisions based on past experience. This “artificial
intelligence” significantly reduces the number of decisions
the guard must make.
The fastest-growing market segment in the video security field is digital video surveillance. The security industry
is rapidly moving toward AVS in which smart cameras and
sensors “learn” and make decisions and provide the security officer with enough information to act.
Prior to the year 2001, camera systems were primarily
used to catch the bad guys after a crime had been committed. If a large competent well-trained security team was
available, the thief or criminal could be caught in the act.
The primary video surveillance functions were to:
•
•
•
•
•
•
Catch perpetrators
Watch workers
Protect from litigation
Watch a perimeter of the facility
Monitor traffic
Protect assets.
With more sophisticated analog video systems and the
migration to wired and wireless digital local area networks
(LAN), intranets, and Internet networks, the security system provided additional functions to:
• Monitor suspicious activities to prevent illegal activity
• Identify and apprehend perpetrators of a crime
• All the other activities listed above.
Historically CCTV systems were closed and proprietary
networks that were controlled by the security manager.
Now analog video systems, access control, intrusion detection, fire, safety, environmental sensors, and control and
communication systems are often open and video images
and information are sent over digital networks to multiple managers and multiple sites. From an economic
point of view it makes sense to have all these sensors distributed throughout a facility or enterprise and monitored
by multiple managers and facilitators. The video security
requirements are now often added to the backbone of the
information technology (IT) structure. This is in contrast
to the analog CCTV methodology that requires individual
video feeds connected to a security console with dedicated
monitors and recorders and printers that do not operate
on a local digital network, an intranet, or the Internet.
The full impact of video surveillance using wireless cameras, monitors, and servers has yet to be realized. Wireless
video surveillance is rapidly growing in popularity for
monitoring remote locations whether from a personal
computer (PC), laptop, personal digital assistant (PDA),
or cell phone.
Digital Transmission—Video, Communications, Control
Remote video surveillance systems have three main functions: (1) recording the surveillance camera image, (2)
playback of the surveillance image and search of specific
event stored video, and (3) remote control of security
equipment. The first step in the transmission process for
remote video surveillance occurs when the cameras capture visual images from the surveillance area. The cameras (the input terminal) view the target areas, compress
the video signals, and transmit them via a transmission
means. The monitoring location(s) or control terminal
receives the signals and de-compresses them back into
visual images, usually achieving near real-time transmission and viewing of them. In an analog system this process
involves converting the input signals from analog to digital form and then back to analog form for display on
a video monitor, and/or recording on an analog VCR.
The video signal is left in digital form when it is recorded
on a digital DVR and displayed on an LCD, plasma, or
other digital monitor. Networked transmission allows the
user to remotely adjust the P/T/Z, focus and aperture
(iris diaphragm) settings of the camera at any time from
the remote monitoring location. Video monitoring is simplified through the use of digital video motion detectors
(DVMDs) and smart cameras. Simultaneous monitoring
and control from multiple geographical locations is often
required. The video security industry is experiencing revolutionary changes brought upon by digital information
technology (IT). This shift in video security from analog
to digital began when the analog VCR was replaced by
the DVR.
The recent phase of this technology has advanced to the
utilization of wired and wireless IT systems and networks.
Video systems are expected to be full-time: 24/7/365 video
surveillance, voice communications, and control.
7.1.2 Local Area Network (LAN), Wide Area
Network (WAN), Wireless LAN (WiFi)
The digital signal transmission channels now available
include local area network (LAN), wide area network
(WAN), wireless LAN (WLAN, WiFi), intranet, Internet,
and World Wide Web (WWW).
7.1.3 Internet
At the core of remote monitoring is a basic network
infrastructure exemplified by network cameras, video
servers, and computers. All these equipments communicate via a standard called the Internet protocol (IP). The
IP is the ideal solution for remote monitoring since it
allows users to connect and manage video, audio, data,
control PTZ, and other communications over a single network that is accessible to users anywhere in the world. This
data is available in most cases by a standard Web browser
201
and Internet access that can be found on any desktop PC,
laptop, and many PDAs and cell phones.
Video servers include an analog camera video input, an
image digitizer, an image compressor, a Web server and
network connection. The servers digitize the video from
the analog cameras and transmit them over the computer
network, essentially turning an analog camera into a network camera.
7.1.4 Wireless 802.11, Spread Spectrum
Modulation (SSM)
A key component to the digital transmission means is
a technology called spread spectrum modulation (SSM).
In this type of modulation a transmission code is combined with the information carrying base-band video signal
and transmitted over the wireless network. The effect of
“spreading” the signal over a wide spectrum of bandwidth
provides the ability to transmit many different signals in
the same allotted bandwidth with high security. This SSM
communication has long been a favorite technology of the
military because of its resistance to interception and jamming and was adopted in the Institute of Electrical and
Electronic Engineers (IEEE) 802.11 series of transmission
standards for digital transmission applications including
digital video. The subsets of 802.11 applicable to video
transmission are 802.11a, b, c, g, i, and the new n. The
SSM technology is used in digital cellular phones, some
advanced alarm systems, and radar—just to name a few
common applications. The advantages of the technology
include cost, bandwidth efficiency, and security. The SSM
signals are difficult to detect and are therefore difficult to
jam because they produce little or no interference. The
products utilizing this technology operate in a licenseexempt category. There are no charges to the user from
any company or government agency.
7.1.5 Digital Video Recorder (DVR), Network
DVR (NDVR)
The digital video recorder (DVR) has been a significant
innovation in the video security market. It has rapidly
replaced the analog VCR as a means for storing video
images. The DVR using lossy or lossless digital compression
provides the ability to store video images with little or no
degradation. The DVR provides a highly advanced search
capability for looking back at recorded images. The DVR
also incorporates features such as video motion detection,
the ability to have multi-users view the recorded video,
and the ability to perform PTZ control functions from
the monitoring and recording site. The DVR provides a
significant upgrade in image quality and flexibility and
serves as an excellent replacement for the analog VCR.
202
CCTV Surveillance
An alternative to the DVR is the network DVR (NDVR).
This digital Internet solution takes the streaming (realtime) and non-streaming video from cameras and records
them on computers on the network. This makes them
available to anyone having access on the network and
makes use of the storage capability of the network. Advantages of the NDVR on the IP surveillance system over DVR
technology make a strong case for it to be the system of
choice for today’s enterprise-level surveillance solutions.
The wide bandwidth and high information content of
the video signal requires that it be compressed by some
means when transmitted over the network. At present
there are several compression technologies that operate
with wired and wireless digital networks. They each have
their own application areas with advantages and disadvantages. Three formats that are very efficient for video transmission are designated by MPEG-2, MPEG-4, and H.264
developed by the Motion Picture Experts Group, an industry standards committee. These compression standards
permit near real-time transmission of video images with
sufficient resolution and quality for surveillance applications and makes the camera scenes available for remote
observation via Internet browsers.
7.1.6 Network Security, Hackers, Viruses,
Reliability
An important aspect of the digital revolution is that of
security from hackers, viruses, and other adversaries. The
digital system must be safeguarded against these intruders
via password protection, virtual private networks (VPNs),
encryption, and firewalls. Viruses are abundant on the
Internet, and must be guarded against when using a
remote digital monitoring system. The VPN is a private
data network that makes use of the public telecommunication infrastructure, maintaining privacy through the
use of firewall protocols and security procedures. Today
many companies are using a VPN for both extranets and
wide area intranets. Higher levels of security are obtained
through the use of WiFi protected access (WPA), digital
encryption standard (DES), and advanced encryption standard (AES). A firewall is typically located at the boundary
between the Internet and corporate network and controls
access into the network. It also defines who has access outside of the network. The firewall is, in physical terms, the
access control for the network.
As with any form of video networking, keeping the
information safe and error-free is imperative. Errors and
contamination are: man-made, due to an equipment failure, external interference, hackers, or viruses. The security
industry must put forth all efforts to ensure the information is accurate. State-of-the-art image authentication
software has increased the reliability of digital video monitoring by preventing signal tampering. These methods
can be incorporated in special compression codes, using
date/time stamping or the summation of pixel changes.
Demonstrating that the video signal and image has not
been tampered with helps ensure the acceptance of this
information in a court of law.
7.2 COMMUNICATION CHANNELS
This chapter treats all of the video digital transmission
networks including the Internet transmission media with
its unique protocols, standards, signal compression, and
security requirements. It addresses the specific algorithms
required to compress the video frame information image
sizes to make them compatible with the existing bandwidths available in wired and wireless transmission channels. It describes the powerful SSM technology used to
transmit the digital signal and the industry standard 802.11
SSM protocols relating to video, voice, command, and
control transmission.
Digital transmission channels include LAN, WAN, MAN,
WiFi, intranet, Internet, and WWW, transmitted via IP and
viewed through a Web browser.
The most common form of digital transmission suitable for video transmission is the LAN that is traditionally interconnected via a two-wire unshielded twisted-pair
(UTP), coaxial cable, or fiber-optic. When connecting to
multiple sites and remote locations the WAN, MAN, and
WiFi are the transmission means. When cables are difficult or impossible to install, WiFi is used to transmit to all
the different communication devices and locations. The
WiFi serves the same purpose as that of a wired or optical
LAN: it communicates information among the different
devices attached to the LAN but without the use of cables.
When implementing WiFi transmission there is no physical cabling connecting the different devices together from
the monitoring site to the camera locations. These digital
channels use 802.11 with all the different variations of the
standard using the SSM technology.
The primary factors dictating the choice of a network
type for interconnecting different surveillance sites are:
(1) the integrity and guaranteed availability of the network
connection, (2) the availability of a backup signal path,
(3) the data carrying capacity (bandwidth) of the network,
and (4) the operating costs for using the network. Wireless can bring a significant reduction for installation labor
required when running or moving cabling within a building or from building to building.
7.2.1 Wired Channels
Where video monitoring already exists, wired digital video
transmission is accomplished by converting the analog
video signal into a digital signal, and then transmitting
the digitized camera video signal over a suitable network
via modem. At the remote monitoring location a modem
Digital Transmission—Video, Communications, Control
converts the digital video signal back into an analog signal. Customers can use their existing telephone service
to transmit the video signal. The systems used in the
1980s and early 1990s were generally referred to as slowscan video transmission (Chapter 6). The video equipment
often interfaces with alarm intrusion sensors to produce an
alarm signal and the video images serve as an assessment
of the alarm intrusion.
Wired digital video transmission works especially well
in panic alarm situations where a remote location is connected to a central alarm station. If an alarm at a remote
location is activated or if a person initiates an alarm with
a panic button, a video clip from the camera prior to the
alarm, during the alarm, and after the alarm at the remote
location is sent to the monitoring station. The operator
at the central-station is able to forward the video clip to
the police, who now are prepared for what the situation
is, how many people were involved, and if there were any
weapons. The police can use the video clip to identify and
apprehend them.
These systems use the dial-up or public switched telephone network (PSTN) sometimes referred to as the plain
old telephone service (POTS) and both are still a common transmitting means. Since the telephone service was
designed for the human voice it is not very suitable for
high-speed, wide bandwidth video transmission. The wired
phone system has a maximum bandwidth of 3000 Hz and
a maximum modem bit rate of 56 Kbps. However, only
about 40 Kbps is normally realized. A slightly improved
version of PSTN is the integrated services data network
(ISDN) that gives direct access to digital data transmission
at data rates of 64 Kbps.
Since many corporations have already set up LAN/WAN
networking systems for IT business applications, the next
logical expansion is to these networks for complete integration of video surveillance. A major advantage of IPaddressed, network-capable video devices is the ability
to receive a signal anywhere using equipment ranging
from a simple network Internet browser to special clientbased application software products. The high bandwidth
requirements for full-frame high-quality video without
compression exceed the capability of most WAN network
connections. On average, a low-quality image transmitted
via networks requires 256 Kbps and can reach 1 Mbps if
image quality and refresh rates are increased. Even LANs
would be strained as large numbers of cameras attempted
to simultaneously pass video signals back to a central video
server, DVR, or both.
7.2.1.1 Local Area Network (LAN)
The most common and most extensively installed LAN is
the Ethernet. This network is specified in the IEEE 802.3
standard and was originally developed by the Xerox Corporation, and further developed by Xerox, DEC, and Intel
corporations. The typical Ethernet LAN uses a coaxial
203
cable or special grade of twisted-pair wires for transmission
(Figure 7-1).
For long ranges or transmission through areas having
electrical interference, the Ethernet can use fiber-optic
transmission technology. The most common lowest bandwidth Ethernet systems is called 10BASE-T and can
provide transmission speeds up to 10 Mbps. For fast Ethernet connections a 100BASE-T is used and provides
speeds up to 100 Mbps. Gigabit Ethernet systems provide an even higher speed of transmission of 1000 Mbps
(1 Gigabit = one billion bits per second). The latter two
are used as the backbone for digital transmission systems.
Video systems generally use the 10Base-T or 100Base-T
networks. WANs connect LANs to form a large structured
network sometimes called an intranet. These networks can
be connected inside buildings and from building to building, and connected to the Internet.
7.2.1.2 Power over Ethernet (PoE)
The PoE, also referred to as power over LAN (PoL) is a
technology that integrates data and power over standard
LAN infrastructure cabled networks (Figure 7-2).
The PoE is a means to supply reliable, uninterrupted
power to network cameras, wireless LAN access points,
and other Ethernet devices using existing, commonly used
category (CAT) cable with four twisted pair conductors
and CAT5 cable infrastructure (Figure 7-3).
The PoE is a technology for wired Ethernet LANs that
allows the electrical power (current and voltage) necessary for the operation of each device to be carried by the
data cables rather than by power cords. This minimizes
the number of wires that must be strung in order to install
the network. The result is lower cost, less downtime, easier
maintenance, and greater installation flexibility than with
traditional wiring. Unlike a traditional telephone infrastructure, local power is not always accessible for wireless
access points, IP video cameras, phones, or other network
devices deployed in ceilings, lobbies, stairwells, or other
obscure areas. Adding new wiring for power may be a difficult and costly option. In cases like this, an option is to
combine the provision of power with the network connection using PoE technology over any existing or new data
communications cabling.
The standard was developed by the IEEE as 802.3af. The
standard Ethernet cable uses only two of those pairs for
10BaseT or 100BaseT transmission. Because the Ethernet
data pairs are transformer-coupled at each end of the
cable, either the spare pairs or the data pairs can be used
to power powered-device (PD) equipment. At the power
source end of the cable, the power source equipment may
apply power to either the spare pairs or the data pair of
that cable, but not to both simultaneously. Also the power
source equipment may not apply power to non-PoE devices
if they are connected to the cable. The PoE uses 48 VDC
designated as safety extra low-voltage (SELV) providing
204
CCTV Surveillance
ANALOG
CAMERA
IP CAMERA
SERVER
BNC
RJ45
INTERNET
INTRANET
ETHERNET/IP
NETWORK
10BASE–T
100BASE–T
FIGURE 7-1
Ethernet local area network (LAN)
SECURITY
SWITCH
PoE INJECTOR
CONTROL
NETWORK CAMERAS
DIGITAL VIDEO RECORDER
LEGEND:
VIDEO / DATA
VIDEO / DATA /POWER
FIGURE 7-2
Digital video network using Power over Ethernet (PoE)
Digital Transmission—Video, Communications, Control
205
POWERED DEVICE (PD)
TX (TRANSMIT) PAIR
POWER SOURCING EQUIPMENT (PSE)
TO ETHERNET DEVICE
RX (RECEIVE) PAIR
TO ETHERNET DEVICE
SENSING CIRCUIT CLOSES
WHEN PD IS DETECTED
DETECTION
RESISTOR
+ 48 VDC
TX
TX
1
2
1
2
RX
3
3
4
4
5
5
6
6
7
7
8
8
48 VDC
RX
FIGURE 7-3
POWERED
ELECTRONICS
• CAMERA
• ROUTER
• SERVER
• OTHER
48 VDC GND.
SWITCH OPEN UNTIL
SUCCESSFUL DETECTION
Power over Ethernet (PoE) connections
an additional safety factor. The PoE has the capability of
powering up to a 13 watt load. Table 7-1 summarizes the
characteristics of UTP CAT Cables.
The PoE avoids the need for separate power and data
cable infrastructure and costly AC outlets near cameras.
CAT CABLE
TYPE
It reduces installation time, a significant saving in cost.
It allows networks cameras to be installed where they are
most effective, and not where the AC power outlets reduce
the number of cameras and further reduce the surveillance implementation costs. Power delivered over the LAN
CROSS TALK
* NEXT (dB)
APPLICATION
BANDWIDTH
(MHz)
IMPEDANCE
(ohms)
CAT-3
16
100
29
10BaseT LAN
STANDARD ETHERNET
CAT-4
20
100
30
10/100BaseT LAN
FAST ETHERNET
CAT-5
100
100
32.3
10/100BaseT
FAST ETHERNET
CAT-5e
100
100
35.3
1 Mb/s
GIGABIT ETHERNET
CAT-6
250
100
44.3
1Mb/s
GIGABIT ETHERNET
CAT-7
600
100
62.1
1 Mb/s
GIGABIT ETHERNET
* NEAR END CROSS TALK
NOTE: CABLE SPECIFICATIONS TYPICAL FOR UNSHIELDED TWISTED PAIR (UTP)
AWG (AMERICAN WIRE GUAGE) 22 AND 24
CAT-3, CAT-5e MOST COMMON FOR VIDEO
Table 7-1
Category (CAT) Cable Specifications
206
CCTV Surveillance
infrastructure is automatically activated when a compatible terminal is identified, and blocked to legacy analog
devices that are not compatible. This allows the mixture
of analog and power over LAN compatible devices on the
same network. Two system types are available: (1) power is
supplied directly from the data ports; (2) power is supplied
by a device between an ordinary Ethernet switch and the
terminals, often referred to as the “power hub.” By backing
up the power over LAN in the communication room with
an uninterrupted power supply (UPS), the entire camera
network can continue operation during a power outage.
This is a real must for high-end surveillance systems.
The inclusion of line detection technology that enables
safe equipment installation without concerns of highvoltage damage to laptops, desktops, and other equipment
due to a misplaced connection is one of the reasons the
power over LAN is much more than an intelligent power
source. To take advantage of PoE the power source equipment must be able to detect the presence of a PD at
the end of any Ethernet cable connected to it. The PD
appliances must assert their PoE compatibility and their
maximum power requirements. When the system is powered up the PoE enabled LAN appliances identify themselves by means of a nominal 25 K resistance across their
power input.
7.2.1.3 Wide Area Network (WAN)
The WANs, in the past, suffered from limited bandwidth.
The most common WAN link was a T1 telephone land line
supplied by AT&T with a maximum data rate of 1.5 Mbps.
Advanced technology WAN systems now incorporate optical OC3 (155 Mbps) and OC12 (622 Mbps) communication links. Figure 7-4 shows a diagram of the WAN as
applied to digital video surveillance.
7.2.1.4 Internet, World Wide Web (WWW)
During the 1990s an open systems revolution swept through
the IT industry, converting thousands of computers connected via proprietary networks to the Internet, a network
of networks based on common standards. These standards
were called transmission control protocol/Internet protocol (TCP/IP) for communications, simple mail transfer
protocol (SMTP) for email, hypertext transfer protocol
(HTTP, http://) for displaying web pages, and file transfer
protocol (FTP) for exchanging files between computers
on the Internet. The Internet has made long-range video
security monitoring a reality for many security applications
(Figure 7-5).
The availability of high-speed computers, large solid
state memory, the Internet, and the WWW has brought
CCTV surveillance from a legacy analog technology to
an OCTV digital technology. The WWW, also known
as The Web, is a salient contributor to the success of
OCTV and AVS. The WWW was developed at the CERN,
the European laboratory for Particle Physics in Geneva,
Switzerland, by Tim Berners-Lee. The web is a multiplatform operating system that supports multimedia communications on the basis of a Graphical User Interface
(GUI). The GUI provides hypertext that enables the user
to click a highlighted text word in search related files,
across web servers, and through hot links: in other words
the Web is hyperlinked. In addition to the video, the Web
supports graphics and audio with levels of quality and
speed depending on the bandwidth available in the network. Since the initial conception of the Web at CERN, its
home has moved to the W3 Consortium (W3C), a cooperative venture of CERN, the Massachusetts Institute of
Technology (MIT), and INRIA a European organization.
Since its organization in 1994 W3C has published numerous technical specifications to improve and expand the
use of the WWW.
Security monitoring is no longer limited to local security rooms and security officers, but rather extends out to
remote sites and personnel located anywhere around the
world. Monitoring equipment includes LCD and plasma
display monitors, PCs and laptops, PDAs, and cell phones.
The requirement for individual personnel to monitor multiple display monitors has changed to a technology of
incorporating smart cameras and VMDs to establish an
AVS system.
The Internet is comprised of LANs using a large array
of interconnected computers through which video and
other communication information is sent over wired and
wireless transmission channels. The location of the sender
and receiver can be anywhere on the network, viewing
scenes from anywhere in the world (Figure 7-6).
The IP is the method by which the digital data can be
sent from one computer to another over the Internet in
the form of packets. Any message on the Internet is divided
into these sub-messages called packets containing both the
senders and receivers address. Because the video message
is divided into many packets, each packet may take a different route through a different gateway computer across
the Internet. These packets can arrive in a different order
than the order in which they were sent. The IP just has
the function to deliver them to the receiver’s address. It
is up to another protocol, the TCP to put them back together
in the right order.
Each computer on the network is known as a host on
the Internet and has at least one address that uniquely
identifies it from all the other computers on the network.
The digital message can consist of an email, a web page,
video, or other digital data. When the video or other data
stream is sent out over the Internet a router (in the form
of software or hardware) determines the next network to
which a packet in the message should be forwarded toward
its final destination. The packet does not go directly from
the sender (transmitted location, i.e. camera, etc.) to the
receiver but generally goes through a gateway computer
SITE 2A—MOBILE
SITE 1 CORPORATE
ROUTER /
BRIDGE
LAN
ACCESS
POINT/BRIDGE
SITE 2—CLIENT
ROUTER /
BRIDGE
WLAN
T1 CABLE—1.5 Mbps MAXIMUM
FIBER OPTIC OC3–A55 Mbps
OC12– 622 Mbps
SITE 4 – CLIENT
WLAN
LAN
ROUTER /
BRIDGE
INTERNET
LAN
SITE 3A—MOBILE
ROUTER—DEVICE THAT MOVES DATA BETWEEN DIFFERENT NETWORK SEGMENTS.
LOOKS AT PACKET HEADER TO DETERMINE THE BEST PATH FOR THE
PACKET TO TRAVEL. CAN CONNECT NETWORK SEGMENTS THAT USE
DIFFERENT PROTOCOLS.
BRIDGE—DEVICE THAT PASSES DATA PACKETS BETWEEN MULTIPLE NETWORK
SEGMENTS USING THE SAME COMMUNICATIONS PROTOCOL. IF THE
PACKET IS BOUND FOR ANOTHER SEGMENT USING A DIFFERENT PROTOCOL
THE BRIDGE PASSES IT ONTO THE NETWORK BACKBONE.
WLAN
ROUTER/
BRIDGE
LAN
CLIENT—NETWORKED PC OR TERMINAL THAT SHARES SERVICES WITH OTHER PCs.
ACCESS POINT—WIRELESS BASED DEVICE FOR CONNECTING ROAMING WIRELESS PC
CARDS DIRECTLY TO THE INTERNET. THE ACCESS POINT PROVIDES
ROAMING AND MOBILITY FROM A STATIONARY INTERNET CONNECTION.
FIGURE 7-4 Wide area network (WAN) diagram
ACCESS
POINT/BRIDGE
Digital Transmission—Video, Communications, Control
SITE 3—CLIENT
207
208
CCTV Surveillance
PDA
CELLPHONE
IP
CAMERA
ANALOG
CAMERA
LAPTOP
SERVER
TCP/IP
(TRANSMISSION CONTROL PROTOCOL /
INTERNET PROTOCOL)
SMTP
(SIMPLE MAIL TRANSFER PROTOCOL)
WIFI
CELLULAR
TOWER
LAN
IT NETWORK
HTTP
(http://)
HYPERTEXT DISPLAYING WEB PAGES
WORLD WIDE WEB (WWW)
INTERNET
FTP
(FILE TRANSFER
PROTOCOL)
INTRANET
TOWER
FIGURE 7-5
Block diagram for remote video surveillance via the Internet
SITE 2
TOWER
SITE 1
ANALOG
CAMERA
SERVER
IP
DOME
IP
CAMERA
C
TOWER
LAN
C
C
SITE 3
INTERNET
C
PDA
LAPTOP
C
WIFI
NETWORK
C
C
LAN
TOWER
INTRANET
C
IP
DOME
SERVER
PTZ
C
• VIDEO DATA PACKETS TAKE DIFFERENT
ROUTES FROM SENDER TO RECEIVER
C
C = HOST COMPUTERS WORLDWIDE
• EACH VIDEO COMPONENT AND COMPUTER (C)
HAS A UNIQUE ADDRESS (MAC)
• TCP PUTS PACKETS OF DATA (VIDEO, CONTROLS, ETC.)
BACK TOGETHER IN THE CORRECT ORDER
• PACKETS TAKE DIFFERENT ROUTES FROM SENDER TO RECEIVER
FIGURE 7-6
Worldwide video monitoring using Internet system
Digital Transmission—Video, Communications, Control
that forwards the packet onto a next computer toward its
final destination.
The Internet allows for complete remote video surveillance, audio communication, and remote control from
any one location to any other location on the network.
As soon as a network is connected to the Internet, any
authorized computer with a browser can receive security
services. For that matter, any security system, even a system
that is not networked, can be potentially made Internet
based, fully or partially, the moment Internet access is
provided.
Traditional central stations are connected to the security systems being monitored by means of a network connection (Ethernet), a telephone dial-up, direct hard wire
connection, satellite uplink, or by radio signal. Product
literature that sites either “IP addressable” or “TCP/IP”
reveals that the product (IP camera, etc.) has some potential for network or Internet-based applications.
An important movement in the Internet industry is
the development of application service providers (ASPs).
A commercial central station could operate as a security ASP, just as an ASP could monitor security alarms
that are reported across the Internet. This could be
carried a step further in an example such as connecting police departments, enabling the police not only to
view and hear what is happening at a crime scene, but
to follow events as they occur before a police response
arrives.
TRANSMISSION
TYPE
TYPICAL
DOWNLOAD
SPEED
7.2.1.5 Leased Land Lines, DSL, Cable
There are several wired transmission means for transmitting the digitally encoded video and other data signals.
The most common options for gaining connection to the
Internet are: (1) leased land lines using PSTN modem,
(2) ISDN telephone, (3) asymmetrical digital subscriber
line (ADSL), and (4) cable. The PSTN and ISDN do not
offer the capacity (bandwidth) to provide multiple channels of high-quality live video, but are a perfectly usable
channel for non-real-time video alarm verification or event
query searching from DVRs. The ISDN is a logical choice
for many video alarm verification applications as it has
an excellent reliability specification, is almost universally
available, and is competitively priced for the data carrying
capacity it provides. Table 7-2 summarizes the bandwidth
carrying capacity of these transmission channels.
7.2.1.5.1 PSTN-ISDN Link
The dial-up PSTN is the most common of the available
transmitting methods for digital video transmission over
long distance wired networks. The service was designed
for human voice, not high-speed video transmission. The
data carrying capacity accessed is at best that of the PSTN
modem or ISDN link, and often much less depending on
network availability and traffic. On paper, ADSL offers a
much faster connection to the Internet. This is based on
the assumption that not all users will require all of the
bandwidth they have paid for, all of the time. Typically, up
TRANSMISSION
TIME FOR 25 kb
IMAGE (SEC.)
MAX. FRAME
RATE FOR
25 kb IMAGE
CONNECTION
MODE
PSTN
45 Kbps
6
10 Frames/min
DIAL-UP
ISDN
120 Kbps
2
0.5 Frames/sec
DIAL-UP
IDSL
150 Kbps
2
0.06
ADSL—LOW END
640 Kbps
0.3
3
0.05
20
ADSL—HIGH END
5 Mbps
HDSL
1.5 Mbps
0.2
6
VDSL
20 Mbps
0.01
80
CABLE MODEM
750 Kbps
0.3
3
T1
1.5 Mbps
0.2
6
10BaseT
5 Mbps
0.05
20
100BaseT
50 Mbps
0.005
200
1000BaseT
500 Mbps
0.0005
2000 Frames/sec
IDSL: ISDN DSL
ADSL: ASYNCHRONOUS DSL
Table 7-2
209
HDSL: HIGH BIT-RATE DSL
VDSL: VERY HIGH DATA RATE DSL
PSTN, ISDN, ASDL, Ethernet, and other Cable Speeds
DIRECT CONNECTION
DIRECT CONNECTION
210
CCTV Surveillance
to 20–50 users share the ASDL bandwidth depending on
the service selected. For occasional access to stored video,
this may be quite acceptable but for multi-channel live
surveillance it is unlikely to be satisfactory. If the Internet
is used for security applications, it is wise to have a backup
communications by a more reliable network and to select
equipment that can automatically revert to this backup
network.
7.2.1.5.2 DSL Link
The DSL technology supplies the necessary bandwidth
for numerous applications including high-speed Internet access, dedicated Internet connectivity, and live video
monitoring. This digital broadband data line directly connects the client computer to the Internet via existing
cables. The speed of DSL varies depending on the connection speed and in some cases the number of people
on the network.
7.2.1.5.3 T1 and T3 Links
The T1 and T3 networks have much higher speeds than
those previously described. “T1” is a term coined by American Telephone and Telegraph (AT&T) for a system that
transfers digital signals at 1.544 Mbps. T3 is the premium
transmission method and has almost 30 times the capacity
of T1. T3 lines can handle 44.736 Mbps. Fiber optics with
its much higher bandwidth and many superior characteristics is replacing T1 and T3 transmission cables.
7.2.1.5.4 Cable
Community Antenna Television (CATV) networks have
developed in parallel with DSL, and now compete for
Internet access and even voice communication, in addition
to the entertainment TV for which they were developed.
Cable provides yet another means for transmitting the
analog and digital video signal. Access to the Internet is
offered by a number of CATV providers. Since the mid1990s a number of these CATV providers have upgraded
much of their traditional coax-based networks with optical fiber, thereby increasing overall network performance
considerably. Both the coax and fiber-optic networks can
support video and two-way Internet access. With the appropriate electronic upgrades, high-speed Internet access
can be provided at end-user costs comparable with DSL
networks.
7.2.1.6 Fiber Optic
Fiber optics is used as the transmission media of choice
for digital signals transmitted over long distances or where
severe electrical disturbances (lightning storms, electrical
equipment) are present. The attributes of fiber optics are:
(1) long-distance transmission—over many miles without
degradation of the signal, (2) ultra-wide bandwidth resulting from the use of optical frequencies, and (3) secure
transmission because of the difficulty to tap the optical
signal.
In analog systems the output signals whether video or
audio are analogs of the input signals. Analog signals are
susceptible to rapid degradation, electrical noise interference, and distortion along the transmission channel.
Analog signals are also degraded when multiple generations or reproductions of signals are required. Digital
signals, on the other hand, are immune to such problems.
Theoretically any number of signal re-generations is possible with zero loss of quality. However, once the digital
signal becomes too small or the interference too large, the
signal “breaks up” or totally drops out.
Amplitude modulation (AM), frequency modulation
(FM), and pulsed-frequency modulation (PFM) are used
in analog video fiber-optic transmission systems. In digitally encoded fiber-optic video transmission the video
signals are sampled at very high rates and converted into
digital signal formats. In both cases these signals are
applied to light emitting diodes (LEDs) or injection laser
diodes (ILDs) inside the optical transmitter units. The
digital optical signals are transmitted through the fibers
and then converted back to analog, base-band electrical
video signals inside the optical receiver units. Figure 7-7
compares the AM, FM, and PFM transmission.
The AM video transmission is limited to short distances
using multi-mode optical fiber and only available at the
850 nm operating wavelength. The FM transmission, on
the other hand, provides very high video transmission performance over long distances and is available for use at 850
and 1300 nm. The 1300 nm wavelength has higher transmission through the atmosphere and is more eye-safe.
The latest generation of fiber-optic video transmission
equipment digitizes the analog base-band video signals to
provide a digital signal. This is accomplished via analog to
digital (A/D) converters or coder-decoders inside the optical transmitters. The digitized signals modulate the LEDs
or ILDs and then inject them optically into and through
the fibers to the optical receivers where they are converted
back into analog base-band signals by internal digital to
analog (D/A) converters. Factors affecting the image quality in digitally encoded video transmission and its effect
on the electrical dynamic range and signal-to-noise ratio
(S/N) of the output video signal is the number of bits
employed in the D/A and the compression employed. No
video compression is needed in fiber-optic transmission
because of the very wide bandwidth capabilities of the fiber
optic. This means that the video is transmitted in real-time
with zero latency (no delay) and standard 30 fps. A summary of the channels available and speeds of transmission
and other parameters are compared in Table 7-3.
7.2.2 Wireless Channels
The WiFi network can be connected to the Internet
through the use of a variety of high-speed connections
including cable modems, DSL, ISDN, satellite, broadband,
Digital Transmission—Video, Communications, Control
TYPICAL FIBER OPTIC FM VIDEO LINE S/N RATIO
TYPICAL FIBER OPTIC AM VIDEO LINE S/N RATIO
S/N RATIO (dB)
S/N RATIO (dB)
65
65
60
60
55
55
50
50
45
45
40
40
35
5
1
211
10
35
20
15
5
1
20
15
10
OPTICAL PATH LOSS (dB)
OPTICAL PATH LOSS (dB)
TYPICAL FIBER OPTIC DIGITALLY EMBEDDED VIDEO LINE S/N RATIO
S/N RATIO (dB)
65
60
55
50
45
40
35
1
5
10
15
20
OPTICAL PATH LOSS (dB)
FIGURE 7-7
Comparison of AM, FM and pulse frequency modulation
TRANSMISSION
TYPE
THEORETICAL *
DOWNLOAD
SPEED
TRANSMISSION
MEDIA
PSTN
45 Kbps
ISDN
120 Kbps
CAT-3
HDSL
1.5 Mbps
CAT-3, 5, 5e
CABLE MODEM
750 Kbps
CAT-3, 5, 5e
5 Mbps
CAT-3, 5, 5e
10BASE T
UTP
CAT-3
50 Mbps
CAT-5e
1000BASE T
500 Mbps
CAT-6
T1
1.5 Mbps
CAT-3, 5e
T3
45 Mbps
100BASE T
UTP
CAT-5, 5e
OC3
155 Mbps
FIBER OPTIC
OC12
622 Mbps
FIBER OPTIC
* REALISTIC SPEED APPROXIMATELY 1/2 OF THEORETICAL
Table 7-3
Comparison of Wired UTP and Optical Transmission Channels
etc. The broadband Internet connection connects to a
video gateway or access point, and its Internet connection
is distributed to all the computers on the network. The
access points or gateways function as the “base stations”
for the network. They send and receive signals from the
WiFi radios to connect the various components of the
security system to each other as well as to the Internet. All
computers in the WiFi network can then share resources,
exchange files, and use a single Internet connection. This
is the central connection among all wireless client devices
(PC, laptop, printers, etc.) and enables the sharing of
the Internet connection with other users on the network.
212
CCTV Surveillance
Access points and gateways have a wide range of features
and performance capabilities and provide this basic network connection service.
7.2.2.1 Wireless LAN (WLAN, WiFi)
The WiFi (Wireless Fidelity) devices “connect” to each
other by transmitting and receiving signals on a specific
frequency of the radio frequency (RF) and microwave
bands. The components can connect to each other
directly, called peer to peer or through a gateway or access
point. The WiFi networks consist of two basic components:
(1) WiFi radios and (2) access points or gateways. The
WiFi radios are attached to the desktop computer, laptop,
or other mobile devices on the network. The access points
or gateways act as “base stations,” i.e. they send and receive
signals from the WiFi radios to connect the various components to each other as well as to the Internet. All the
computers in the WiFi network then share resources and
exchange files over a single Internet connection.
The IEEE developed a series of 802.11 protocols to meet
the requirements of disparate applications, and continues to formulate new ones. The 802.11a, b, g, i, and n
standards are most useful for the wireless digital video
transmission applications. Table 7-4 summarizes some of
the parameters of the standards.
A peer-to-peer network is composed of several WiFi
equipped computers talking to each other without using a
base station (access point or gateway). All WiFi Certified™
equipment supports this type of wireless setup, which is a
good solution for transferring data between computers or
when sharing an Internet connection among a few computers.
Many laptop computers and mobile computing devices
come with a WiFi radio built into them and are ready to
operate wirelessly. For other laptops without such a device,
a WiFi radio embedded in a simple Personal Computer
Memory Card International Association (PCMCIA) card
can be inserted into expansion slot of a laptop computer.
There are other ways to include the desktop PC into the
network. Since many PCs do not have card slots for PC
cards, the simplest method is to use a universal serial bus
(USB) WiFi radio that plugs into an available USB port
on the computer.
7.2.2.2 Mesh Network
The mesh network is a topology that provides multiple
paths between network nodes. Wired networks have used
OPERATING
FREQUENCY
BAND (GHz)
DATA†
RATES
(Mbps)
OPERATING FREQUENCY BANDS
(GHz)
802.11 *
2.4
1, 2
2.4–2.8
(LEGACY)
IR
IEEE
STANDARD
MODULATION
METHOD
MAX POWER
OUTPUT
(EIRP)
DSSS
FHSS
IR
APPLICATIONS/
COMMENTS
ORIGINAL 802.11
STANDARD FOR
WIRELESS LAN
300 MHz IN 3 BANDS of 100 MHz each:
5.2, 5.8
6, 12, 24,
9, 18, 36, 48
54 MAXIMUM
5.150 to 5.250 (UNII LOWER BAND)
5.250 to 5.350 (UNII MIDDLE BAND)
5.725 to 5.825 (UNII UPPER BAND)
802.11b
2.4
1, 2, 5.5, 11
11 MAXIMUM
83.5 MHz FROM 2.40 GHz to 2.4835 GHz
(ISM BAND)
802.11g
2.4
802.11a **
1, 2, 5.5, 11
6, 9, 12, 18,
24, 36, 48, 54
2.4–2.4835
108
20–40 MHz
* IEEE ESTABLISHED STANDARD IN 1997 TO DEFINE MAC (MEDIA ACCESS CONTROL)
AND PHY (PHYSICAL) LAYER REQUIREMENTS FOR WIRELESS LAN.
** IEEE ESTABLISHED 802.11a IN 1999
†
THEORETICAL MAXIMUM RATES. REALISTIC MAXIMUM APPROXIMATELY ONE–HALF
‡
ADVANCED ENCRYPTION STANDARD
ISM—INDUSTRIAL, SCIENTIFIC, MEDICAL
UNII—UNLICENSED NATIONAL INFORMATION INFRASTRUCTURE
COFDM—CODED ORTHONOGONAL FREQUENCY DIVISION MULTIPLEXING
FDMA—FREQUENCY DIVISION MULTIPLE ACCESS
DSSS—DIRECT SEQUENCE SPREAD SPECTRUM
FHSS—FREQUENCY HOPPING SPREAD SPECTRUM
EIRP—EQUIVALENT ISOTROPICALLY RADIATED POWER
IR—INFRARED
Table 7-4
DSSS
FDMA
DSSS
COFDM
40 mW
200 mW
800 mW
1 WATT
TYPICAL: 30 mW
INDOOR
INDOOR
OUTDOOR
USES FDMA, DSSS
DUAL BAND
2.4 GHz
ADDS HIGH LEVEL
AES ENCRYPTION‡
802.11i
802.11n
COFDM
Comparison of IEEE 802.11 Standards
VERY HIGH
DATA RATE
Digital Transmission—Video, Communications, Control
the mesh topology to get redundancy and reliability. Mesh
networks make the most sense with wireless transmission
because wireless nodes can be set up to form ad hoc networks that connect many nodes. In the wireless application
if interference or excess distance between nodes causes a
dropped video link the mesh system will find an alternate
path through the mesh automatically. The nodes themselves may generate messages to be sent elsewhere or be
available to receive data or both. The nodes act as repeaters
to move the video and other data from point-to-point when
they are not transmitting or receiving their own data. What
results is a very robust network at low cost. The Mesh
network using many closely spaced repeater transceivers
(nodes) is shown in Figure 7-8.
Each node can communicate with its nearby neighbors that are within range. The nodes can exchange data
between themselves, store it, or forward data meant for a
more distant node that is out of range of a nearby node.
One of the nodes can also serve as a wired or wireless
connection to an Internet node or access point. A particular attribute of the wireless Mesh network using multiple nodes is that it allows the signal to be transmitted
over a longer range than would be possible with a normal
line-of-sight (LOS) link. In mesh networks multiple paths
IP CAMERA
213
exist through the network system, increasing the probability that the video signal from the camera will reach the
monitoring location. The Mesh configuration is also more
reliable since if one of the nodes fails due to a power
loss, jamming or other defect, communication is still maintained, i.e. the video, voice, communication, or control
signals can be routed through another path. In addition
to the reliability aspect, the Mesh configuration offers the
benefit of requiring very low transmitted power at any
given node because the distance between nodes is usually
short. Mesh networks are especially useful in monitoring a
large network of image and/or alarm sensors. In portable
and rapid deployment applications, low transmit power
means low device power consumption and longer battery
life. The military has already adopted mesh networks in
battlefield systems and many forms of video security are
ideal applications for this growing technology.
7.2.2.3 Multiple Input/Multiple Output (MIMO)
Most wideband WiFi networks operate with data rates
between 11 and 54 Mbps. There is however a need for
greater network bandwidth capacity for wireless LANs.
The wireless radio channel for moving video and other
OUTDOOR
MESH
NODE
IP CAMERA
DOME
WIRELESS
MESH
NETWORK
BNC
ANALOG
CAMERAS
IP CAMERAS
MESH
NODE
SERVER
IP CAMERAS
BNC
CENTRAL MONITORING
FIGURE 7-8
Wireless mesh transmitting network
214
CCTV Surveillance
digital information over the air waves has a highly variable
nature. Unlike the relatively stable environment that exists
on wire, cable, or fiber-optic networks, the ability of the air
to carry information can and does change over time and
often from moment to moment. With this fundamental
variability and the overhead inherent in any networking
protocol, the actual throughput available from a 54 Mbps
connection is often much less than this peak number. As a
consequence it is necessary to improve the performance
of wireless LANs at the physical layer if higher throughputs are to be achieved. One popular approach is to gang
together multiple radio channels and to use compression
and related techniques to gain some additional advantage
in information throughput. The ideal solution is to come
up with a technology that simply packs more information
per unit of bandwidth and time. This technique applied to
wireless transmission is known as modulation efficiency—the
number of bits per unit of bandwidth and time that can
be transmitted through the air at any given time.
Radio signals are subject to serious degradation as
they move through space, primarily due to the distance
between the transmitter and receiver, interaction with
objects in the environment, and interference from other
radio signals and reflections of the signal in question itself
(known as multi-path). All these artifacts result in a number of forms of fading, the loss in power of the radio signal,
as it moves from the transmitter to the receiver.
The technique available today that has been put into
practice in a wireless LAN is called multiple input, multiple output (MIMO). This technology adds an additional
dimension to the radio channel—a spatial dimension—
allowing a more complex but inherently more reliable
radio signal to be communicated (Figure 7-9).
Whereas conventional radio transmission uses a single
input, single output, a true MIMO system uses at least
two transmit antennas, working simultaneously in a single
channel, and at least two receive antennas at the other end
of the connection working in the same channel. Generally
the number of receive antennas in a MIMO system is usually greater than the number of transmit antennas and the
performance of transmission improves with the addition
of more receive antennas. Going from a single antenna
to two antennas can result in a 10 × 10 dB improvement
in the S/N, a key indicator of reliability and signal quality. Adding a third antenna adds an additional 4 × 5dB
improvement. Figure 7-10 illustrates a six-antenna MIMO
receiver.
The MIMO technology relies upon the interactions of
the signal with the environment in the form of multipath for its benefits—a counterintuitive element in the
technology. The phenomenon is attributed to reflections
and multi-path transmissions from walls, ceilings, floors,
and other objects. By improving the performance of the
antennas and the number of them used in the WLAN, the
REFLECTING
OBJECT(S)
TRANSMITTER
MULTIPATH
SIGNALS
RECEIVER
MIMO
SIGNAL
PROCESSING
(RF + DSP)
MIMO
SIGNAL
PROCESSING
(RF + DSP)
MIMO
TRANSMITTER
ELECTRONICS
MIMO
SIGNAL
PROCESSING
(RF + DSP)
DUAL
TRANSMIT
ANTENNA
MIMO
SIGNAL
PROCESSING
(RF + DSP)
QUAD
RECEIVE
ANTENNA
FIGURE 7-9
Multiple input, multiple output (MIMO) receiver
MIMO
RECEIVING
ELECTRONICS
Digital Transmission—Video, Communications, Control
FIGURE 7-10 Six antenna wireless LAN MIMO receiver
overall performance is significantly improved. The MIMO
technology introduces a third spatial dimension beyond
the frequency and time domains, which would otherwise
define the radio channel.
The major difference between the MIMO and traditional wireless systems is a utilization of the physical multipath phenomenon. Unlike traditional modems that are
typically impaired by multi-path, MIMO takes advantage
of multi-path. The typical radio signal from a point source
(single antenna) typically bounces off different objects
during transmission, particularly indoors as it interacts
with objects in the environment. The result of these interactions is multi-path fading, as the signal interferes often
destructively with itself. The MIMO takes advantage of
multiple paths, using signal processing implemented on
digital signal processor (DSP) chips, and using clever algorithms at the transmitter and receiver. Somewhat counter
intuitively, MIMO actually depends upon multi-path to
function correctly and produce improvements, making it
even better suited to in-building applications. The MIMO
can offer a dramatic improvement in signal throughput
over competing WLAN technologies. The new 802.11n
standard including MIMO processing in its specification
should produce performance of 144–200 Mbps.
7.2.2.4 Environmental Factors: Indoor–Outdoor
Indoor and outdoor environmental effects must always be
considered when implementing a wireless analog or digital
video system. Atmospheric conditions, objects in the signal’s path, incorrect antenna pointing angle can all cause
fading and dropouts in the digital video signal. All of these
factors affect the quality of service (QoS) in the resulting
video image or other communication data. Most analog
and digital video transmission takes place using the FCC
allocated 902 MHz, 2.4 GHz, and 5.8 GHz bands, each of
215
which exhibit signal degradation under different conditions. The 902 MHz and 2.4 GHz bands provide the best
transmission through most non-metal, dry solid objects,
but the 5.8 GHz band exhibits severe attenuation when
objects are placed in the path between the transmitter and
receiver. The 5.8 GHz band should only be used for short
range indoor applications and clear LOS outdoor applications or where specific metal reflectors can be placed to
re-direct the microwave beam to the receiver. The 802.11b
technology operates at 2.4 GHz and a data rate of 11 Mbps
and can handle up to three video data streams at a time.
The 802.11g technology operating at 2.4 GHz and a data
rate of 54 Mbps, and the 801.11a technology operating
at 5.8 GHz and a data rate of 54 Mbps can manage multiple standard video streams. They all require innovative
techniques to provide high QoS and quality video images.
One system using a diversity antenna array (not MIMO)
provides wireless connections at data rates of up to
54 Mbps over a time domain multi-access (TDMA) proprietary link that uses the 802.11a, 5.8 GHz frequency band.
The system permits multiple streams of DVD, cable and
satellite digital video, audio, and data to be delivered over
the wireless links without degrading quality. The key to
the improved QoS is in the front end of the receiver.
The RF transceiver employs a spatial wave-front receiver
that uses five antennas and two full receiver channels to
eliminate multi-path (ghost) signals. It does this by using
the five-antenna array to capture the RF signals and then
selects the best two of five signals. This approach takes
advantage of the multi-path signals as opposed to other
techniques that try to eliminate them. After the two signals are selected they are fed into separate independent
receive channels that amplify, filter, frequency convert,
and eventually feed them to the base-band processor. The
base-band chip converts the two analog signals into digital
streams and then, using DSP techniques, combines them
into one high-quality data stream. When a system is set
up it scans the available channels for one that is not in
use by any nearby 802.11 WiFi network. The chip then
continuously monitors all channels for possible interference and, if a potential interference is detected, the chip
looks for another unused channel. The signals that can
be processed can come from any source since the chip
can process video in any standard format from MPEG-1
to MPEG 4, H 264. It should be pointed out that most
systems in use do not use diversity antenna arrays and are
therefore limited to transmitting fewer channels of video.
7.2.2.5 Broadband Microwave
Microwave transmission uses ultra-high frequencies to
transmit video signals over long distances. There are several frequency ranges assigned to the microwave systems
all in the gigahertz ranges. Table 7-5 lists the broadband
microwave frequencies bands available for transmission.
216
CCTV Surveillance
FREQUENCY
BAND
CHANNEL
FREQUENCY
900 MHz*
1.2 GHz
2.4 GHz*
5.8 GHZ*
902–928
1.2–1.7
2.4–2.5
5.6–5.8
L
1.7–1.9
S
2.2–2.5
C1
C2
3.1–3.5
C3
X
6.2–6.4
8.2–8.6
K
21.2–23.6
4.4–5.0
NUMBER OF
CHANNELS
OUTPUT POWER/
RANGE
4, 2 SIMULTANEOUSLY
4, 2 SIMULTANEOUSLY
4 SIMULTANEOUSLY
11 SIMULTANEOUSLY
50–500 mW
SHORT/
MEDIUM
RANGE
300–2000 ft
SYSTEM
DEPENDENT
0.25–5 W
LONG
RANGE
1–20 MILES
SITE 1
SITE 2
NARROW BEAM
TRANSMITTER/
DISH/HORN
DISH/HORN/LOW
NOISE RECEIVER
LONG-RANGE TRANSMISSION
CAMERA
MONITOR
* NO FCC LICENSE REQUIRED FOR THESE LOW POWER TRANSMITTERS
FCC LICENSE REQUIRED OR FOR GOVERNMENT USE ONLY FOR ALL OTHERS
Table 7-5
Broadband Microwave Frequencies for Video Transmission
The wavelength of these frequencies is very short and
gives rise to the term “microwave.” These high-frequency
signals are especially susceptible to attenuation and must
therefore be amplified frequently if long distances (20–
50 miles) separate the transmitter and receiver. Repeaters
at intermediate locations between the transmitter and
receiver are used when distances exceed 20–30 mi. In
order to maximize the strength of the high-frequency signal, focused antennas are used at both ends. Since the
microwave frequencies have characteristics similar to light
waves, these antennas can take the form of concave metal
dishes that collect the maximum amount of incoming signal and reflect it to the receiver detector. The requirement
for these tightly focused antennas limits the microwave
application, and it is clearly a point-to-point rather than
a broadcast transmission system. These microwave signals
will not be passed through buildings, uneven terrain, or
any other solid objects. Broadband microwave technology
used as a video transmission media is used to interconnect LANs between buildings and over long distances. The
microwave dishes must be line-of-sight from transmitter
to receiver to collect the microwave signals reliably. Using
the microwave technology requires FCC licensing, however once the license is granted for any particular location,
that frequency band cannot be licensed to anyone else for
any purpose within a 17.5 mi. radius.
7.2.2.6 Infrared (IR)
Infrared (IR) links use IR signals to transmit video, data,
and control signals. These IR transmission paths must be
set up in a line of sight configuration or the IR signal can
be reflected off an infrared reflecting surface (mirror).
The major advantage of infrared transmission is its ability
to carry a high-bandwidth signal and its immunity to tapping. Its major disadvantage is that the IR beam can be
obstructed and it cannot pass through most solid objects.
The IR emitter is in the form of an LED or ILD.
7.3 VIDEO IMAGE QUALITY
In both legacy analog and digital video surveillance systems, the criteria for image quality include resolution,
frame rate, and color rendition. In digital video monitoring and surveillance applications each camera generates
a stream of sequential digital images typically at a rate
of 2–20 per second, or 30 per second for real-time. In
the video application the data network must be capable
of sustaining a throughput required to deliver the packets comprising the video streams being generated by all
the cameras. This is one measure of QoS, but QoS also
encompasses latency (the delay between transmitting and
receiving packets) and jitter (the variations in that delay
from packet to packet). The QoS criterion is generally
applied to the forward video signal direction since the vast
majority of traffic results from these video streams from
the camera to the monitor and recorder. The QoS does
apply in some cases where the cameras offer centralized
in-band control, whether to simply adjust settings from
time-to-time or to PTZ the cameras in real-time.
The Internet and other IP-based networks increasingly
are being used to support real-time video applications,
voice, and audio, all of which are extremely demanding in
terms of latency, jitter, and signal loss. The Internet and
its original underlying protocols were never intended to
support QoS, which is exactly what each of these traffic
types requires. The real-time streaming protocol (RTSP)
is an application layer (Layer 7, Section 7.8.3) protocol for
control over the delivery of data that has real-time properties including both live video data feeds, stored video
clips, and audio.
7.3.1 Quality of Service (QoS)
The QoS describes the video image quality and intelligence in the digital video image as determined by the
video frame rate and resolution (number of pixels).
Digital Transmission—Video, Communications, Control
The QoS is defined as the control of four network categories: (1) bandwidth, (2) latency, (3) jitter, and (4) traffic
loss. Bandwidth is defined as the total network capacity.
Latency is the total time it takes for a frame to travel
from a sender to a receiver. Latency can be crucial with
receivers having QoS requirements. Packets arriving too
early require buffering, or worse they may be dropped.
Packets arriving to late are not useful and must be discarded. Jitter is the variation in the latency among a group
of packets between two nodes. Jitter requires a receiver to
perform complex buffering operations so that packets are
presented to higher levels with a uniform latency. Traffic
loss refers to the packets that never arrive at the receiver.
The video signal requires compression to fit into the
bandwidth available in the communication channel, and
for the practical compression techniques used, this compression always results in signal degradation (exception
lossless transmission). Data transmission is generally considered moving in one direction in a video monitoring or
surveillance application: that is the vast majority of traffic
results from the video streams flowing from the camera
to the monitor or video recorder. There is some traffic
that flows in the other direction including controls for the
camera functions.
7.3.2 Resolution vs. Frame Rate
Resolution is a measure of how clear and crisp an image
appears on the monitor. Each of the individual video
components included within a system contributes to the
overall image quality, either recorded or displayed on the
monitor. The resultant image quality is only as good as
the equipment component having the lowest resolution.
When a high resolution monitor is combined a with a low
resolution camera, the result is a low resolution image display. This fact becomes increasingly important when using
the system for recording, as the playback image from the
recorder is generally less than that obtained when displayed directly on the monitor.
The image quality of the video signal is dependent on:
(1) the video frame rate required to reproduce motion
in the scene, (2) the resolution required to convey the
intelligence required in the scene, and (3) the bandwidth
available for the transmission. For a practical transmission
with existing communication channels the video signal
must first be digitally compressed to fit into the available
bandwidth. To achieve the necessary intelligence in the
image, the resolution required for the application must be
specified and the network must have sufficient bandwidth.
When more than one video image (or additional information) is to be displayed on a video monitor, a format
called Common Intermediate Format (CIF) is used. Most
digital video systems with standard 4 × 3 formats display
three different resolutions: (1) full screen 704 × 480 pixels 4 × 3 resulting in the highest resolution, (2) 1/4
217
screen 352 × 240 having a proportionally lower but often
adequate resolution, and (3) full screen having 704 × 240
pixels. The 320 × 240 pixels requires 1/4 the bandwidth
and has a 4× faster image transfer rate. The 704 × 240
has 1/2 the bandwidth of the 704 × 480 system. The 1/4
CIF format has a resolution of 352 × 240 pixels with the
NTSC system and 352 × 288 pixels with the PAL system.
The three formats described above referred to as CIF are
summarized in Table 7-6. Their relative sizes are shown in
the inset drawing.
It is often desirable to display the digital video image
on only part of the display screen when the screen is
being shared with other systems functions (alarms, access
control, etc.). In this case the 1/4 CIF is most appropriate.
Since the 1/4 CIF requires only 1/4 the bandwidth, it can
display the image at 4 × the CIF rate.
7.3.3 Picture Integrity, Dropout
It is very important during digital video signal transmission that the video image have integrity throughout the
transmission. The various compression and transmission
technologies used for transmitting the video signal have
different vulnerabilities to noise and external interference
and cause the video image to be degraded in different
ways. The temporary loss of the digital signal causes image
pixelation or picture breakup which results in the loss of
“blocks” of pixels causing parts of the image to be absent
and displaying an incomplete picture. In worst case, when
the video signal strength (S/N) is sufficiently low and synchronization is lost, video frame “lock-up” occurs and the
last full frame transmitted may be displayed as a full frame,
partial frame, or none at all. For general surveillance video
surveillance applications, degradation or temporary loss
of a few frames of video signal can be tolerated. However,
in most security applications and especially in strategic
surveillance applications this is unacceptable.
7.4 VIDEO SIGNAL COMPRESSION
Video signal compression is the process of converting
analog video images into smaller digital files for efficient transfer across a network. Compression provides
reduced bandwidth, quicker file transfers, and reduced
storage requirements. Compression and decompression
are accomplished through the use of special software or
hardware or in some cases both.
From the earliest days, video (consumer television)
has been a bandwidth hog. Standard broadcast channels
require from 4 to 6 MHz of bandwidth to produce a complete picture and sound at full frame rates of 30 fps. In
digitized form the signal requires data rates on the order
of 2 Mbps.
218
CCTV Surveillance
QCIF
CIF
2 CIF
4 CIF
PIXEL FORMAT
ASPECT RATIO:
1.222
PIXEL COUNT
1/4
352 × 240 (NTSC)
352 × 288 (PAL)
88,480
101,376
1/4 SCREEN
2 CIF
FULL
704 × 240 (NTSC)
704 × 288 (PAL)
168,960
202,752
FULL SCREEN-1/2 VERTICAL RESOLUTION
QCIF
1/16
176 × 120 (NTSC)
176 × 144 (PAL)
21,120
25,344
4 CIF *
FULL
704 × 480 (NTSC)
704 × 576 (PAL)
337,920
405,504
CIF FORMAT
CIF
SCREEN AREA
DISPLAY
1/16 SCREEN
FULL SCREEN-TWICE THE VERTICAL AND
HORIZONTAL RESOLUTION OF CIF
* 4 CIF RESOLUTION IS SLIGHTLY HIGHER THAN THAT OF VGA (640 × 480)
Table 7-6
Common Intermediate Format (CIF) Parameters
In the 1980s this bandwidth limitation for transmitting video signals was addressed by the US government
Defense Advanced Research Project Agency (DARPA) to
compress NTSC and HDTV type video streams to fit within
available bands of the radio frequency spectrum. One
result of the initial work done by DARPA and MPEG
was the evolution of a family of video compression standards that apply directly to real-time video applications.
The MPEG group was founded under the International
Organization for Standardization (ISO) and created the
first compression standard MPEG-1, in 1992. This standard was directed toward single speed applications like
CD-ROM and is still used in today’s camcorders and video
CD movie rentals. Two years later, MPEG-2 followed, which
added frame interlace support and was directed toward
applications such as digital TV (DTV) and digital video
disk (DVD).
A video stream consists of a series of still images or
frames displayed in rapid succession. Each digital image
is in the form of a rectangle consisting of an array of picture elements known as pixels. Each pixel represents the
light intensity that the camera sees in either black and
white (monochrome) or color, at that pixel location. The
NTSC display contains 720 × 480 pixels which is known
as a 4 × 3 aspect ratio. High definition television (HDTV)
has a higher pixel count of 1920 × 1024. Table 7-7 summarizes the Advanced Television Systems Committee (ATSC)
digital television standards.
In monochrome cameras the intensity is represented by
a single pixel. In color cameras the sensors are grouped
together in three pixels: one for red, one for green, and
one for blue (RGB). The combination of these three colors
in different proportions produces every other color. To
convert to digital form the output from each pixel in the
camera sensor is converted to digital values by use of an
A/D converter. For the monochrome camera each pixel
is converted into an 8-bit value representing the intensity
of the image on the pixel. For the color camera the 8-bit
value and an additional 16 bits are used to digitize all three
colors (red, green, and blue), resulting in 24 bits (eight
bits for each color).
Why is digital video signal compression required? Without video compression an enormous amount of bandwidth
is required to efficiently transfer video across a network.
A 24-bit color video stream at 640 × 480 resolution transferring 30 frames in one second creates almost 30 MB
(megabyte) of data.
Compression schemes for sending data over a restricted
bandwidth have existed for years with the “zip” file of
lossless compressed data being a popular program. This
lossless compression, however, is not sufficient or suitable
for video transmission and does not take into account
an advantage of unique features of video transmission.
In particular, individual frames of video often contain
repetitious material and often have only small portions of
the image or frame that change from frame to frame. The
zip compression program does not take advantage of this
feature.
There are two generic types of digital video compression: lossless and lossy. Lossless as the name implies means
that all the information to reproduce every pixel present
in the camera output is transmitted to the monitoring
Digital Transmission—Video, Communications, Control
DTV
FORMAT
INDEX
VERTICAL
RESOLUTION
(PIXELS)
HORIZONTAL
RESOLUTION
(PIXELS)
SCREEN FORMAT
ASPECT RATIO
SCAN
TYPE
REFRESH
RATE (Hz)
INTERLACED
1
2
640
3
4×3
PROGRESSIVE
4
5
6
7
INTERLACED
480
704
4×3
PROGRESSIVE
8
INTERLACED
9
10
704
11
16 × 9
PROGRESSIVE
FORMAT
DESCRIPTION
H × V, fps i or p *
30
640 × 480, 30i
24
640 × 480, 24p
30
640 × 480, 30p
60
640 × 480, 60p
30
704 × 480, 30p
24
704 × 480, 24p
30
704 × 480, 30p
60
704 × 480, 60p
30
704 × 480, 30i
24
704 × 480, 34p
30
704 × 480, 30p
12
60
704 × 480, 60p
13
24
1280 × 720, 24p
30
1280 × 720, 30p
14
720
1280
16 × 9
PROGRESSIVE
15
INTERLACED
16
17
1080
1920
16 × 9
PROGRESSIVE
18
DTV—DIGITAL TELEVISION
ATSD—ADVANCED TELEVISION SYSTEMS COMMITTEE
SDTV—STANDARD DEFINITION TELEVISION
EDTV—ENHANCED DIGITAL TELEVISION
HDTV—HIGH DEFINITION TELEVISION
Table 7-7
60
1280 × 720, 60p
30
1920 × 1080, 30i
24
1920 × 1080, 24p
30
1920 × 1080, 30p
219
FORMAT
TYPE
SDTV
EDTV
HDTV
* i—INTERLACED SCAN
p—PROGRESSIVE SCAN
fps—FRAMES PER SECOND
ATSC Digital Television Standard Scanning Formats
site and reconstructed without any loss in picture quality.
This means that the compression algorithms must be able
to accurately reconstruct the uncompressed video signal.
Lossy compression means that the reconstructed (decompressed) signal can not exactly re-create the original video
signal.
The following is a calculation of the number of uncompressed RGB signal bits that must be transmitted for a
single frame of NTSC video if no compression were to
take place:
To transmit 1 frame = 720 pixels × 480 pixels × 24 pixel
Spatial redundancy means that neighboring pixels within
a video frame are more likely to be close to the same value
(in both brightness and color) than far apart. Temporal
redundancy means that neighboring frames in time tend
to have a great deal of similar content, such as background
information, that is either stationary or moving in predictable ways. Any compression system will perform better
if the video signal is preconditioned properly. In practice this means removal of the noise that would otherwise
consume precious bits. Figure 7-11 illustrates some examples of spatial and temporal redundancies in a typical
video image.
= 8294400 bits
To transmit 1 second of video = 8294400 × 30 fps
= 248832000 bits
From the above it can be seen that it takes over 248 Mb
to transmit 1 second of uncompressed full-color video.
Clearly few transmission channels can afford to provide
this much bandwidth for transmitting any video signals.
For this reason some scheme of compression of video signals is required to make a practical remote video security
system.
Video compression takes advantage of enormous spatial and temporal redundancies in natural moving imagery.
7.4.1 Lossless Compression
Lossless compression is the process of compressing 100%
of video data with zero loss. This type of compression does
not compress as much as lossy compression since every
piece of data is retained. The benefit of this compression
is that video data can be compressed and decompressed
over and over without any video data degradation. Lossless
compression algorithms compress the video data into the
smallest package possible without losing any information
in the scene. The zip file for standard data (not video) is
an example of a lossless compression algorithm since the
220
CCTV Surveillance
(A) SPATIAL
ONLY MOVEMENT:
WATER IN POOL
AND SWIMMERS
SWIMMERS ENTER POOL—ONLY AREA OF INTEREST
(B) TEMPORAL
ONLY MOVEMENT:
PERSON
TRAVERSING
FENCE LINE
(C) SPECTRAL
PERSON DRESSED
IN RED DETECTED
• MJPEG OPERATES ON SPATIAL REDUNDANCY, NOT TEMPORAL
• MPEG OPERATES ON SPATIAL, SPECTRAL (COLOR) AND TEMPORAL REDUNDANCY
FIGURE 7-11
Spatial and temporal redundancies in video images
data that is compressed can be decompressed and an exact
duplicate of the original re-created at the receiver end.
Lossless compression generates an exact duplicate of the
input data scene after many compression/decompression
cycles: no information is lost. This method, however, can
only achieve a modest amount of compression. Typical
compression ratios for lossless transmission are from 2:1
to 5:1.
7.4.2 Lossy Compression
In the case of the video signal it is often not necessary that
each bit of data be re-created exactly as in the original
camera image. Depending on the video quality required
at the monitoring location, often much of the video information can be tossed away without noticeably changing
the video image that the user sees. The exclusion of this
extraneous video results in the ability to achieve high compression rates.
Lossy compression achieves lower bit counts than lossless compression by discarding some of the original video
data before compression. Video data degradation does
occur with lossy compression when it is compressed and
decompressed over and over. In other words, every time
video data is compressed and decompressed, less of the
original video image is retained.
Two common methods for compression are discrete
cosine transform (DCT) and discrete wavelet transform (DWT).
7.4.2.1 Direct Cosine Transform (DCT)
The DCT is a lossy compression algorithm that samples
the image at regular intervals. This transform divides the
video image into 8 × 8 blocks and analyzes each block
individually. It analyzes the components of the image and
discards those that do not affect the image as perceived by
the human eye. JPEG, MPEG, M-JPEG, H.261, H.263, and
H.264 incorporate DCT compression. Lossy compression
can eliminate some of the data in the image at a sacrifice
to the quality of the image produced. This reduction in
bits transmitted, however, provides greater compression
ratios than lossless compression and therefore requires
less bandwidth. The choice of lossless or lossy compression
results in a trade-off of file size vs. image quality. Lossy
compression discards redundant information and achieves
much higher compression at the sacrifice of not being
able to exactly reproduce the original video scene. Typical
compression ratios for lossy transmission are from 20:1
to 200:1.
Digital Transmission—Video, Communications, Control
7.4.2.2 Discrete Wavelet Transform (DWT)
Wavelet video compression, rather than operating on
pieces of the image, operates on the entire image. The
transformation uses a series of filters that determines the
content of every pixel in the image. Because the technology works on the entire image there is no mosaic effect
when the image is viewed as is sometimes experienced with
DCT. While wavelet technology is a lossy compression technique, the lossy effects are not apparent until very high
compression ratios of 350:1 are reached. Wavelet compression uses multiple single recorded frames to create a video
sequence. It differs from others in that it compresses files
more tightly with average file sizes for a wavelet image of
about 12 Kb or 360 Kbps at 30 fps. Wavelet compression is
based on full frame information and on frequency, not on
8 × 8 pixel blocks as in DCT. Wavelet compression compresses the entire image with multiple filtering at both the
high and low frequencies and repeats the procedure several times. This compression method offers compression
ratios up to 350:1.
7.4.3 Video Compression Algorithms
Many compression algorithms have evolved over the years
to address specific digital data transmission requirements.
The International Telecommunications Union (ITU) and
the International Organization for Standards (ISO) have
developed video compression technology and standards
that meet and exceed the requirements for most of
today’s video security applications as well as anticipated
future requirements. The compression standards that are
specifically directed toward transmitting single frame and
streaming video signals include: (1) MPEG-2, (2) MPEG4, (3) JPEG, (4) M-JPEG, (5) JPEG-2000, (6) wavelet, (7)
H.263, (8) H.264, and (9) super motion image compression technology (SMICT).
The required video frame rates for a security application are primarily determined by the motion in the scene
(activity) and the number of pixels required for the specified resolution. When there is little motion in the scene
or if the motion is slow, very often less than 30 fps are
sufficient to obtain the necessary intelligence in the scene.
This reduces the required bandwidth for the transmission
of the digital video signal. Frame rates as low as 5 fps can
be useful.
7.4.3.1 Joint Picture Experts Group: JPEG
The JPEG is the oldest and most established compression
technique and is generally applicable to still images or
single frames of video. This compression technique divides
the image into 8 × 8 blocks of pixels with each block a
signed number (plus or minus) and code (Figure 7-12).
The DCT compression software examines the blocks and
their size and determines which blocks are redundant and
221
not essential in creating the image. The program transmits
the blocks that are essential, which is a reduced number
based on the level of compression determined by the system settings. The compression ratio is limited to approximately 10:1. New compression algorithms are evolving that
have built upon JPEG and provide higher compression
ratios and have higher signal quality with smaller bandwidth requirements. The JPEG uses still images to create
a video stream and has an average image file size of about
25 Kb per frame or 750 Kbps at 30 fps.
7.4.3.2 Moving—Joint Picture Experts Group:
M-JPEG
The M-JPEG compression technology creates a video
sequence (stream) from a series of still frame JPEG
images. The average file size of an M-JPEG image is about
16 Kb per frame or 480 Kbps at 30 fps. The M-JPEG is a
lossy compression method designed to exploit some limitations of the human eye, notably the fact that small
color changes are perceived less than small changes in
brightness. With a compression ratio of 20:1, compression can be achieved with only a small fraction of image
degradation.
7.4.3.3 Moving Picture Experts Group: MPEG-2,
MPEG-4, MPEG-4 Visual
7.4.3.3.1 MPEG-2 Standard
The MPEG-2 is the successor to MPEG-1 and has the
primary goal of transmitting broadcast video at bit rates
between 4 and 9 Kbps. It produces high-quality live camera
images using a relatively small amount of bandwidth per
camera. It is capable of handling high-definition television
(HDTV) and has been adopted as the digital television
standard by the FCC and is the compression standard
for DVDs. The MPEG-2 NTSC standard has a resolution
of 720 × 480 pixels and incorporates both progressive
and interlaced scanning although progressive scanning
is rarely used in video security applications. Interlaced
scanning is the method used in the video security industry
to produce images on surveillance monitors.
The MPEG-2 and MPEG-4 are based on the group of
images (GOI) concept as defined by an I-frame, P-frame,
and B-frame (Figure 7-13).
The technology’s basic principle is to compare two compressed image groups for transmission over the network.
The first frame group is called the I-frame (intra-frame),
and uses the first compressed image as a reference frame.
This image serves as the reference point for all frames
following it that are in the same group. Following the
I-frame come the P-frames (predictive), that are coded
with reference to the previous frame and can either be
an I-frame or another P-frame. The P-frames include the
changes, i.e. movement and activity from the leading Iframe. B-frames (bi-directional) are compressed with a low
222
CCTV Surveillance
IMAGE CUT INTO 8 × 8 TILES
8 × 8 TILE FOR EACH COLOR (R, G, B)
VIDEO IMAGE
B
G
R
8 × 8 TILES ZIGZAG SCANNED
AT 64 FREQUENCIES
DC
EACH TILE PROCESSED BY
COMPUTER USING DCT ALGORITHM
DCT CONVERTS 8 × 8 TILE
DC = OVERALL TILE BRIGHTNESS
HIGH FREQUENCY =
DETAILS IN TILE IMAGE
DCT
DISCRETE COSINE TRANSFORM
ACHIEVES COMPRESSION BY DISCARDING INTRAFRAME
SPATIAL AND SPECTRAL (COLOR) REDUNDANCIES
FIGURE 7-12
HIGH FREQUENCIES
JPEG lossless compression technique
bit rate using both the previous and future references
(I and P). B-frames are not used as references. Typical
GOI lengths are usually 12 or 16 frames. The network
viewing stations reconstruct all images based on the reference I images and the difference data in the B- and
P-frames. The detail relationship between the three frame
types are described in the MPEG standard. The MPEG-2
and MPEG-4 can achieve compression ratios up to approximately 60–100 to 1.
7.4.3.3.2 MPEG-4 Standard
The MPEG-4 standard was introduced in 1998 and has
evolved into the first true multimedia and Web compression standard because of its low bit-rate transmission
and incorporation of audio and video with point-and-click
interaction capabilities. The MPEG-4 uses the GOI concept and I-, P-, B-frames but in addition uses object-based
compression where individual objects within a scene are
tracked separately and compressed together. This method
offers a very efficient compression ratio that is scalable
from 20:1 to 300:1. The primary uses for the MPEG-4
standard are web–streaming media, CD distribution, video-
phone, and broadcast television. The MPEG-4 consists of
several standards-termed layers:
• Layer 1 describes synchronization and multiplexing of
video and audio.
• Layer 2 is a compression codec for video signals.
• Layer 3 is a compression codec for perceptual coding
of audio signals.
• Layer 4 describes procedures for testing compliance.
• Layer 5 describes systems for software simulation.
• Layer 6 describes delivery multimedia integration framework.
• Layer 10 is an advanced codec for video signals, also
called H.264.
7.4.3.3.3 MPEG-4 Visual Standard
The MPEG-4 Visual became an international standard in
1999 with its main feature being the support of objectbased compression. Objects in the scene after appropriate
identification (segmentation) can be coded as separate
bit streams and manipulated independently. This is an
important attribute for video security applications. If the
target can be automatically recognized, tracked, and segmented from the scene, it can be coded separately from
Digital Transmission—Video, Communications, Control
223
I, P, AND B FRAMES AND MOTION PREDICTION
I
B
B
P
B
B
P
B
B
P
I = REFERENCE FRAME
P = PREDICTIVE FRAME
B = DIFFERENCE FRAME
• I–FRAME IS ENCODED AS A SINGLE IMAGE WITH NO REFERENCE TO PAST OR FUTURE FRAMES.
• P–FRAME IS ENCODED RELATIVE TO THE PAST REFERENCE FRAME. A REFERENCE FRAME CAN BE A P OR AN I–FRAME
• B–FRAME IS ENCODED RELATIVE TO THE PAST REFERENCE FRAME, THE FUTURE REFERENCE FRAME,
OR BOTH FRAMES. THE FUTURE REFERENCE FRAME IS THE CLOSEST FOLLOWING REFERENCE FRAME (I OR P)
FIGURE 7-13
MPEG-2 and MPEG-4 compressed image frames: Reference I, difference B, and predictive P
and where appropriate, with higher quality (resolution)
than the other areas of the scene.
The MPEG-4 Visual has enhanced functionality compared to MPEG-2. Spatial prediction within I-frames and
enhanced error resiliency are two such features. Improved
prediction and coding separately improve compression by
15–20% compared to MPEG-2. An advanced feature of
MPEG-4 Visual is global motion compensation (GMC).
This is especially useful for PTZ applications and mobile
applications involving moving ground vehicles, aircraft,
and ships, in which camera movement induces most of the
image motion. The GMC mode reduces the motion information change to a few parameters per frame as opposed
to a separate motion vector for each block of the image.
The GMC can lead to significant bit-rate savings in these
PTZ motion applications. The MPEG-4 Visual compressors
and decompressors (CODECS), having both chips and
software, are most often used for the Internet and cell
phone applications.
7.4.3.4 MPEG-4 Advanced Video Coding
(AVC)/H.264
An improvement over MPEG-4 Visual: MPEG-4 Advanced
Video Coding (AVC), also referred to as H.264, offers
greater flexibility and greater precision in motion vectors
(activity in the scene). The intent of the standard was
also to create one that would be capable of providing
good video quality and bit rates that were half or less than
previous standards relative to MPEG-2, H.263, or MPEG-4.
The MPEG-4 AVC/H.264 is the most recent video compression standard introduced in 2003. The AVC was jointly
developed by MPEG and ITU—a developer of video conferencing standards that calls it H.264. The MPEG-4 AVC
achieves better performance than MPEG-2 by about a
factor of two, producing similar quality at half the bit rate.
The improved performance is mainly due to increased
prediction efficiency both within and the between frames.
The MPEG-4, MPEG-4 Visual (with or without GMC), and
MPEG-4 AVC are superior to MPEG-2 in terms of raw efficiency (quality per bit) and are also more network friendly
than MPEG-2.
The H.264 compression system dramatically lowers the
bandwidth (by 2 times) required to deliver digital TV (DTV)
channels and provides new security business models at a
significantly lower cost. Current standard-definition (SD)
and the high-definition (HD) digital video are based almost
entirely on MPEG-2, the 10-year-old standard that has nearly
reached the limit of its video compression efficiency.
The MPEG-4 AVC compression was developed specifically by and for television broadcasting, whether via terrestrial, cable, satellite, or Internet delivery. It uses the
same protocol and modulation techniques as MPEG-2 so
that MPEG-4 AVC is immediately deployable. By using
224
CCTV Surveillance
the same protocol and modulation techniques, MPEG-4
AVC compression reduces the bandwidth by a factor of
two, thus requiring 50% less bandwidth or storage capacity
compared with MPEG-2 to deliver the same video quality.
This means that instead of having to transmit HDTV at
19 Mbps and SD at 4 Mbps, equivalent HD picture quality is obtained at about 8 Mbps, and SP at 2 Mbps, and
DVD quality video at less than 1 Mbps. The technology
offers greater efficiency and reception with cell phones,
PDAs, and specialized pagers. MPEG-4 AVC permits both
progressive and interlaced scanning.
The MPEG-4 AVC reaches compression ratios for low
motion images of 800:1 to 1000:1. With images containing a high level motion, MPEG-4 AVC reaches compression
ratios of 80:1 to 100:1.
7.4.3.5 JPEG 2000, Wavelet
A newer standard for JPEG compression is JPEG 2000
based on wavelet compression algorithms. It has the
potential to provide higher resolution at compression
ratios of 200:1. The JPEG 2000 was created as the successor to the original JPEG format developed in the late
1980s and is based on state-of-the-art wavelet techniques
that provide better compression and advanced system-level
functionality. Wavelet video compression operates on the
entire image at once, rather than on pieces of the image
(Figure 7-14).
Wavelet compression in contrast to JPEG and MPEG
algorithms is based on full-frame information and on
signal frequency components. It does not divide the image
into 8 × 8 pixel blocks but analyzes the entire image as a
single block.
The JPEG 2000 improves download times of the still
image by compressing images to roughly half the size of
JPEG. In addition JPEG 2000 permits viewing “something”
(a low resolution picture) while waiting for the full highresolution picture to develop on the screen. JPEG 2000’s
progressive display initially presents a low-quality image
and then updates the display with increasingly higher
VIDEO IMAGE SCANNED FREQUENCY PLANE
DC TO HIGHEST FREQUENCY
DC
• ALGORITHM CONSISTS OF PAIRS OF HIGH-PASS
AND LOW-PASS FILTERS
• IMAGE ANALYZED BEGINNING WITH DC AND
PROGRESSING TO HIGHEST FREQUENCY
HIGHEST
FREQUENCY
FREQUENCIES ASCEND IN THE ORDER:
LL3, HL3, LH3, HH3, HL2, LH2, HH2, HL1, LH1, HH1
DC
LOW
FREQUENCY
DC
LL 3 H L 3
L H 3 HH 3
LH 2
HL 2
HL 1
HH 2
LH 1
• DWC = DISCRETE WAVELENGTH COMPRESSION
• LL = LOWEST FREQUENCY COMPONENTS
• HH = HIGHEST FREQUENCY COMPONENTS
FIGURE 7-14
Wavelet compression technology
ONLY HIGH FREQUENCY
COMPONENTS ANALYZED
HERE
HH 1
Digital Transmission—Video, Communications, Control
quality images. Wavelet compression is similar to JPEG in
that it uses multiple single recorded frames to create a
video sequence. The average file size for a wavelet image
is about 12 Kbps at 30 fps.
Wavelet compression compresses the entire image with
multiple filtering, and filters the entire image, both high
and low frequencies, and repeats this procedure several
times. There is no mosaic effect once the images are
viewed because the technology works on the entire image
at once.
7.4.3.6 Other Compression Methods:
H.263, SMICT
7.4.3.6.1 H.263 Standard
The H.263 standard was developed for video conferencing using transmission networks capable of rates below
64 Kbps. It works much the same way MPEG-1 and MPEG2 work but with reduced functionality to allow very low
transmission rates. The H.263 is similar to JPEG except
that it only transmits the pixels in each image that have
changed from the last image, rather than full images. Often
the two consecutive images (frames) from a camera are
essentially the same and so the H.263 standard takes advantage of this characteristic and uses a frame differencing
technique that sends only the difference from one frame to
the next.
TYPE
COMPRESSION
TRANSFORM
TRANSFORM
BIT RATE
JPEG
FRAME -BASED
DCT *
8 Mbps
MJPEG
FRAME -BASED
DCT
10 Kbps to
3 Mbps
MPEG -1
STREAM -BASED
DCT
1.5 Mbps
MPEG -2
STREAM -BASED
DCT
MPEG -4
PART 2
STREAM -BASED
H.263
225
7.4.3.6.2 SMICT Standard
The super motion image compression technology
(SMICT) standard has almost the same characteristics of
H.264. Based on redundancy in motion, it combines digital signal processing (DSP) hardware compression, with
CPU software compression. Utilizing an intelligent nonlinear super motion CODEC, SMICT intelligently analyzes
the motion changes in the scene that occurred within
the frame, eliminates the redundant portion of the image
that need not be stored, and compresses the delta (or
change) based on motion. Table 7-8 compares the significant parameters of some of the video compression
techniques.
The MPEG-7 and the MPEG-21 are new standards being
considered.
7.5 INTERNET-BASED REMOTE
VIDEO MONITORING—NETWORK
CONFIGURATIONS
Wired and wireless digital video networks using LANS,
WANS, WiFi, and the Internet have made AVS possible.
The digital video signal must be transmitted from the
camera location to the monitoring location. For the case
of wireless networks there are four basic configurations
that are used: (1) point to point—also known as peer to
RESOLUTION
FRAME
RATE
(fps)
LATENCY
(TIME LAG)
0–5
APPLICATIONS
COMMENTS
STORING STILL
VIDEO FRAMES
NOT SUITABLE FOR
MOTION VIDEO
0–30
LOW
IP NETWORKS
BROADCAST
JPEG PLAYED IN
RAPID SUCCESSION
352 × 288 (PAL)
352 × 240 (NTSC)
UP to 30
MEDIUM
VIDEO CD
SOME DVRS
CIF SIZE,
VHS TAPE QUALITY
2 Mbps to
15 Mbps
720 × 576 (PAL)
720 × 480 (NTSC)
24–30
MEDIUM
HDTV
BROADCAST QUALITY
DCT AND
WAVLET
10 Kbps to
10 Mbps
640 × 480 to
4096 × 2048
1–60
MEDIUM
CCTV WHEN HIGH FRAME
STREAMING VIDEO RATES REQUIRED OR
WHEN SCENE ACTIVITY
INTERNET (WEB)
IS LOW TO MEDIUM
STREAM -BASED
DCT
30 Kbps to
64 Kbps
128 × 96 to
704 × 480
10–15
LOW
TELECONFERENCE VIDEO STREAMING
H.264/AVC
MPEG -4
PART 10
STREAM -BASED
DCT
64 Kbps to
240 Mbps
4096 × 2048
0–30
LOW
HIGH SPEED
VIDEO
NEAR BROADCAST QUALITY.
COMPRESSES VIDEO FAR
MORE EFFICIENTLY THAN
MPEG -4, PART 2
JPEG2000
WAVELET
FRAME -BASED
WAVELET
30 Kbps to
7.5 Mbps
160 × 120
320 × 240
8–30
HIGH
SOME CCTV
RECORDING
LAG, LIMITED USE IN
SECURITY
BROAD RANGE
SMART CARD
MULTI-MEDIA CONTENT
NOT YET IN SECURITY
ANY SIZE
MPEG -7
* DIRECT COSINE TRANSFORM. USES INTRA FRAMES (I), PREDICTED FRAMES (P), AND BI -DIRECTIONAL FRAMES (B). I, P, AND B ARE CALLED GROUP OF PICTURES (GOP).
AVC—ADVANCED VIDEO CODING
Table 7-8
Comparison of Most Common Compression Standards
226
CCTV Surveillance
peer, (2) multi-point to point, (3) point to multi-point,
and (4) mesh. This section describes the four configurations used.
7.5.1 Point to Multi-Point
The point-to-multi-point wireless systems use IP packet
radio transmitters and standard Ethernet interfaces to
enable high-speed network connections to multiple Ethernet switches, routers, or PCs from one single location
(Figure 7-15).
The network cameras can be connected and conveniently located wherever necessary. Transmission capacities vary from 10 to 60 Mbps and operate at distances up
to 10 miles. The point to multi-point (multi-casting) is like
a radio or television station in which one signal (station
or channel) is broadcast and can be heard (or viewed) by
many different users in the same or different locations.
With IP multi-cast, the video server needs to transmit only
a single video stream for each multicast group regardless
of the number of clients that will view the information.
where only a single camera or sensor and a single monitoring location is used and only one to one camera control
functions are required (Figure 7-16).
These systems offer higher capacities and greater distances than the point-to-multi-point systems. They are ideal
for transmitting video signals from a local central site
where a base station is located, to a central command and
control center that is located much farther away. Pointto-point systems can connect to remote sites up to 40
miles away from the monitoring site and have transmission
bandwidth capacities ranging from ten to several hundred
megabits per second.
7.5.3 Multi-Point to Point
The multi-point to point is most commonly used when
multiple video cameras are multiplexed into a central control point. Multi-point-to-point systems transmit the video
signal from multiple cameras to the remote systems monitoring location (Figure 7-17).
7.5.4 Video Unicast and Multicast
7.5.2 Point to Point
Point-to-point wireless video transmission is used in simpler systems to provide connectivity between two locations
A video broadcast sends out a video data packet intended
for transmission to one or multiple nodes on the network. A unicast signal is sent from source to viewer as a
standalone stream and required that each viewer have his
SITE 2
PDA WITH
WIFI CARD
BRIDGE WIRELESS
LINKS: 802.11
SITE 3
SITE 1—BASE STATION
ACCESS
POINT
SERVER
ANALOG
CAMERA
BNC
RJ45
IP
CAMERA
SITE 4
LAN
TOWER
SERVER
IP
DOME
BNC RJ45
ANALOG
PTZ
FIGURE 7-15
Point to multi-point wireless network
SITE 5
LAPTOP
Digital Transmission—Video, Communications, Control
TWO LOCATIONS
SITE 1—BASE STATION
SITE 2
SERVER
ANALOG
CAMERA
BRIDGE WIRELESS
LINK: 802.11
IP
CAMERA
RJ45
PDA
BNC
ACCESS
POINT
LAN
TOWER
LAPTOP
SERVER
IP
DOME
BNC
RJ45
ANALOG
PTZ
FIGURE 7-16
Point to point wireless network
IP CAMERA
DOME
SERVER
SITE 1
BNC
RJ45
ANALOG
CAMERA
10BASE–T
ETHERNET/IP
NETWORK
CENTRAL CONTROL
ACCESS
POINT
SECURITY
IP
CAMERA
S I TE 2
10BASE–T
ETHERNET/IP
NETWORK
SERVER
LAPTOP
BNC RJ45
CPU
CONTROL
ANALOG
PTZ
VIDEO
STORAGE
DOME
IP CAMERA
SITE 3
100BASE–T
FIBER OPTIC
FIGURE 7-17
Multi-point to point wireless network
227
228
CCTV Surveillance
own video viewer. A multicast stream allows multiple viewers on a network to all share the same feed. The benefit is
in bandwidth consumption: for 20 people to view a 1 Mbps
video stream as unicast feeds, they would consume a total
of 20 Mbps of bandwidth 20 × 1 Mbps. If those same 20
viewers connected to the same feed as a multicast stream,
assuming they are all on the same network, they would
consume a total of 1 Mbps of bandwidth (Figure 7-18).
7.6 TRANSMISSION TECHNOLOGY PROTOCOLS:
WiFi, SPREAD SPECTRUM MODULATION
(SSM)
Most wireless LAN systems use spread spectrum technology, a wideband radio frequency technique developed by the military for use in reliable, secure, missioncritical communications systems. Spread spectrum modulation (SSM) is designed to trade off bandwidth efficiency for reliability, integrity, and security. In other
words, more bandwidth is consumed than in the case
of narrowband transmission, but the trade-off produces
a signal that is, in effect, louder and is easier to
detect, provided that the receiver knows the parameters of the spread spectrum signal being broadcast.
If a receiver is not tuned to the right frequency, a
spread spectrum signal looks like background noise
(Figure 7-19).
In contrast to SSM, a narrowband radio system transmits
and receives information at a specific radio frequency. Narrowband radio keeps the radio signal frequency as narrow
as possible, just enough to pass the information. A private
telephone line is much like a narrowband radio frequency.
When each home in a neighborhood has its own private
telephone line, people in one home cannot listen to calls
made to other homes. SSM, privacy and non-interference
are accomplished by the use of separate radio frequencies,
and the radio receiver filters out all radio signals except
the one to which it is tuned.
The first publicly available patent on SSM came from
the inventors Hedy Lamarr, the Hollywood movie actress,
and George Antheil, an avant-garde composer. The patent
was granted in 1942 but the engineering details were a
closely held military secret for many years. The inventors
never profited from their invention, they simply turned
the patent over to the US government for use in the
World War II effort, and commercial use was delayed
VIDEO MULTICAST
1 Mbps SHARED AMONG ALL VIEWERS
LCD
DISPLAY
TOWER
VIDEO
SOURCE
KEYBOARD
VIDEO UNICAST
VIDEO
SOURCE
FIGURE 7-18
1 Mbps FOR EACH VIEWER
Video unicast and video multicast configuration
LAPTOP
Digital Transmission—Video, Communications, Control
229
POWER
CONTINUOUS WAVE
(CW) SIGNAL
SPREAD SPECTRUM
SIGNAL
FREQUENCY SPECTRUM
FIGURE 7-19
Spread spectrum modulation (SSM) compared to narrow band transmission
until 1985. It was initially developed by the military to
avoid jamming and eavesdropping of communication signals. The present global positioning system (GPS), cellular phone, and wireless Internet transmission systems
now represent the largest commercial SSM technology
applications.
The SSM technology provides reliable and secure communications in environments prone to jamming and/or
signals prone to interception by third parties. Most SSM
systems operate in the 900 MHz, 2.4 GHz, and 5.8 GHz
bands, and require no licensing application and ongoing
fees to anyone, providing the strict rules on signal specifications: bandwidth and power output, are adhered to.
The SSM technology is currently the most widely used
transmission technique for wireless LANs. The technique
spreads the digital signal power over a wide range of frequencies within the band of transmission. The bands for
commercial security video transmission range from 902
to 928 MHz, 2.4 to 2.484 GHz, and 5.1 to 5.8 GHz, all of
which do not require an FCC license.
There are two types of spread spectrum radio: frequency
hopping (FH) and direct sequence (DS). In the 1960s
Aerojet General first used the FH concept, the predecessor to SSM for military applications in which the signal
frequencies were rapidly switched. The SSM is a similar
concept to FH only performed at a much faster rate. The
radio signal required very little transmitter power and was
immune to noise and interference from other similar systems employing the exact same carrier frequency. The
radio signal was secure and completely undetectable by
signal spectrum analyzers then available.
7.6.1 Spread Spectrum Modulation (SSM)
7.6.1.1 Background
The purpose of SSM is to improve (reduce) the bit error
rate of the signal in the presence of noise or interference. This is achieved by spreading a transmitted signal
over a frequency range greater than the minimum bandwidth required for information transmission. By spreading
the data transmission over a large bandwidth, the average power level of any one frequency is reduced and less
interference is caused to others in the band. Implemented
appropriately, others will interfere less with the signal even
if others do not employ SSM techniques. While the channel data may be analog or digital, for simplicity a basic
digital system is considered.
Frequency hopping the transmitter repeatedly changes
the carrier frequency from one to another, referred to
as hopping. The hopping pattern is usually controlled by
a pseudo noise (PN) code generator. Any narrowband
interference can only jam the FH signal for a short period
of time in every PN code period.
Direct sequence spread spectrum (DSSS) is the technology in most use today, and spreads the spectrum
by modulating the original signal with PN noise. The
PN is defined as a wideband sequence of digital bits,
called chips that are employed to minimize confusion.
The DSSS receiver converts this wideband signal into
its original, narrow base-band signal by an operation
known as de-spreading. While de-spreading its own signal,
the receiver spreads any narrowband interfering signals,
230
CCTV Surveillance
thereby reducing the interference power in the narrowband detection system.
A typical spread spectrum radio transmitter transmits
a sequence of coding bits, referred to as PN code, and
spreads the signal over a radio spectrum 20 MHz wide per
channel. At the receiver end both the desired and foreign
signals are de-spread to effectively regenerate the desired
signal and suppress the foreign signals. In a typical wireless
LAN configuration, a transmitter/receiver (transceiver)
device, called an access point, connects upstream to the
wired network from a fixed location using standard
cabling.
7.6.1.2 Frequency Hopping Spread Spectrum
Technology (FHSS)
Frequency hopping spread spectrum (FHSS) uses a narrowband carrier that changes frequency and a pattern
known to both the transmitter and receiver. Properly synchronized, the net effect is to maintain a single logical
channel. To an unintended receiver the FHSS appears
to be short duration impulse noise. Figure 7-20 illustrates
how FHSS works.
The FHSS technique broadcasts the signal over a seemingly random series of radio frequencies and a receiver
hops and follows these frequencies in synchronization
while receiving the signal message. The message can only
be fully received if the series of frequencies is known. Since
only the intended receiver knows the transmitter’s hopping sequence, only that receiver can successfully receive
all the signals.
7.6.1.3 Slow Hoppers
With this technique the data signal is transmitted as a
narrowband signal with a bandwidth only wide enough to
carry the required data rate. At specific intervals this narrowband signal is moved or hopped to a different frequency
within the allowed band. The sequence of frequencies follows a pseudo-random sequence known to both the transmitter and the receiver. Once the receiver has acquired
the hopping sequence of the transmitter, one or more
packets are transmitted before the frequency is hopped to
the next channel. Many data bits are transmitted between
hops. This technique is useful for narrowband data radios
but not for wideband video signals.
7.6.1.4 Fast Hoppers
Similar in manner to slow hoppers, fast hoppers make
many hops for each bit of data that is transmitted. In this
way each data bit is redundantly transmitted on several different frequencies. At the receiving end, the receiver need
only receive a majority of the redundant bits correctly in
order to recover the data without error. The real benefit
of the fast hopper is that true process gain is provided by
DWELL-TRANSMIT TIME
FREQUENCY
SLOTS (MHz)
928
TRANSMITTED FREQUENCY
HOPS AS A FUNCTION
OF TIME: f 1, f 2, • • •, f 7, • • •
f2
f7
f5
f3
f1
BLANK-OFF
TIME
f4
f6
902
0
FIGURE 7-20
1
2
3
4
5
Frequency hopping spread spectrum (FHSS) technology
6
7
TIME
Digital Transmission—Video, Communications, Control
the system due to this real-time redundancy of data transmission. This allows interference to exist in the band that
would effectively block one or more narrowband channels
without causing loss of data.
7.6.1.5 Direct Sequence Spread Spectrum (DSSS)
The DSSS method is the most widely used SSM technique
and is currently used in most WiFi systems. The DSSS
increases the rate of hopping so that each data bit can be
even more redundantly encoded (more process gain) or
that a higher bit rate can be transmitted as required in
video signals.
The DSSS generates a redundant pattern for each bit to
be transmitted. This bit pattern is called a chip (or chipping code). It follows that the longer the chip, the greater
the probability that the original data can be recovered
and, of course, the more bandwidth required. Even if one
or more bits in the chip are damaged during transmission,
statistical techniques embedded in the radio can recover
the original data without the need for retransmission. To
an unintended receiver, DSSS appears as low-power wideband noise and is rejected (ignored) by most narrowband
receivers. Figure 7-21 illustrates how this technology works.
The FCC rules on signal specifications limit the practical data throughput for DSSS protocol to 2 Mbps in the
902 MHz band, 8 Mbps in the 2.4 GHz band, and 100 Mbps
SIGNAL
VOLTAGE
in the 5.8 GHz band. The FCC also requires that transmitters must hop through at least 50 channels in the 902 MHz
band and 75 channels in the 2.4 GHz band.
The DSSS transmitters spread their transmissions by
adding redundant data bits called “chips” to them. The
DSSS adds at least 10 chips to each data bit. Once a
receiver has received all of the signal and chip bits, it uses
a correlator to remove the chips and collapses the signal
to its original length. The IEEE 802.11 standard requires
11 chips for DSSS transmission.
The DSSS system can operate when other systems
such as microwave radio, two-way communications devices,
alarm systems, and/or other DSSS devices are transmitting
in close proximity. It also has the ability to select different
channels to provide workarounds on the rare occasions
that interference occurs.
The magical and non-intuitive element of the DSSS system breakthrough is that by multiplying the PN DDSS
spread signal with a copy of the same pseudo noise,
the original data signal is recovered. This process is
called correlation and only occurs if the codes are identical and perfectly aligned in time to within a small fraction of the code clock. By using concurrently different
pseudo-random codes, multiple independent communications links can simultaneously operate within the same
frequency band. To recover the specific encoded data
channel, the inverse function is applied to the received
signal. A major breakthrough in DSSS came when it was
“ONE”
DATA BIT
1
“ZERO”
DATA BIT
0
TIME
0
1
MAXIMUM HOP RATE:
• 2 Mbps IN 902 MHz BAND
• 8 Mbps IN 2.4 GHz BAND
• 100 bps IN 5.8 GHz BAND
10-CHIP CODE
WORD FOR EACH
“ONE” DATA BIT
0
TIME
0
1
SAME CHIP CODE
WORD BUT INVERTED
FOR “ZERO” DATA BIT
0
0
FIGURE 7-21
231
Direct sequence spread spectrum (DSSS) technology
TIME
232
CCTV Surveillance
capability of carrying a large bandwidth of data, specifically
video image transmissions for surveillance applications.
realized that a pseudo-random digital code or pseudorandom noise contains the frequencies from DC to that
of the code clock rate. When the narrowband data signal
is multiplied by the pseudo-random code sequence, the
spectrum of the signal is spread to a bandwidth twice that
of the code (Figure 7-22).
The amount of performance improvement that is
achieved against interference is known as the processing
gain of the system. An ideal estimate for processing gain is
the ratio of the spread spectrum bandwidth to the signal
information rate:
Processing Gain =
7.6.2 WiFi Protocol: 802.11 Standards
Using a wireless LAN (WLAN, WiFi) dramatically reduces
the time and cost of adding PCs and laptops to an established network. For a small or medium company, a complete wireless network can be set up within hours, with
minimal disruption to the business. A laptop or PDA with
WLAN allows mobile employees to be more productive by
working from public “hotspots,” at airports, hotels, etc.
Among the most fundamental steps to take when planning a WLAN is to learn about the various IEEE 802.11
standards, decide which one is appropriate for the application requirements, and apply it according to the standard.
The WiFi Alliance is responsible for awarding the WiFi
certified logo that ensures 802.11 compatibility and multivendor interoperability. The original 802.11 PHY (physical) standard established in June 1997 defined a 2.4 GHz
system with a maximum data rate of 2 Mbps. This technology still exists but should not be considered for new
deployment. In 1999 the IEEE defined two additions to
the 802.11 PHY, namely 802.11b and 802.11a.
There are two basic categories of IEEE 802.11 standards.
SSM Bandwidth
Signal Bandwidth
It is important to note that data rate (signal bandwidth)
and process gain are inversely proportional. In a digital
data system, the process gain can be directly determined
by the ratio of the pseudo-random code bits, called chips,
and data or symbol rate of the desired data. For example,
a system that spreads each symbol by 256 chips per symbol
has a ratio of 256:1. The process gain is generally expressed
in dB, the value of which is determined by the expression:
P gain in dB = 10 Log Base 10 Chips/Symbol
This corresponds to 24 dB for the example of 256
chips/symbol.
In any case, the SSM technique results in a system that
is extremely difficult to detect by observers outside the
system, does not interfere with other services, and has the
1. The first are those that specify the fundamental protocols for the complete WiFi system. These are called
802.11a, 802.11b, and 802.11g standards and the new
802.11n standard.
POWER
FHSS
DSSS
DSSS
HOP
#25
HOP
#12
HOP
#8
HOP
#5
HOP
#60
HOP
#1
FHSS
FRF – R C
FIGURE 7-22
FCH
FRF
Direct sequence spread spectrum (DSSS) modulation signal
FRF + RC
FREQUENCY
Digital Transmission—Video, Communications, Control
233
2. Second, there are extensions that address weaknesses
that provide additional functionality to these standards.
These are 802.11d, e, f, h, i, and j. Only the 802.11i and
802.11e standards relating to quality of service (QoS)
security are considered.
other technologies this rate may be reduced further due
to interference issues.
Table 7-9 shows the parameters of these fundamental
802.11 standards.
Each of these standards has unique advantages and disadvantages. Their specific attributes must be considered
before choosing one.
The 802.11a technology uses the 5 GHz radio spectrum to
deliver data at a rate of 54 Mbps, and allows for 12 channels
to be used simultaneously. The 802.11a standard occupies
300 MHz in three different bandwidths of 100 MHz each:
7.6.2.2 802.11a Standard
7.6.2.1 802.11b Standard
1. 5.150–5.250 GHz, lower band
2. 5.250–5.350 GHz, middle band
3. 5.725–5.825 GHz, upper band.
The 802.11b technology uses the 2.4 GHz radio spectrum
to deliver data at a rate of 11 Mbps, and allows for three
non-overlapping channels to be used simultaneously. The
802.11b standard occupies 83.5 MHz (for North America)
from 2.4000 to 2.4835 GHz. The standard 802.11b should
be considered if there is no high bandwidth requirement,
i.e. near real-time video is not required but there is a need
for a wide coverage area. If price is a primary consideration
the 802.11b system costs roughly one quarter as much
has an 802.11a network covering the same area at the
same data rate. Its main disadvantage is its lower maximum
link rate. Also since it occupies the 2.4 GHz band used by
Table 7-10 lists nine (4 non-overlapping) 20 MHz bandwidth channels available in the 5.8 GHz band.
The 802.11a standard should be considered if the application requires high bandwidth, as required in high frame
rate video transmission. It also should be considered when
there is a small, densely packed concentration of users.
The greater number of non-overlapping channels allows
access points to be placed closer together without interference. Two disadvantages of the 802.11a standard is that
it is not backward compatible with the older 802.11b standard, and costs roughly four times as much to cover the
same area.
IEEE
STANDARD
OPERATING
FREQUENCY
(GHz)
DOWNLOAD
SPEED *
(Mbps)
BANDWIDTH
(MHz)
802.11a
5.8
54
TOTAL: 300
EACH CHANNEL: 20
802.11b
2.4
11
TOTAL: 83.5
EACH CHANNEL: 22
CHANNELS
APPLICATIONS/COMMENTS
12
12 NON-OVERLAPPING
HIGH BANDWIDTH, HIGH FRAME RATE
MANY NON-OVERLAPPING CHANNELS
11
3 NON-OVERLAPPING
LOW INTERFERENCE IN AREA
REALTIME VIDEO NOT REQUIRED
LOW COST
DEFINES QUALITY of SERVICE (QoC)
802.11e
—
—
—
802.11g
2.4
11, 54
EACH CHANNEL: 22
3 NON-OVERLAPPING
12 NON-OVERLAPPING
—
—
—
108
200
—
SUPPORTS
MIMO DEPLOYMENT
802.11i
802.11n
—
—
—
* THEORETICAL MAXIMUM RATES. REALISTIC MAXIMUM APPROXIMATELY 1/2.
IEEE—INSTITUTE of ELECTRICAL and ELECTRONIC ENGINEERS
MIMO—MULTIPLE–IN MULTIPLE–OUT
Table 7-9
IEEE 802.11 a, b, g, i, and n WiFi Standard Characteristics
(a) BANDWIDTH
(b) LATENCY
(c) JITTER
(d) SIGNAL LOSS
WIDE–AREA COVERAGE
HIGH BANDWIDTH
DUAL BAND
BACKWARD COMPATIBLE WIH 802.11b
ENHANCED SECURITY–
AUTHENTICATION PROTOCOL
IMPROVED SECURITY KEY
ADDS HIGH LEVEL AES ENCRYPTION
NEWEST STANDARD: HIGH DATA RATE
AND BANDWIDTH. HIGH THROUGHPUT
UP TO 600 Mbps. SUPPORTS MIMO
DEPLOYMENT.
234
CCTV Surveillance
CHANNEL
NUMBER
FREQUENCY
(GHz)
1
5.735
1A
5.745
2
5.755
2A
5.765
3
5.775
3A
5.785
4
5.795
4A
5.805
5
5.815
BAND *
UNII UPPER BAND
MAXIMUM
POWER
OUT
800 mW
MODULATION
METHOD
CHANNELS
COFDM
9 MAXIMUM
4 NON–OVERLAPPING
* 802.11a OCCUPIES 300 MHz IN THREE DIFFERENT BANDWIDTHS OF 100MHz EACH
TOTAL OF 9 CHANNELS AVAILABLE: 4 NON-OVERLAPPING.
COFDM—CODED ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING.
UNII—UNLICENSED NATIONAL INFORMATION INFRASTRUCTURE
Table 7-10
Wireless Transmission Channels in 5.8 GHz Band
7.6.2.3 802.11g Standard
The 802.11g technology uses the 2.4 GHz radio spectrum
to deliver data at a rate of 54 Mbps, and allows for three
channels to be used simultaneously. The 802.11g standard is applicable to high-bandwidth video applications
that require wide-area coverage. It should also be considered if backward compatibility with 802.11b is required.
The main disadvantage of 802.11g is that maximum data
throughput is reduced when 802.11g and 802.11b equipment shares the same network. Since it shares the 2.4 GHz
frequency spectrum used by microwave ovens, cordless
phones, garage door openers, and other wireless gadgets,
it faces the same interference issues as 802.11b.
Manufacturers such as Intel are supplying chipsets that
include the IEEE 802.11a, b, and g technologies so that
PCs and laptops can continue to connect to corporate
wireless LANs without a hardware upgrade requirement
even if the enterprise upgrades to a new infrastructure.
7.6.2.4 802.11n Standard
The new 802.11n WiFi standard has high throughput and
was created to provide over 100 Mbps effective throughput,
complementing all broadband access technologies including fiber optic, DSL, cable, and satellite. The goal of the
802.11n protocol standard is to increase the 54 Mbps transmission to over 100 Mbps.
The goal of the newest generation 802.11n standard
more than triples the real throughput of WiFi and pushes
the 30 Mbps standard to at least 108 Mbps. The new
802.11n standard including MIMO processing in its specification should produce performance of 144–200 Mbps.
Figure 7-23 compares the throughput and distance
improvements using the MIMO-based wireless LAN.
7.6.2.5 802.11i Standard
The 802.11i standard provides enhanced security for wireless transmissions. It includes the use of authentication
protocol, an improved key distribution framework, and
stronger encryption via AES.
7.6.3 Asynchronous Transfer Mode (ATM)
Two common protocols adopted to transmit video, voice,
data, and controls over the Internet are the IP and asynchronous transfer mode (ATM). The ATM is a broadband
network technology that allows very large amounts of data
to be transmitted at a high rate (wide bandwidth). It does
this by connecting many links into a single network. This
feature has an important implication for transmitting highquality video with a guaranteed QoS.
The ATM was developed in concept in the early 1980s.
Since the early 1990s ATM has been highly touted as the
ultimate network switching solution. This is because of
its high speed and its ability to serve video and all other
information types, and its ability to guarantee each type an
appropriate QoS. The ATM is a fast-packet, connectionoriented, cell-switching technology for broadband signals.
It has been designed from concept up, to accommodate any form of information: video images, voice, facsimile, and data whether compressed or uncompressed
at broadband speeds and on an unbiased basis. Further,
all such data can be supported with a very small set of
Digital Transmission—Video, Communications, Control
235
DATA RATE/RANGE IN TYPICAL BUSINESS INDOOR ENVIORNMENT
MAX RELIABLE
RATE (Mbps)
120
100
802.11n
80
60
802.11a
40
20
0
0
20
40
60
80
100
120
140
160
180
200
RANGE (Ft)
MIMO = MULTIPLE IN/MULTIPLE OUT
FIGURE 7-23
Rate/range comparison of 802.11a vs. 802.11n MIMO indoors
network protocols, regardless of whether the network is
local, metropolitan, or wide area in nature. The ATM generally operates at minimum access speeds of 50 Mbps up
to 155 Mbps. The ATM has, however, been slow to be
accepted, is clearly on the rise, but it is a long time away
before it may ultimately replace all of the circuit-, packet-,
and frame-switching technologies currently in place.
7.7 TRANSMISSION NETWORK SECURITY
The WLANs transmit video and data over the air using
radio waves. Any WLAN client in the area served by the
data transmitter can receive or intercept the information signal. Radio waves travel through ceilings, floors,
and walls and can reach unintended recipients on different floors and outside buildings. Given the nature of the
technology there is no way to assuredly direct a WLAN
transmission to only one recipient.
Users must be conscious of security concerns when planning wireless 802.11 networks. The first step of WLAN
security is to perform a network audit to locate rogue
access points within the network. The second step involves
the basics of configuring and implementing the best security practices at all access points of the WLAN. In 2001,
researchers and hackers demonstrated their ability to
crack wired equivalency policy (WEP), a standard encryption for 802.11 wireless LANs. Because these encryption
and authentication standards were vulnerable, stronger
methods were developed and should be deployed to more
completely secure a WLAN. The 802.11i standard has
accounted for weaknesses in previous protocols but is still
subject to some vulnerability if improperly implemented
or by-passed by rogue devices.
Every enterprise network needs a policy to ensure
security on the network, and WLANs are no different.
While policies will vary based on individual security and
management requirements of each WLAN, a thorough
policy and enforcement of the policy can protect an
enterprise from unnecessary security breaches and performance degradation.
7.7.1 Wired Equivalent Privacy (WEP)
The IEEE 802.11 WLAN standards include a security component called wired equivalent privacy (WEP). The WEP
defines how clients and access points identify each other
and communicate securely using secret keys and encryption algorithms. Although the algorithms used are well
understood and not considered vulnerable, the particular
way in which the keys are managed has resulted in a number of easily exploitable weaknesses. The WEP security
relies on the user name/password method. Many WLAN
access points are shipped with the WEP security disabled by
default. This allows any WLAN-enabled device to connect
236
CCTV Surveillance
to the network unchallenged. However, even when WEP is
enabled there are still ways to breach the security; it just
takes a little longer. As a first basic layer of security it is
imperative that network administrators turn WEP “ON”
prior to deploying access points in the corporate network.
Most enterprises using wireless LANs do not enable the
WEP and consequently users should presume that any data
sent over such a wireless link can be intercepted. Furthermore with WEP now cracked by malicious hackers, organizations must explore additional measures including virtual
private networks (VPNs) and vendor specific authentication schemes to provide more robust protection of the
data passed over the wireless link.
Wireless LAN signals do not necessarily stop at the outer
walls of a building, a corporate campus border, or a physical plant perimeter. Physical security is ineffective in protecting against wireless LAN intrusions. In some metropolitan areas, hackers armed with portable computers or even
PDAs with LAN cards make a game of drive-by invasions of
corporate networks. As a first step, existing wireless LANs
should be checked to ensure that WEP security protection
is enabled.
7.7.2 Virtual Private Network (VPN)
Network architects considering WLAN deployments must
look beyond current WEP technology to ensure that security is not compromised. Currently the “best practices” recommendations are to overlay a VPN on top of the WLAN
to establish an encrypted tunnel for users and devices
to exchange sensitive information securely. Many current
out-of-the-box VPN products support alternate methods
for authenticating users and devices such as the use of
digital IDs. It is extremely important to take advantage of
enhanced identification methods for VPN, as a high level
of trust is needed to grant users full access to security information. Companies must invest in products that provide
secure identification and authentication capabilities with
the VPN.
Having a VPN overlay and basic security with a WLAN
is comparable to having a security guard in the lobby of
a building. The guard calls to let you know that John
Doe is there to see you. If you are expecting him you
let him through. But, is he who he who really says he is
and how would you know until you saw him walk through
the door? The security guard alone still leaves the hole
in the system. But if the security guard must check John
Doe’s passport (or a credential he knows to be authentic),
there is no way he is coming in without authenticated
documentation to prove his identity. Likewise to achieve
mutual authentication, the security guard must present
his or her own passport to Mr. Doe, so he knows he is
at the correct building and not about to meet with an
impostor.
To deploy a VPN, the WLAN access point is placed outside the firewall and a VPN gateway is placed between
the two. Since the WLAN access point is outside the firewall, it is effectively being treated as an untrustworthy
network resource since it blurs the security parameter.
Even if WEP security is compromised, no access to corporate resources is possible without a subsequent authenticated VPN.
Most enterprises deploying wireless LANs will be forced
to embrace vendor-specific security architecture or use
VPNs. A VPN cannot be used everywhere in the wireless
LAN architecture due to lack of VPN client support from
manufacturers on certain handheld devices and proprietary operating systems.
7.7.3 WiFi Protected Access (WPA)
WiFi protected access (WPA) is an interim standard developed by the WiFi Alliance. It combines several technologies that address known 80211× security vulnerabilities.
It provides an affordable, scalable solution for protecting
existing corporate WLANs without the additional expense
of the VPN/firewall technology. It includes the uses of the
80211× standard in the extensible authentication protocol. For encryption it uses the temporal key integrity
protocol and WEP 128-bit encryption keys. The WPA is
a subset of the 802.11i standard. The WPA interim standard upgrades legacy systems and is an improvement over
the WEP system. After upgrading to the WPA standard,
firewalls and VPNs are no longer necessary. The national
Institute of standards and technology (NIST) will not certify WPA under the FIPS 140-2 security standard. The federal government is mandated to procure a new system that
conform to the FIPS 140-2 security standard and will not
certify WPA onto this new standard.
7.7.4 Advanced Encryption Standard (AES),
Digital Encryption Standard (DES)
The data encryption standard (DES) is probably the most
popular secret-key system in use on wired networks today.
The much trickier triple DES is a special mode of DES
that is used primarily for highly sensitive information.
Triple DES uses three software keys. Data is encrypted
with the first key, decrypted with the second key, and then
encrypted again by the third key. The security chips used
in equipment contain a triple-DES encryption/decryption
engine that secures the content, avoiding troublesome
theft-of-service issues for content providers. Moreover it
prevents accidental viewing by another receiver sensor
since it locks the data stream to a particular receiver. It
provides capability for additional network entry, authentication, and authorization.
Digital Transmission—Video, Communications, Control
The advanced encryption standard (AES) was selected
by NIST in October 2000 as an upgrade from the previous DES standard. The AES uses a 128-bit block cipher
algorithm and encryption technique for protecting digital
digital information. With the ability to use even larger 192bit and 256-bit keys, if necessary, it offers higher security
against brute-force attack than 56-bit DES keys. The 128bit key size AES standard makes hacking of data nearly
impossible.
The AES is replacing both triple DES on wired networks
and WEP on wireless LANs. For wireless networks, AES
is being built into equipment complying with the new
802.11i protocol.
7.7.5 Firewalls, Viruses, Hackers
A firewall can be a software program, a hardware device,
or a combination of both. Basically a firewall is a system
or group of systems that enforces an access control policy
between two networks. The term “firewall” has become
commonplace in discussions of network security. While
firewalls certainly play an important role in securing the
network, there are certain misconceptions regarding them
that lead people to falsely believe that their systems are
totally secure once they have a firewall. Firewalls are effective against attacks that attempt to go through the firewall,
but they cannot protect against attacks that don’t go through
the firewall. Nor can a firewall prevent individual employees with modems from dialing into or out of the network,
bypassing the firewall entirely.
The purpose of the firewall is to protect networked
computers from intentional hostile intrusion from outside
the network. Any private network that is connected to a
public network needs firewall protection. Any enterprise
that connects even a single computer to the Internet via a
modem should have personal firewall software.
What can the firewall protect against? Generally firewalls
are configured to protect against unauthenticated interactive logins from outside the network. Firewalls help prevent pranksters and vandals from logging into the network
computers. A firewall examines all traffic routed between
two networks to see if the traffic meets certain criteria.
There are two distinct types of firewalls that are commonly
used: (1) the packet filtering router and (2) the proxy
server. The first type of firewall, the packet filtering router,
is a machine that forwards packets between two or more
networks. It works on a set of rules and codes and decides
whether to forward or block packets based on the rules
and codes. The second type of firewall, the proxy server,
has had the normal protocols FTP (file transfer protocol)
and Telnet replaced with special servers. It relies on special
protocols to provide authentication and to forward packets. In some instances the two types of firewalls are combined so that a selected machine is allowed to send packets
through a packet filtering router onto an internal network.
237
7.8 INTERNET PROTOCOL NETWORK CAMERA,
ADDRESS
The fastest-growing technology segment in the video security industry is that of networked or IP addressable cameras and associated equipment. As the video industry shifts
from traditional legacy analog CCTV monitoring to an
OCTV networking system, IP cameras with internal servers
are going to completely change the way surveillance is
configured (Figure 7-24a,b).
The camera configurations and set up and viewing of
video images will be done via a LAN, WAN, MAN, or
WLAN backbone, and a standard Web browser. Some security equipment manufacturers are referring to the next
generation of video as Internet protocol television (IPTV).
The devices making up a digital video surveillance system are comprised of an IP network camera, a video server,
and PC or laptop computer. In portable surveillance applications the laptop, PDA, and cell phone are the monitoring devices. The following sections describe each of these
devices and what functional part they play in the overall
camera surveillance and control functions.
The industry offers two different methods for networking cameras. The first method is that of incorporating
an IP addressable camera into an existing LAN, WAN, or
MAN configuration (Figure 7-25).
In this method each camera is assigned a static IP
address. With proper security codes or passwords this video
information can be viewed on a standard Web browser on
the network. These IP cameras with their built-in servers
generally have capability for four video inputs. At the
receiving and monitoring location there are two choices:
(1) the system converts the video back into an analog format so that it can be displayed and/or recorded on an analog display and recorder, or (2) the video remains in digital form and is directly displayed on an LCD, PC, or laptop,
and recorded on a digital video recorder (DVR). The second method for implementing remote or networked cameras is adapting the existing or standard legacy cameras
and configured systems into a local network (Figure 7-26).
The video outputs from the cameras, matrix switchers,
and digital recorders are sent via interface adapters onto
the input of the LAN, WAN, WLAN, or Internet network.
The system starts as a standard security system before the
video outputs and system control lines are connected to a
standalone or plug-in Ethernet network interface unit.
The security industry is transitioning from an analog to
digital system by transporting the digital video images over
an IP-based network using IP cameras as the video image
source. Networked cameras can connect directly into the
existing network via an Ethernet port, and eliminate the
coaxial or UTP cabling that is required for analog cameras
(Figure 7-27).
When computers are already in place, no additional
equipment is needed for viewing the video image from
the network camera. The camera output can be viewed
238
CCTV Surveillance
(A) LEGACY ANALOG
P/T/Z
MONITOR
CAT 3,5
RG59 COAX
ANALOG CAMERAS
UTP
SINGLE PAIR
RG59 COAX
RG59 COAX
ANALOG
DOME
RG59 COAX
BNC
MULTIPLEXER
LAN
CAT 3,5
FIBER OPTIC
INTERNET
INTRANET
WAN
WLAN
(B) IP DITIGAL
SERVER
CAT 3,5
BNC
UTP
ANALOG
CAMERA
TOWER
UTP
CAT 3,5
LCD MONITOR
UTP
IP CAMERAS
IP
DOME
FIGURE 7-24 (a) Analog CCTV with coaxial, UTP, or other cabling, (b) Digital IP cameras and digital video server
on wired LAN network
in its simplest form on a Web browser and the computer monitor. If analog cameras are already present at
a site the addition of the video server will make those
camera images available in any location. To connect to
the Internet many different kinds of transmission types
are available. These include standard and ISDN modems,
DSL modems, cable TV modems, T1 connections, and
10BaseT and 100BaseT Ethernet connections. In addition,
cellular-phone modems and various 802.11 wireless network options are also available.
7.8.1 Internet Protocol Network Camera
The network camera has its own IP address and built-in
computing functions to handle any network communication (Figure 7-28). Everything needed for viewing images
over the network is built into the camera unit. The network camera can be described as a camera and a computer combined. It is connected directly to the network
as any other network device and has built-in software for
a web server. It can also include alarm input and relay
output. More advanced network cameras can be equipped
with functions such as motion detection and analog video
output.
An IP compliant network camera contains a lens, a video
imaging chip, a compression chip, and a computer. The
network camera lens focuses the image onto a CCD or
CMOS sensor that captures the image scene and digital
electronics transforms the scene into electrical signals. The
video signals are then transferred into the computer function, where the images are compressed and sent out over
the network (Figure 7-29).
For storing and transmitting images over the network, the
video data must be compressed or it will consume too much
disk space or bandwidth. If bandwidth is limited the amount
of information being sent must be reduced by lowering the
frame rate and accepting a lower image quality.
7.8.2 Internet Protocol Camera Protocols
To facilitate communications between devices on a network they must be properly and uniquely addressed. Just
as the telephone companies must issue phone numbers
that are not duplicated, the computers and devices on the
Digital Transmission—Video, Communications, Control
239
INTERNET
INTRANET
WAN
WLAN
CAT 3,5
FIBER OPTIC
LAN
UTP
CAT 3,5
DOME
TOWER
PTZ
LCD
MONITOR
DVR
HARD DISK
STORAGE
IP CAMERAS
FIGURE 7-25
Incorporating IP cameras in an existing LAN, WAN, or MAN
network must be carefully programmed so that data transmissions can be transmitted and received from one to the
other. Each network device has two addresses: (1) media
access control (MAC) physical address and (2) IP logical
address. The MAC addresses are hard-coded into a device
or product at the factory (manufacturer) and typically are
never changed. The IP addresses are settable and changeable, allowing networks to be configured and changed.
The IP address uniquely identifies a node or device just
as a name identifies a particular person. No two devices
on the same network should ever have the same address.
There are two versions of IP existing in use today. Most
networks now use IP version 4 (IPv4) but new systems will
begin to use the next-generation IP version 6 (IPv6), a
protocol designed to accommodate a much larger number
of computer and device address assignments.
The Internet Corporation for Assigned Names and Numbers (ICANN) is a non-profit organization formed in 1999 to
assume responsibilities from the federally funded Internet
Assigned Numbers Authority (IANA) for assigning parameters for IPs, managing the IP address space, assigning
domain names, and managing root server functions. The
ICANN assigns IP addresses to organizations desiring to
place computers on the Internet. The IP class and the
resulting number of available host addresses an organization receives depends on the size of the organization.
The organization assigns the numbers and can reassign
them on the basis of either static or dynamic addressing.
Static addressing involves the permanent association of an IP
address with a specific device or machine. Dynamic addressing assigns an available IP address to the machine each time
a connection is established. As an example, an Internet Service Provider (ISP) may hold one or more Class C address
blocks. Given the limited number of IP addresses available,
the ISP assigns an IP address to a user machine each time
the dial-up user accesses the ISP to seek connection to the
Internet. Once the connection is terminated, that IP address
becomes available to other users.
7.8.3 Internet Protocol Camera Address
Unlike traditional analog CCTV systems, network video is
based on sets of transmission standards and protocols. These
rules are necessary because the video system is no longer a
closed system but an open system interconnecting with many
clients and users. There are two primary sets of standards
that control networking: (1) 802 created by the IEEE and
(2) Open Systems Interconnect (OSI) seven-layer model,
created by the International Organization for Standardization (IOS). The following sections summarize the standards.
240
CCTV Surveillance
INTERNET
INTRANET
WAN
WLAN
CAT 3,5
FIBER OPTIC
LAN
CAT 3,5
IP/ETHERNET NETWORK
VIDEO
SERVER
ANALOG
CAMERAS
MONITOR
BNC
UTP
SINGLE PAIR
RG59
COAX
RG59
COAX
CAT 3,5
RG59
COAX
CAT 5
RG59
COAX
PTZ
UTP
1
4
8
12
16
SWITCHER / MULTIPLEXER
ANALOG
DOME
RG59
COAX
VCR / DVR
CAMERA /LENS
CONTROLS
FIGURE 7-26
Diagram to connect legacy analog cameras to the digital network
The OSI seven-layer model is the standard cited in
almost all network documents and is the central part of
any network foundation. Although all the OSI layers are
necessary for communication, the four considered in this
analysis are: (1) Physical, (2) Data Link, (3) Network, and
(4) Transport. Figure 7.30 summarizes the seven layers of
the OSI networking model.
The Physical Layer 1 deals with the hardware of the
system. This includes items like servers, routers, hubs, network interface cards, etc. This physical layer has the function of converting digital bits into electronic signals and
connecting the devices to the network.
The Data Link Layer 2 provides the interface, or link,
between the higher layers in the network hardware. The
Data Link Layer has three functions: (1) make sure
a connection is available between two network nodes,
(2) encapsulate the data into frames for transmission, and
(3) ensure that incoming data is received correctly by performing some error checking routines. Layer 2 is divided
into two sub layers: logical ink control and media access control.
The media access control layer is better known by MAC
which is a hard-coded address assigned to every network
interface on any device made to attach to a network. This
address is assigned by the manufacture of the device. The
MAC addresses are unique throughout the entire world.
The address itself is a 48-bit address, consisting of six octets
(eight-digit numbers, in binary).
Connections between devices on a network are ultimately made by MAC address, not IP addresses or domain
names. Those methods simply assist a device in finding
the MAC of another device. The first part of the MAC
address, or the first three of octets, is unique to the manufacturer of the device. It is called the organizational unique
identifier. Every company manufacturing network devices
Digital Transmission—Video, Communications, Control
ANALOG
CAMERA
241
SERVER
CONNECTION OPTIONS:
BNC
RJ45
COAX
• ISDN MODEM
• DSL MODEM
• CABLE TV MODEM
LAN
• T1 CABLE
• 10BaseT ETHERNET
UTP
CAT 3,5
DOME
PC
LCD
MONITOR
INTERNET
• 100BaseT ETHERNET
• FIBER OPTIC
• 802.11 WIRELESS
• CELLULAR WIRELESS
IP CAMERAS
FIGURE 7-27
Diagram to connect networked cameras, Ethernet and Internet
(A) FIXED
FIGURE 7-28
(B) PAN / TILT/ ZOOM
IP network camera
has one or several. The second part of the MAC or the
last three octets, is unique to each device. No two devices
in the world should have the same MAC address.
The second sub-layer in Layer 2 is the logical link control
(LLC). The LLC takes the raw data bits from the upper layers and encapsulates them in preparation for transmission.
It organizes the data into frames, giving information such as
addressing, error checking, etc. After framing and addressing is complete, the frames are then sent to Layer 1 to be
converted into electrical pulses and sent across the network.
Layer 3 is the Network layer and is primarily responsible
for two functions: addressing and routing. This layer contains the IP protocol, part of the TCP/IP protocol. The
“IP address” common to all of us is the Layer 3 responsibility and is unique throughout the entire world. The IP
address is a 32-bit address and must be assigned by a user
or administrator somewhere and is not set at the factory.
Since it is user assignable there is great flexibility in how
the address is assigned. The IP address consists of four
sets of numbers separated by periods or dots, however,
computers actually see the IP address in binary form. The
current IP address format is called IP version four, or IPv4,
in which there are over 4.3 billion possible addresses.
The last OSI level considered here is the Transport
Layer 4. This layer is responsible for reliably getting the
packets from point A to point B. This layer supports
242
CCTV Surveillance
COLUMN/ROW
PIXEL
SCANNING
TIMING AND
SYNCHRONIZING
SDRAM / FLASH
MEMORY
LENS
VIDEO SIGNAL
COMPRESSION
MJPEG, MPEG-4
SENSOR:
CCD, CMOS
CENTRAL
INTERNET
INTRANET
LAN/WAN
WIRED ETHERNET
INTERFACE PORT
PROCESSING
UNIT
IRIS, ZOOM,
P/ T, SHUTTER
WIRELESS PORT
DSP
LENS, P/ T
DRIVERS
NTSC /PAL
OUTPUT
WIFI
802.11 a /b/g
LCD DISPLAY
ALARM TRIGGERS
ALARM OUTPUTS
FIGURE 7-29
IP network camera block diagram
FIGURE 7-30 Seven layer
open systems interconnect
(OSI) model
LAYER
APPLICATION
7
PRESENTATION
6
UPPER
LAYER
(SOFTWARE)
SESSION
5
TRANSPORT
4
NETWORK
(PACKET *)
3
DATA LINK
(FRAME *)
2
1
LOWER
LAYER
(HARDWARE)
PHYSICAL
(BIT *)
* EXCHANGE UNIT
two different transmission methods: connection-oriented
and connectionless. Connection-oriented transmissions are
handled by TCP. These are point-to-point connections for
guaranteed reception of data. An email message, accessing a Web page or downloading a file are all examples of
connection-based exchanges. Error checking is performed
on these exchanges because there is a guarantee of data
reception. This transmission method does not work well
for video since video is near real-time requiring large
amounts of data to be transmitted and it would fail to
produce an acceptable stream of video images for viewing
or recording. If an error occurred and the sending
device retransmitted parts of the video clip the video
stream would not be viewable. For video transmission
the connectionless protocol user datagram protocol (UDP),
which does not guarantee delivery of error-free data, is
Digital Transmission—Video, Communications, Control
used. The UDP is the foundation of video multitasking,
which is a one-to-many method of video streaming. It is a
crucial element of networked video systems.
In spite of the large number of addresses possible in the
IPv4 standard, the popularity of TCP/IP protocol, especially the IP-based Internet, has placed a good deal a strain
on the IPv4-based numbering scheme. To alleviate this
problem, at least partially, in 1993 the concept of supernetting (subnetting) was devised. This technique used the
number of 1 bits in the network address to specify the
subnet mask. This technique reduced the number of routes
and therefore the size and complexity of the routing tables
that the Internet switches and routers had to support. This
subnet technique goes a long way toward easing the pressure on the IPv4 addressing scheme but does not solve
the basic problem of the lack of addresses in the future.
The new IPv6 protocol resolves this issue through the
expansion of the address field to 128 bits, thereby yielding
virtually unlimited potential addresses.
A proper IP address consists of four sets of numbers,
separated by periods or dots. Each of the four sets of
numbers is called an octet. The addressing architecture
defines five address formats each of which begins with
one, two, three, or four bits that identify the class of the
network. The host portion of an IP address is unique for
each device on a network while the network portion is
the same on all devices that share a network. The way to
distinguish which part of an address is which is called the
subnet mask. The subnet mask is another 32-bit number
that looks similar to an IP address, but does something
entirely different. The five address formats are: Class A, B,
C, D, and E. Figure 7-31 shows a breakdown of the three
classes of network addresses of interest: Class A, B, C.
243
Each line in each class represents an IP address in binary
from bit zero to 32. Under Class A the first eight bits
are the network information. This identifies the network
itself and is shared by all devices on that network segment. To the right of the vertical divider line, the host
information part of the address uniquely identifies each
device. A host is any device with an assigned address.
When the classes are compared, it is seen by looking at
each class the dividing line between network and host
moves. Class B addresses are divided in the middle with
two octets for the network ID and two for the host. Class
C addresses have the first three octets for the network
and the last one for the host device. Moving the dividing
line and changing classes determines how many different networks can be created and how many hosts are on
each. When the IP address and subnet mask are compared, anywhere where there is a one indicates the network
portion of the IP address. Anywhere where there is a zero
shows the host portion. If two addresses are not on the
same subnet they will not be able to talk to each other.
Figure 7-32 shows a dissection of an IPv4 IP address with
its subnet mask.
The subnet mask is uncovered by comparing the IP
address and the subnet mask in binary. Anywhere where
a one appears in the comparison indicates the network
portion of the IP address. Anywhere where there is a zero
shows the host portion of the address.
The IP addresses are used to identify the camera equipment in a network whether local or on the Internet.
These addresses are configured by software: they are not
hardware-specific. An IP address can be either static or
dynamic. Static addresses do not change and are usually found on LAN and WAN networks. However, if the
network interfaces via dial-up modem, high-speed cable
HOW IP VERSION 4 ASSIGNS IP ADDRESSES
BREAKING DOWN 3 CLASSES OF NETWORK ADDRESSES
INTO BINARY
CLASS A
BIT
#0
7 8
NETWORK
INFORMATION
31
NETWORK BEGINNING
OCTET
CLASS
A
B
C
HOST INFORMATION
1–126
128–191
192–223
NUMBER OF
NETWORKS
HOST ADDRESSES
PER NETWORK
126
> 16, 000
> 2, 000, 000
16, 777, 2 1 4
65, 5 3 4
254
CLASS B
BIT
#0
15
NETWORK INFORMATION
BIT
#0
HOST INFORMATION
CLASS C
23
NETWORK INFORMATION
FIGURE 7-31
31
16
24
31
HOST
INFORMATION
Class A, B, C network addresses
THE GRAY LINES REPRESENT IP ADDRESSES
IN BINARY FORM FROM BIT 0 TO 32. UNDER
CLASS A, THE FIRST 8 BITS ARE TITLED NETWORK
INFORMATION. THESE BITS IDENTIFY THE NETWORK
ITSELF AND ARE SHARED BY ALL DEVICES ON THAT
NETWORK SEGMENT. AFTER THE VERTICAL DIVIDER
LINE THE HOST INFORMATIN PART UNIQUELY
IDENTIFIES EACH HARDWARE DEVICE.
244
CCTV Surveillance
DISSECTING THE IP ADDRESS AND SUBNET MASK
DECIMAL NOTATION
BINARY NOTATION
IP ADDRESS
154.140.76.45
10011010
10001100
01001100
00101101
SUBNET MASK
255.255.255.0
11111111
11111111
11111111
00000000
THIS OCTET IS PART
OF AN EXTENDED
NETWORK PREFIX
FIGURE 7-32
THIS OCTET REPRESENTS
HOST INFORMATION
IP address and subnet mask
modem, or DSL, the IP address is usually dynamic, which
means it changes each time the Internet connection
is made.
The dynamic host configuration protocol (DHCP) is an
IP for automating the configuration of equipment that
uses the TCP/IP protocol. It is the IP-addressing method
where the network router supplies a temporary IP address
to the computer connected to it. If a device is programmed
to use DHCP it is likely the device will function on the
LAN, but not be accessible from outside the LAN using
the Internet.
The DHCP lets network administrators automate and
centrally manage the assignment of IP addresses in
an organization’s network. The DHCP lets the network
administrator supervise and distribute IP addresses from
a central point and automatically send a new IP address
FIGURE 7-33 Converting the
decimal IP address to binary
when a computer is plugged into a different location in
the network.
The IP address consists of four groups, or quads (octet).
The groups are decimal digits separated by periods. An
example is: 153.99.12.227. In binary form the IP address
is a string of zeros and ones. Part of the IP address represents the network number or address and part represents
the local machine address, also known as the host number
or address. The most common class used by large organizations is Class B, which allows 16 bits for the network
number and 16 for the host number. Therefore, in the
example, 153 and 99 represent the network address and
12 and 227 represent the host address. The decimal and
binary equivalent IP address would be divided as shown in
Figure 7.33
<HOST ADDRESS>
<NETWORK ADDRESS>
DECIMAL FORM
153. 99. 12. 227
OCTET
I
II
III
IV
10011001. 01100011. 00001100. 11100011
BINARY FORM
TO CALCULATE THE FIRST OCTET:
1 × 27 + 0 × 26 + 0 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 0 × 21 + 1 × 20 =153
THE SECOND OCTET:
0 × 27 + 1 × 26 + 1 × 25 + 0 × 24 + 0 × 23 + 0 × 22 + 1 × 21 + 1 × 20 = 99
THE THIRD OCTET:
7
6
5
4
3
2
1
0
0 × 2 + 0 × 2 + 0 × 2 + 0 × 2 + 1 × 2 + 1 × 2 + 0 × 2 + 0 × 2 = 12
THE FOURTH OCTET:
1 × 2 + 1 × 2 + 1 × 2 + 0 × 2 + 0 × 2 + 0 × 2 + 1 × 2 + 1 × 2 = 227
Digital Transmission—Video, Communications, Control
For LAN and WAN systems a special networking
board/card must be incorporated into the user’s computers. This networking card uses TCP/IP protocol and is
capable of interconnecting all of the PCs to the system. By
adding a network interface to a camera site which serves as
a bridge between analog-based CCTV systems and a digital
network, one can view the video image over a computer
network as well as control PTZ functions.
Network computers have one IP address for the LAN,
and a second one for the LAN connected to the Internet
(WAN). The following three methods describe step-by-step
procedures to obtain the LAN address of the computer
using Windows XP.
Method 1
1. Click on START in Windows XP.
2. Open the Control Panel within the START window and
click on Network and Internet Connections.
3. Clicking on Network Connections opens a window displaying icons for Network Connections.
4. Right-click the Network Connection that is currently
“enabled” and click Properties.
5. Scroll down the center of the Properties window and
highlight Internet Protocol.
6. Click the Properties button, and a window will display
the following information: “Obtain IP Address Automatically.” If this button is lit the computer network is
using DHCP.
If the button “Use The Following IP Address” is lit,
the network is using “static” IP addresses that do not
change periodically. The information boxes below will
have values such as:
• IP—192.168.1.105
• Subnet Mask—255.255.255.0
• Default Gateway—192.168.1.4.
The Subnet Mask indicates the class of network (A,
B, or C) being used. The Default Gateway is the LAN
IP address of the network router.
8. Click OK twice to close the IP address window without
changing the settings.
Method 2
Another way to access the LAN IP of a specific computer
is to:
1. Click START and RUN in Windows XP. Type Command
and press Enter.
2. Type IP Config\All in the Command window.
The same LAN IP information detailed above will be displayed on the screen.
Method 3
To obtain the WAN (Internet) IP of the network:
1. Open a Web browser such as Internet Explorer.
2. Type http://www.whatismyip.com in the address line.
3. Click Go.
245
The IP address of the network will be displayed on the
computer screen.
7.9 VIDEO SERVER, ROUTER, SWITCH
A Server is a computer or software program that provides services to clients—such as a file storage (file server),
programs (application server), printer sharing (printer
server), or modem sharing (modem server).
A Router is a device that moves data between different digital network segments and can look into a packet
header to determine the best path for the packet to travel.
Routers can connect network segments that use different
protocols and allow all users in a network to share a single
connection to the Internet or a WAN.
A Switch is a device that improves network performance
by segmenting the network and reducing competition for
bandwidth.
7.9.1 Video Server
A server is a computer or program that provides services to
other computer programs in the same or other computers.
A computer running a server program is also frequently
referred to as a server. Specific to the Web and a web server
is the computer program that serves requested HTML
pages of files.
Video servers transform analog video into high-quality
digital images for live access over an intranet or the Internet. A video server enables the user to migrate from an
existing analog CCTV system into the digital world. Most
single video servers can network up to four analog cameras, a cost-effective solution for transmitting high-quality
digital video over computer networks. By bridging the analog to digital technology gap, video servers complement
previous investments in analog cameras.
A video server digitizes analog video signals and distributes digital images directly over an IP-based computer
network, i.e. LAN, intranet, Internet. The video server converts analog cameras into network cameras and enables
users to view live images from a Web browser on any network computer, anywhere and at anytime.
The video server can deliver up to 30 fps in NTSC format
(25 fps PAL) over a standard Ethernet. It includes one or
more analog video inputs, image digitizer, image compressor, a web server, and network/phone modem interface
and serial interfaces (Figure 7-34).
The video server receives analog video input from the
analog camera which is directed to the image digitizer.
The image digitizer converts the analog video into a digital format. The digitized video is transferred to the compression chip, where the video images are compressed to
J-MPEG, MPEG-2, MPEG-4, H.264, or other format. The
CPU, the Ethernet connection and serial ports, and the
246
CCTV Surveillance
(A) BLOCK DIAGRAM
IMAGE
DIGITIZER
CENTRAL
PROCESSING
UNIT (CPU)
COMPRESSION
ENGINE
ETHERNET/IP
NETWORK
ETHERNET
DRIVER
PAN/ TILT/ ZOOM
DRIVER
FLASH
MEMORY
ANALOG VIDEO
CAMERAS
DRAM
MEMORY
MODEM
CAMERA /
PLATFORM
PHONE LINE
(B) FOUR CAMERA SERVER
FRONT
FIGURE 7-34
REAR
(a) Video server block diagram, (b) Typical equipment
alarm input and relay output represent the brain or computing functions of the video server. They handle the
communication with the network. The CPU processes the
actions of the web server and all of the software for drivers
for controlling different PTZ cameras. The serial ports
(RS-232 and RS-485) enable control of the camera’s PTZ
functions and other surveillance equipment. There is a
modem for connections to telephone or other transmission channels. The alarm input can be used to trigger the
video server to start transmitting images. The relay output
can start actions such as opening a door. The video server
is equipped with image buffers and can send pre-alarm
images of an alarm event. The flash memory is the equivalent to the hard disk of the video server and contains all
software for the operating system and all applications.
information packet. The router can be located at any
juncture of a network or gateway including each Internet
point-of-presence. The router is often included as part of
a network switch (Figure 7.35).
7.9.3 Video Switch
A switch port receives data packets and only forwards
those packets to the appropriate port for the intended
recipient. This further reduces competition for bandwidth
between the clients, servers, or workgroups connected to
each switch port.
7.10 PERSONAL COMPUTER, LAPTOP, PDA,
CELL PHONE
7.9.2 Video Router/Access Point
The video router on the Internet is a device or in some
cases software in a computer, that determines the next
network to which a packet of digital information should
be forwarded, toward its final destination. It connects at
least two networks and determines which way to send each
Personal computers (PC) and laptops are the most widely
used appliances for monitoring video surveillance images
on the digital network. The Personal digital assistant
(PDA) and cell phone are the choice when the absolute
minimum in size is required, and image quality is not the
primary factor.
Digital Transmission—Video, Communications, Control
(A)
FIGURE 7-35
(B)
Typical router/access point
7.10.1 Personal Computer, Laptop
Personal computers and laptops have the computing
capacity, digital storage, and network interfaces to monitor digital video and other surveillance functions through
wired or wireless connections. They contain the displays,
operating systems, application software, and communications devices to receive and communicate with all of
the cameras and other devices on the security network.
Laptops have the added functionality of being mobile,
transportable, and battery-operated. This is a very useful
attribute for rapid deployment video systems.
7.10.2 Personal Digital Assistant (PDA)
The full impact of video surveillance using wireless
cameras, monitors, and servers has yet to be realized.
Wireless video surveillance is rapidly growing in popularity
(A) PDA
FIGURE 7-36
247
for monitoring remote locations, whether from a laptop
or a PDA. WiFi video digital transmission provides the ability to deliver near real-time, full-motion video surveillance
at 20 fps to PDAs and cell phones at any location having
access to the Internet via the WiFi connection. A video
server at the surveillance site compresses images and sends
them wirelessly to the PDA or cell phone. The systems can
provide secure access to validated mobile phones without
any eavesdropping. The IP security cameras connected
to the network transmit digital video via MPEG-4 video
compression wirelessly to PDAs and cell phones. Remote
video and alarm surveillance is only a phone call away:
anytime of day, anywhere in the world. Software is available that allows PDA users running Microsoft Pocket PC
2002 to receive video, thereby remotely monitoring security areas while mobile. The Axis Camera Explorer (ACE)
lets you watch live network video from anywhere, on a
PDA (Figure 7-36).
(B) WIRELESS LINK
Personal digital assistant (PDA) and cellphone used as video receiver
(C) CELLPHONE
248
CCTV Surveillance
Giving personnel the ability to remotely monitor secure
areas greatly increases security functionality. Access to the
system via the Internet is accomplished by assigning an IP
address to every surveillance device entering an address in
a Web browser to connect with the system. Just about any
PDA or laptop using Windows CE or Linux with a wireless
card and the wireless Web modem can obtain a wireless
remote video transmission.
PDAs and Pocket PCs have a slot for a compact flash format WiFi radio. There are also small format WiFi radios for
PDAs and mobile data devices offering additional options
for wireless connections. A PDA is a very useful monitoring
device for a rapid deployment video system.
7.10.3 Cell Phone
The cellular phone network has a sub-carrier that can
be used to transmit and receive control data for video
cameras and other components. This sub-carrier channel
information called the cellular digital packet data (CDPD)
transmits the digital data over the cellular telephone network using the idle time between cellular voice calls. A
mobile data base station (MDBS) resides at each cellular
phone cell site that uses a scanning receiver to scan and
detect the presence of any voice traffic, based on the signal strength. Providing that there are two channels idle
(for transmitting and receiving) the MDBS will establish
an air link. The type of sub-carrier available depends on
the security service provider.
7.11 INTERNET PROTOCOL SURVEILLANCE
SYSTEMS: FEATURES, CHECKLIST,
PROS, CONS
The following is a summary of features and key questions
that should be considered in selecting a video surveillance
transmission technology. Most comments apply to both
wired and wireless networks. Some apply to wireless networks only. A list of pros and cons follows the list.
7.11.1 Features
• The IP surveillance provides worldwide remote accessibility. Any video stream, live or recorded, can be
accessed and controlled from any location in the world
over the wired or wireless network.
• Video images from any number of cameras can be stored
in digital format in a host service. This enables the viewing of images from multiple cameras and playback of
an entire sequence of events.
• The cost of developing the infrastructure for the Internet and security system services has and will be borne
primarily outside of the security industry.
• As long as there is access to the Internet, any location in
the world that has a PC and a browser can be provided
with security system services.
• The IP surveillance uses a more cost-effective infrastructure than analog technologies. Most facilities are
already wired with an UTP IT infrastructure. The installation of future directed hybrid systems will be capable
of accommodating new analog as well as digital systems
and thereby ensure compatibility.
• The IP surveillance technology provides an open, easily
integrated platform to connect and manage the enterprise data, video, voice, and control, making management more effective and cost-efficient.
• The IP digital surveillance brings intelligence to the
camera level. The VMD, event handling, sensor input,
relay output, time and date, and other built-in capabilities allow the camera to make intelligent decisions on
when to send alarms, and when and at what frame rate
to send video.
• The cost savings for commercial companies and governmental agencies implementing IP technology could
be massive. Multinational corporations and government
agencies with plants and offices around the world
already have worldwide communications networks onto
which the security function could be added.
7.11.2 Checklist
• How much bandwidth is available for network
transmission?
• How much total storage space is available to store the
video images?
• Will video be viewed and recorded remotely?
• Must the video be of high enough quality to be used for
personnel identification purposes?
• Does the application require real-time video?
• Are different frame rates needed during certain events
or specific times?
• Is a peer-to-peer network or one with a base station
(access point) required?
• How many base stations (access points or gateways) are
needed?
• How will the WiFi network be connected to the Internet?
• What are the WiFi radio options for PCs, laptop’s, PDAs,
and cell phones?
• How many users will use a single access point?
• What is the total number of users and computers?
• Will each computer use a WiFi connection?
• Is the video to be interfaced with existing networks?
• What is the available bandwidth that can be reserved for
the video signal?
• What image quality and resolution are needed for the
application?
• What resolution is needed to identify the person or
activity in the scene?
Digital Transmission—Video, Communications, Control
249
• What frame rate is needed to be activity specific and
sufficient to capture motion in the scene?
• Is a wired or wireless network more suitable? Is a wired
network preferable to minimize security problems?
• What are the security requirements: standard—
strategic?
vice. In strategic applications some form of encryption is
needed.
7.11.3 Pros
7.12 SUMMARY
There are many advantages to the implementation of IP
surveillance technologies using either wired or wireless
networks in small or large surveillance applications:
Video imaging and storage is going through more technological changes and structural redefinition than any other
part of the physical security market. The Internet and
WWW has made long-range video security monitoring a
reality for many security applications. Likewise the availability of high-speed computers, large solid state memory,
and compression technologies have made the sending of
real-time video over these networks practical and effective.
New methods of wireless transmission including MIMO
mesh have improved the range, reliability and QoS of
wireless transmission.
This chapter has described the digital video security
and Internet transmission media with its unique modulation and demodulation requirements. The specific compression algorithms required to compress the video frame
image file sizes to make them compatible with the existing wired and wireless transmission channels available are
described. A powerful technology used to transmit the
digital signal called SSM has made wireless video transmission a reality. The 802.11 spread spectrum protocols are
described as relating to video, voice, and command and
control transmission.
Security monitoring is no longer limited to local security rooms and security officers, but rather extends out to
remote sites and personnel located anywhere around the
world. Monitoring equipment includes flat panel displays,
PCs, laptops, PDAs, and cell phones. The requirement for
individual personnel to monitor multiple display monitors
has changed to a technology of incorporating smart cameras with VMDs to establish an AVS system from local and
remote sites.
A key factor to be considered in any wired or wireless digital video network system is protecting the data
from unfriendly intruders and viruses. Using WEP, VPN,
firewalls, and anti-virus and encryption techniques is
paramount.
• The IP surveillance scales from one to thousands of
cameras in increments of a single camera. There are no
16 channel jumps as in analog systems.
• Automatic transmittal of images over the Internet to a
remote location to provide video images of events that
just happened.
• Embedding the video images as HTML pages in a web
server built right into the camera.
• Transmitting video images over wireless media to PDAs,
laptop’s, and cell phones at local or remote monitoring
locations.
• Remote guard tours to provide increased efficiency of
guards and services at a greatly reduced cost.
• Intelligent monitoring and control, including the transmission of images triggered by alarm conditions with
pre-alarm images.
• Remote surveillance from anywhere to anywhere, online
any time—24/7/365.
• Wireless for convenience and cost considerations.
• Wireless a must when no wired installed network is
available.
• Can now integrate video, alarm intrusion, access control,
fire, etc. into a seamless security system.
7.11.4 Cons
Security personnel can question the security of Internetbased security systems. Section 7.7 described several
important video surveillance security concerns when using
digital IP networks. These included: viruses and hackers,
and eavesdropping. Another factor of concern is that of
reliability of the IP network, i.e. temporary loss of ser-
• Some locations do not have high-speed Internet access.
• Some Internet service providers (ISPs) may not provide
reliable service.
This page intentionally left blank
Chapter 8
Analog Monitors and Digital Displays
CONTENTS
8.1
8.2
8.3
8.4
8.5
8.6
8.7
Overview
Analog Monitor
8.2.1 Cathode Ray Tube Technology
8.2.1.1 Beam Deflection
8.2.1.2 Spot Size, Resolution
8.2.1.3 Phosphors
8.2.1.4 Interlacing and Flicker
8.2.1.5 Brightness
8.2.1.6 Audio/Video
8.2.1.7 Standards
8.2.2 Monochrome Monitor
8.2.3 Color Monitor
8.2.4 Color Model
Flat-Screen Digital Monitor
8.3.1 Digital Technology
8.3.1.1 Pixels, Resolution
8.3.2 Liquid Crystal Display (LCD)
8.3.2.1 Brightness
8.3.2.2 Liquid Crystal Display Modes of
Operation
8.3.3 Plasma
8.3.4 Organic LED (OLED)
Monitor Display Formats
8.4.1 Standard 4:3
8.4.2 High Definition 16:9
8.4.3 Split-Screen Presentation
8.4.4 Screen Size, Resolution
8.4.5 Multistandard, Multi-Sync
8.4.6 Monitor Magnification
Interfacing Analog Signal to Digital Monitor
Merging Video with PCs
Special Features
8.7.1 Interactive Touch-Screen
8.7.1.1 Infrared
8.7.1.2 Resistive
8.7.1.3 Capacitive
8.7.1.4 Projected Capacitance Technology
(PCT)
8.8
8.9
8.10
8.7.2 Anti-Glare Screen
8.7.3 Sunlight-Readable Display
Receiver/Monitor, Viewfinder, Mobile Display
Projection Display
Summary
8.1 OVERVIEW
In the late 1990s digital flat-screen devices began to
be used in video surveillance systems. Dropping prices,
improved performance, and the obvious space advantages
of flat-panel displays has caused a rapid shift away from
traditional CRT monitors for video security use. Digital
video monitors and projectors are reaching new heights
of performance and are replacing the longtime workhorse
in the industry, the CRT monitor.
This chapter analyzes the monitoring hardware used for
video security systems. This hardware consists of a variety
of monochrome and color monitors: CRTs, LCDs, plasma
screens, and organic LEDs. These monitors vary in size
from 5 to 42 inches diagonal. The monitor size depends in
part on how many cameras are to be monitored, how many
security personnel will be monitoring, and how much
room is available in the security room. The question of
how many cameras will be viewed sequentially or simultaneously on a single monitor will be analyzed. There is
discussion of the consequences of displaying 1, 2, 4, 9, 16,
and 32 individual camera pictures on a single monitor.
Special features and accessories for these analog and
digital displays will be described. These include the touch
screen that allows an operator to input a command to the
security system by touching defined locations on the monitor. Chapter 20 describes the integration of the analog
CRT and flat-panel LCD monitor displays into the security
console room.
251
252
CCTV Surveillance
There are several different hardware technologies that
exist for displaying the video image, computer data and
graphics on video monitors. The technologies include:
CRT, LCD, plasma, and OLED. The video projector is used
to display images on a large screen for multiple personnel viewing. Monitors can receive an analog video signal
and digital information in formats such as SVGA, NTSC,
PAL, and SECAM. Color CRT monitors are versatile and
often have resolutions from the standard 640 × 480 pixels
to high 2048 × 1536 pixels with a 32-bit color depth (24-bit
common), and a variety of refresh rates from 60 to 75 fps.
The sharpness of the analog display is described by the
number of TV lines it can display and that of the digital display by the number of pixels. In general the more
the pixels the sharper the picture image. Resolution and
image quality on analog and digital monitors are described
with different parameters: for the analog monitor TV lines,
for the digital monitor pixels.
The horizontal resolution of a 9-inch monochrome analog CRT monitor is approximately 800 TV lines, and for
a 9-inch color monitor approximately 500 TV lines. The
horizontal resolution of a typical 17-inch monochrome
monitor is 800 TV lines and for a 17-inch color monitor
approximately 450 TV lines. Vertical resolution is about
350 TV lines on both types as limited by the NTSC standard of 525 horizontal lines.
The horizontal and vertical resolution for a 15-inch digital LCD with a 4 × 3 format is 1024 pixels by 768 pixels
XGA (extended graphics array).
Most monitors are available for 117-volt, 60 Hz (or
230-volt, 50 Hz) AC operation, and many for 12 VDC operation with a 117 VAC to 12 VDC wall converter. Video
signal connections are made via RCA plug, BNC, 9 or 25
pin connectors. The two-position switch on the rear of
some monitors permits terminating the input video cable
in either 75 ohms or high impedance (100,000 ohms). If
only one monitor is used, the switch on the rear of the
monitor is set to the 75-ohm or low-impedance position
matching the cable impedance for best results. If multiple
monitors are used, all but the last monitor in the series are
set to the high-impedance position. The last monitor is set
to the low-impedance 75-ohm position. If a VCR or DVR
recorder or a video printer is connected, all the monitors
are set to high impedance and the recorder or printer
devices set to low impedance. Recorder and printer manufacturers set the impedance to 75 ohms at the factory.
Only one 75 ohm terminated device can be used at a time.
Most cameras and monitors have a 4 × 3 geometric display format, that is, the horizontal-to-vertical size is 4 units
by 3 units. The high definition television (HDTV) has a
16 by 9 aspect ratio.
For any application, the security director, security systems provider, and consultant must decide:
• Should each camera be displayed on an individual
monitor?
• Should several camera scenes be displayed on one
monitor?
• Should the picture from each individual camera be
switched to a single monitor via an electronic switcher
or multiplexer?
If there is high scene activity, i.e. many people passing
into or out of an area, all cameras should be displayed on
separate monitors. For installations with infrequent activity or casual surveillance, a manual, automatic, or other
switcher or split screen should be used.
Since each installation requires a different number of
monitors and has different monitoring criteria depending
on the application, each installation becomes a custom
design. The final layout and installation of a system should
be the collaboration between the security department,
management, outside consultants, and security equipment
providers (dealer, installer, system integrator).
8.2 ANALOG MONITOR
Up until the late 1990s the CRT monitor has been the technology used in virtually all security applications, including
video surveillance, access control, alarm, and fire monitoring. With the widespread use of computer displays in many
security departments and the availability of flat-panel technologies for displaying data and video images, the CRT
display is still used in most security monitoring applications. The continuing success of the CRT monitor is based
on an extremely simple concept, with a relatively simple
structure and using electronic solid-state semiconductor
circuitry for all other electronic functions in the monitor.
While the CRT still utilizes vacuum-tube technology, its
combination with semiconductor technology provides the
most cost-effective solution to displaying a video image, be
it monochrome or color. The CRT monitor has become
less expensive while improving in quality and lifetime.
These monitors cost less than flat panel digital displays
because of their simple construction and long successful
history of high-volume production.
While the CRT has enjoyed many years of use, it is likely
that plasma displays, LCDs, OLED displays, and other
new technologies will eventually make CRT-based displays
obsolete in video security applications. These new designs
are less bulky, consume less power, and are digitally based.
As of mid-2003 some LCDs became directly comparable
in price to CRTs.
8.2.1 Cathode Ray Tube Technology
The CRT, invented by Karl Braun, is the most common display device used in video surveillance, computer displays,
television sets, and oscilloscopes. The CRT was developed
using Philo Farnsworth’s work and has been used in all
Analog Monitors and Digital Displays
EXTERNAL
SYNC
253
(1) ELECTRON SOURCE
(2) ELECTRON BEAM CONTROL
(3) ELECTRON BEAM FOCUSING
EXTERNAL
INTERNAL
VERTICAL
DEFLECTION
SUPPLY
VERTICAL
SYNC
SEPARATER
VIDEO OUT
IMPEDANCE *
(3)
ELECTRON BEAM
FOCUSING COIL
SYNC
STRIPPER
VIDEO IN
VERTICAL
DRIVE
VIDEO
PROCESSING
AMPLIFIER
(1)
CATHODE
VERTICAL
DEFLECTION
COIL
ELECTRON
BEAM
HIGH
(2)
ELECTRON BEAM
CONTROL (GRID)
LOW
75 ohm
HORIZONTAL
DEFLECTION
COIL
100,000 ohm
HORIZONTAL
DEFLECTION
SUPPLY
HORIZONTAL
DRIVE
* HIGH = 100,000 ohm
LOW = 75 ohm
FIGURE 8-1
CRT
ELECTRON BEAM
FOCUSES TO
SPOT ON SCREEN
Block diagram of cathode ray tube (CRT) monitor
television sets until the late-20th century. Components of
the monochrome CRT monitor include the video amplifying and deflection circuitry, the video-processing circuits
to remove the synchronizing signals from the video signal,
and the CRT (Figure 8-1).
The CRT is composed of four basic components:
(1) heated cathode (2) electron gun, (3) glass envelope, and (4) phosphor screen. The color CRT requires
three electron guns of similar construction to display the
three primary colors, red, green, and blue (RGB). The
monochrome CRT is relatively easy to manufacture since
the screen consists of a uniform coating of a single phosphor material. The yield during manufacture of the CRTs
is high (compared to the LCD or plasma displays) since
the human eye is far less sensitive to variations in phosphor flaws than it is to the defective pixel or cell failures
in flat panel digital displays. The homogeneous and continuous phosphor layer has very high resolution since it is
continuous, as contrasted to flat-panel cells (pixels). The
resolution of a CRT is limited by the electron beam diameter and the electronic video bandwidth that determines
how fast the electron beam can turn on and off.
The lifetime of standard CRTs is legendary, especially
under adverse operating conditions with which it is used:
consider the abuse that standard consumer TVs receive
but they still continue to operate. The CCTV monitors
may be cycled on and off and adjusted over a wide range
of brightness and contrast, sometimes beyond their design
limits and still operate satisfactorily for many years.
8.2.1.1 Beam Deflection
Cathode rays are streams of high speed electrons emitted
from the heated cathode of the vacuum tube. In a CRT the
electrons are carefully directed into a beam and this beam
is deflected by a magnetic field to scan the surface at the
viewing end (anode) which is lined with a phosphorescent
material that produces the visible image on the face of the
tube for viewing. In the case of the video monitor, and
television and computer monitors, the entire front area
of the tube is scanned in a fixed pattern called a raster.
The picture is created by modulating the intensity of the
electron beam according to the scene light intensity represented in the video signal. The magnetic field is applied
to the neck of the tube via the use of an electromagnetic
coil, and the process is referred to as magnetic deflection.
The CRT bends the electron beam at extremely high
speed with exact timing and gating to produce a complex
254
CCTV Surveillance
picture. Electron beams can be deflected so quickly that
pictures on the screen can be refreshed without noticeable
flicker. The CRT’s electron beam strikes and excites the
phosphor screen, which has a high luminous efficiency.
Most CRT monitors use magnetic deflection to deflect
the electron beam in the horizontal and vertical directions to produce the scene on the monitor face. Figure 8-1
illustrates the placement of the vertical and horizontal
deflection coils at the neck of the CRT. When current
flows through the horizontal coils, a horizontal magnetic
field is produced across the neck. The amount of horizontal deflection of the electron beam depends on the
strength of the magnetic field and therefore the current
through the coil. The direction of the beam deflection
(left to right) while passing through the horizontal coil
depends on the polarity of the field. Likewise for the vertical deflection coil the electron beam is deflected up or
down depending on the strength of the magnetic field
which in turn depends on the vertical deflection current.
The energizing of both the horizontal and the vertical
coils causes the raster scan and picture on the CRT monitor. As with the scanning in a tube or solid-state video
camera, the video monitor has an aspect ratio of 4:3 with
a diagonal of 5 units. The size of the tube is measured
from one corner of the screen to the opposite corner and
referred to as the diagonal.
A disadvantage of the CRT is its relative size, particularly
its depth compared to digital displays that have a short
depth. However, if there is sufficient space behind the
monitor there is no disadvantage to the CRT monitor size.
8.2.1.2 Spot Size, Resolution
The term “image resolution” describes how much image
detail the image can display on an analog or digital monitor. Higher resolution means more image detail. In analog
monitors the resolution is generally defined in TV lines and
defined as the number of black-and-white line pairs that
are distinguishable in the horizontal and vertical direction.
There is sometimes confusion defining the horizontal resolution, since it is sometimes defined as the number of
horizontal TV lines in a width equivalent to the vertical
height and at other times it is defined as the total number
of TV lines along the horizontal axis. The correct definition is the number of TV lines for an equivalent vertical
height.
The spot size of the light beam is the diameter of the
focused electron beam, which ultimately determines the
resolution and quality of the picture. The spot size should
be as small as possible to achieve high resolution. Typical
spot sizes range from about 0.1 to 1.0 mm. The spot size is
smallest at the center of the CRT and largest at the corners
(5–10% larger). Deflection along the edges elongates the
spot and decreases resolution.
The convention to describe the image resolution in
digital raster image displays is with a set of two positive
integers where the first number is the number of pixel
columns (horizontal width) and the second is the number
of pixel rows (height). The second most popular convention is to describe the total number of pixels in the image,
which is calculated by multiplying the pixel columns by
the pixel rows.
8.2.1.3 Phosphors
Cathode ray tube phosphors glow for a time determined
by the phosphor material and must be matched to the
refresh rate. The use of the white P4 phosphor has been
widespread as the standard monochrome television monitor phosphor in the past. It is capable of achieving good
focus and small spot size. Its low cost and ready availability contribute to its continued popularity in monitors. P4
is a medium, to medium-short-persistence phosphor. The
phosphor “glow” activated by the electron beam fades away
fairly rapidly, leaving no cursor trail or temporary “ghost”
scene when the monitor is turned off. The P4 phosphor
is moderately resistant to phosphor “burn” a term used
to describe the permanent dark pattern caused by fixed
bright scenes in the video image on the CRT face. The
susceptibility to burn is somewhat proportional to persistence, with longer-persistence phosphors more liable to
burn-in.
Color tubes use three different phosphorescent materials that emit red, green, and blue light. These colors are
emitted from closely packed patterns of clusters or strips
(Sony Trinitron) as determined by a shadow mask. There
are three electron guns (one for each color). The shadow
mask ensures that the electrons from each color gun reach
only the phosphor dots of its corresponding color, with
the shadow mask absorbing electrons that would otherwise
hit the wrong phosphor.
8.2.1.4 Interlacing and Flicker
Flicker is the visible fading between image frames displayed
on any monitor. In the CRT monitor it occurs when the
CRT is driven at too low a refresh rate (frame rate) allowing the screen phosphors to lose their excitation between
sweeps of the electron gun. On computer monitors using
progressive scan (no interlace), if the vertical refresh rate
is set at 60 Hz, most monitors will produce a visible flickering effect. Refresh rates of 75–85 Hz and above result in
flicker-free viewing on progressively scanned CRTs. Above
these rates no noticeable flicker reduction is seen and
therefore higher rates are uncommon in video surveillance applications. Although it has become acceptable
to call 60 Hz non-interlaced displays flicker-free, a large
percentage of the population can see flicker at 60 Hz in
peripheral vision when common P4-type phosphors are
used. A 19-inch CRT viewed at 27 inches covers more than
the central cone vision, and therefore most people see
some flicker. While this situation is not ideal, it cannot be
Analog Monitors and Digital Displays
overcome because of the inherent 60 Hz power-line frequency. On LCDs, lower refresh rates around 75 Hz are
often acceptable.
Interlacing is one of the most common and cost-effective
methods used to achieve increased resolution at conventional 60 Hz scan rates. One critical design consideration
in interlaced operation is that a long-persistence P39 phosphor must be used; P4 phosphor is not suitable for interlaced operation (the European equivalent of P4 is W). The
glow of short- to medium-persistence P4 phosphor begins
to fade before it can be refreshed. At the standard US
non-interlaced 60 Hz refresh rate this presents no problem: the viewer’s eye retains the image long enough to
make any fading imperceptible. In an interlaced monitor,
the beam skips every other row of phosphor as it moves
down the CRT face in successive horizontal scans. Only
half the image is refreshed in a vertical sweep cycle, so
the frame-refresh rate is effectively 30 Hz (two 1/60 second
fields equal 1/30 second frame). The eye cannot retain
the image long enough to prevent pronounced flicker in
the display if a short- to medium-persistence phosphor is
used. The phosphor glow must persist long enough to
compensate for the slower refresh rate.
The “flicker threshold” of the human eye is about 50 Hz,
with a short- to medium-persistence phosphor. Monitor
manufacturers designing for European-standard 50 Hz
operation therefore pay particular attention to the phosphor used.
Interlaced scanning is the method used in most video
systems to reduce flicker, and since video scene content
consists of large white areas, no objectionable flicker is
apparent. In computer alphanumeric/graphic displays,
most display data consists of small bright or dark elements. Consequently, an annoying flicker results when
alphanumeric/graphic data are displayed using interlaced
scanning unless a longer-persistence phosphor is used.
Therefore the phosphor type used in the video monitor is
different from that used for computer terminal monitors.
255
that video and audio from the camera location are displayed on and heard from the monitor. The video input
impedance is 75 ohms and the audio input impedance is
600 ohms.
8.2.1.7 Standards
The NTSC television format uses 525 lines per frame with
about 495 horizontal lines (any number of lines between
482 and 495 may be transmitted at the discretion of the
TV station) for the picture content. To produce satisfactory horizontal picture definition—that is, a gray scale and
sufficient number of gradations from dark and light per
line—a bandwidth of at least 4.2 MHz is required.
CCTV monitors generally conform to EIA specifications
EIA-170, RS-330, RS-375, RS-420, and most often to UL
specification 1410 for signal specifications and safety. The
analog circuitry is usually capable of reproducing a minimum of ten discernible shades of gray, as described in the
RS-375 and RS-420 specification.
The outer glass on the front of the CRT allows the light
generated by the phosphors to get out of the monitor.
However, for color tubes the glass must block dangerous
X-rays generated by the impact of the high-energy electron beam. For color monitors the type of glass used is
leaded glass. Modern CRTs are safe and well within safety
limits for humans because of this and other shielding and
protective circuits designed to prevent the anode voltage
from rising to high levels and producing X-ray emission.
CRTs operate at very high voltages that can persist long
after the monitor has been switched off. The CRT monitor and especially the tube should not be tampered with
unless the technician has had proper engineering training
and appropriate precautions have been taken. Since the
CRT contains a vacuum, care should be taken to prevent
tube implosion caused by improper handling.
8.2.2 Monochrome Monitor
8.2.1.5 Brightness
The luminance (brightness) of the CRT monitor picture
is proportional to the electron beam power, while the
resolution depends on the beam diameter. Both of these
properties are determined by the electron gun. Very high
resolution monitors are available having a resolution of
3000 lines—close to the ergonomic limit of the human eye.
Present security systems do not take advantage of this high
resolution but some systems display 1000-line horizontal
resolution.
8.2.1.6 Audio/Video
As with home television cameras and receivers, some monitors are equipped with audio amplifiers and speakers so
Figure 8-1 shows the block diagram including the videoprocessing circuits to remove the synchronizing signals
from the video signal, the video amplifying and deflection
circuitry, and the CRT. In its simplest form, the analog
CRT monochrome monitor consists of:
•
•
•
•
•
•
•
input video terminating circuit
video amplifier and driver
sync stripper
vertical-deflection circuitry
horizontal-deflection circuitry
focusing electronics and
CRT: cathode electron generator, electron gun, faceplate.
The video input signal to the monitor is a negative sync
type, with the scene signal amplitude modulated as the
256
CCTV Surveillance
positive portion of the signal (see Figure 5-4), and the
synchronizing pulses the negative portion of the signal.
Via frequency-selective circuits, the horizontal and vertical synchronization pulses are separated and passed onto
the horizontal and vertical drive circuits. The sync-stripper
circuit separates the analog video signal from the horizontal and vertical synchronizing pulses. These synchronizing
pulses produce the scanning signals for the horizontal and
vertical deflection of the electron beam and are similar
to those used in the camera to produce scanning of the
image sensor. The vertical- and horizontal-deflection electronics drive the vertical and horizontal coils on the neck
of the CRT to produce a raster scan.
The CRT consists of: (1) a cathode source that emits
electrons to “paint” the picture, (2) a grid (valve) that
controls the flow of the electrons as they pass through it,
(3) the electron beam that passes a set of electrodes to
focus the beam down to a spot, and (4) a phosphor-coated
screen that produces the visible picture (Figure 8-2).
When the focused beam passes through the field of the
tube’s deflection yoke (coils), it is deflected by the yoke to
strike the appropriate spot of the tube’s phosphor screen.
By varying the voltage on the horizontal and vertical coils,
the electron beam and spot are made to move across the
CRT with the familiar raster pattern. The screen then
emits light with intensity proportional to the beam inten-
sity resulting in the video image on the monitor. The CRT
monitor accomplishes all this using relatively simple and
inexpensive components. In this way the scene received by
the camera is reconstructed at the monitor. The block diagram is for any monochrome analog monitor. The color
monitor has electronics for the three primary colors: red,
green and blue (RGB).
Analog video monitors accept the standard video baseband signal (20 Hz to 6.0 MHz) and display the image on
the CRT phosphor. The monitor circuitry is essentially
the same as a television receiver but lacks the electronic
tuner and associated RF amplifiers and demodulators to
receive the VHF or UHF broadcast, cable, and satellite signal. All monochrome and color security monitors accept
the standard 525-line NTSC input signals. The video signal
enters the monitor via a BNC connector and is terminated by one of two input impedances: 75 ohm to match
the coaxial-cable impedance or high impedance: 10,000–
100,000 ohms (Figure 8-3).
The high impedance termination does not match the
coaxial-cable impedance and is used when the monitor
will be terminated by some other equipment such as a
looping monitor, a VCR or DVR, or some other device
with a 75-ohm impedance. If two or more monitors receive
the same video signal from the same source, only one
of the monitors—the last one in line—should be set to
MONOCHROME
COLOR
(3) ELECTRON
BEAM GUN ASSEMBLY
(3) ELECTRON
BEAMS
VERTICAL
DRIVE
CRT
ELECTRON BEAM
FOCUSING COIL
VIDEO
IN
VERTICAL
DEFLECTION
COIL
ELECTRON
BEAM
CATHODE
ELECTRON BEAM
CONTROL (GRID)
CRT
HORIZONTAL
DEFLECTION
COIL
HORIZONTAL
DRIVE
FIGURE 8-2
R G
B
(3) ELECTRON BEAMS
FOCUS TO FORM ONE
COLOR PIXEL CLUSTER
ELECTRON BEAM
FOCUSES TO SINGLE
SPOT ON SCREEN
Cathode ray tube (CRT) components
Analog Monitors and Digital Displays
257
MONITOR (CRT/LCD)
INPUT IMPEDANCE:
HIGH: 10,000 TO 100,000 ohm
LOW: 75 ohm
LOW
MONITOR (CRT/LCD)
RECORDER (DVR/VCR)
PRINTER
HIGH
VIDEO
INPUT
OUTPUT
OUTPUT
INPUT
HIGH
75 ohm
10,000 TO
100,000 ohm
LOW
FIGURE 8-3
VIDEO
INPUT
75 ohm
TERMINATION
(INTERNAL)
OUTPUT
CCTV monitor terminations and connections
the 75-ohm position. If a recorder rather than a second
monitor is used, the recorder automatically terminates the
coaxial cable with a 75-ohm resistor.
As shown in the block diagram in Figure 8-3, the monitor has two BNC input connectors in parallel. When only
one monitor is used, the impedance switch is moved to the
75-ohm, low-impedance position terminating the coaxial
cable. If more than one monitor or auxiliary equipment is
used, the terminating switch is left in the high-impedance
position, opening the connection to the 75-ohm resistor
so that the final termination is determined by a second
monitor, recorder, or printer. Some monitors contain an
external synchronization input so that the monitor may
be synchronized from a central or external source.
The operator controls available on most monochrome
monitors are power on/off, contrast, brightness, horizontal hold, and vertical hold. Three other controls sometimes available via screwdriver adjust (front or rear of the
monitor) are horizontal size, vertical size, and focus.
8.2.3 Color Monitor
Until recently the major CRT technology used in color
monitors employed three electron guns (one for each primary color) arranged in a triangle called the delta-delta
system (Figure 8-4a).
A device called a shadow mask aligns each electron gun
output so that the beam falls on the proper phosphor
dot. The shadow mask is a thin steel screen in the CRT
containing fine holes that concentrate the electron beam.
This technique provides the highest resolution possible
but requires the guns to be aligned manually by a technician, as well as expensive convergent-control circuitry.
The composite video input signal in the color monitor
contains the information for the correct proportion of R,
G, B signals to produce the desired color. It also contains
the vertical and horizontal synchronization timing signals
needed to steer the three video signals to the correct color
guns. Composite video color monitors decode the signal
and provide the proper level to generate the desired output from the three electron guns.
Today the most widely used CRT color technique is the
precision-in-line (PIL) tube that eliminates most of these
difficulties (Figure 8-4b). The PIL tube uses the shadow
mask found in its predecessors, but the electron guns are
in a single line. The spacing between the holes is termed
the dot pitch or dot-trio spacing and ultimately determines
the tube’s resolution. The highest-resolution production
PIL tube has approximately a 0.31 mm pitch, and is preconverged by the manufacturer so that no adjustment is
necessary in the field. There is a slight decrease in resolution for the PIL as compared with the original deltadelta, but this is a small sacrifice considering that no field
adjustment is required. A third CRT color tube called the
Trinitron (trademark of Sony Corporation) consists of the
phosphor layer consisting of alternate RGB vertical stripes
(Figure 8-4c).
Analog CRT monitors are available in many sizes. The
5- and 9- inch diagonal sizes are suitable for side-by-side
258
CCTV Surveillance
RED(R), GREEN(G), BLUE(B)
FORMS SINGLE WHITE DOT
0.31 mm
PITCH
IN-LINE
ELECTRON GUN
R
G
R
B
G
B
R
G
B
FIGURE 8-4
Color monitor technology
mounting in the standard EIA 19 inch rack. Figure 8-5
shows a few examples of these monitors.
8.2.4 Color Model
Video monitors and computer displays use the RGB color
model utilizing the additive model in which red, green,
and blue light are combined in different proportions to
(A) 9" MONOCHROME
FIGURE 8-5
TRINITRON
(C)
PRECISION IN-LINE (PIL)
(B)
DELTA
(A)
Typical CRT monitors
create all the other colors. The idea of the RGB model
itself came from the additive light model. Primary colors
are related to biological rather than physical concepts,
and based on the physiological response of the human
eye to light. The human eye contains receptor cells called
cones which respond most to yellow, green, and blue lights
(wavelengths of 564, 534, and 420 nm, respectively). The
color red is perceived when the yellow–green receptor is
stimulated significantly more than the green receptor.
(B) 14" COLOR
Analog Monitors and Digital Displays
The RGB model is used to display colors on the CRT,
LCD, plasma, and OLED monitors. Each pixel on the
screen is represented in the video signal or computer’s
memory as an independent value for red, green, and blue.
These values are then converted into intensities and sent
to the CRT or flat-panel display. Using an appropriate
combination of red, green, and blue light intensities, the
screen can reproduce the colors between its black and
white levels. Most computer models use a total of 24 bits
of information for each pixel commonly known as bits per
pixel or bpp. This corresponds to eight bits each for red,
green, and blue giving a range of 256 possible values or
intensities for each color. With this system approximately
16.7 million discrete colors can be reproduced.
In the 24 bpp, RGB values are commonly specified using
three integers between 0 and 255, each representing red,
green, and blue intensities in that order. For example:
•
•
•
•
•
•
•
•
(0, 0, 0) is black
(255, 255, 255) is white
(255, 0, 0) is red
(0, 255, 0) is green
(0, 0, 255) is blue
(255, 255, 0) is yellow
(0, 255, 255) is cyan
(255, 0, 255) is magenta.
The colors used in Internet Web design are commonly
specified using the RGB model. They are used in the
HTML, and related languages with a limited color palette
of 216 RGB colors as defined by the Netscape Color Cube.
However, with the predominance of 24-bit displays the
use of the full 16.7 million colors is used. The RGB color
model for HTML was formally adopted as an Internet
standard in HTML 3.2.
8.3 FLAT-SCREEN DIGITAL MONITOR
There are several technologies used to manufacture flatpanel digital displays and more are in development. They
all have one feature in common: a much better depth
profile than the traditional CRT monitor display. Typical flat-panel displays are from 1/2 to 4-inches in depth
compared with 10–20 inches for CRT monitors. The most
common flat-panel displays are:
• liquid crystal display (LCD)
• plasma
• organic LED (OLED).
The LCD and plasma displays have been in commercial
production and in widespread use in the video surveillance industry for several years. The OLED monitors are
beginning deployment in small sizes, but are expected to
be introduced in large sizes tailored for the security and
computer markets. Flat-panel displays offer a small footprint and trendy modern look but have higher costs, and
259
in many cases inferior images compared with traditional
CRTs. In some applications, specifically modern portable
devices such as laptops, cell phones and PDAs, these negatives are being overcome.
8.3.1 Digital Technology
A raster graphics image, digital image, or bitmap is the display
format used by most digital video flat-screen monitors. In
general the technology represents a rectangular grid of
pixels or points of color on a computer monitor. The color
of each pixel is individually defined (RGB) and generally
consists of: colored pixels defined by three bytes, one byte
for each red, green, and blue pixel. For a monochrome
image only black-and-white pictures are required with a
single byte for each pixel. This raster presentation is distinguished from vector graphics in that vector graphics
represents an image generated through the use of geometric objects such as lines, curves, arcs, and polygons.
The bitmap on the monitor corresponds to the format
of the image on the camera, and is stored identically to
it in the video displays’ computer memory. Each pixel in
the map has a specific width and height, and the bitmap
representing the image has an overall width and height
consisting of a specific number of rows and columns of
pixels. The quality of the raster image is determined by
the total number of pixels (resolution) and the amount
of information in each pixel (often called color depth).
The standard for most high-quality displays in 2004 had
an image that stored 24 bits (3 bytes) of color information
per pixel. Such an image in a typical surveillance application is sampled at 640 × 480 pixels (total 307,200 pixels).
This image will look good as compared to an excellent image
sampled at 1280 × 1024 (1,310,720 pixels). High-quality,
high-resolution pictures such as these generally require
compression techniques to reduce the size of the image
file stored in the computer’s memory and to fit the signal into the limited bandwidth communication transmission channels available. Raster graphics cannot be scaled
to a higher resolution i.e. larger screen size without a
loss of resolution and image quality (this is in contrast
to vector graphics which can easily scale and retain their
quality and size on the larger device on which they are
displayed).
8.3.1.1 Pixels, Resolution
A pixel (contraction of picture elements) represents the
smallest resolution element made up in the monitor picture,
in a computer memory, or in the camera image sensor.
Usually the pixels (dots) are so small and so numerous
that they cannot be distinguished on the monitor and
appear to merge into a smooth image when viewed at a
normal distance from the monitor. The pixel dots in the
260
CCTV Surveillance
flat-panel display are analogous to the dots used in hardcopy printed matter used to produce a printed image. The
color and intensity of each pixel is such that it represents
the scene image at that location.
The more the pixels used to represent the image the
higher the resolution and the closer the image resembles
the original scene. The number of pixels in the image
determines the image resolution. The normal VGA display
has 640 × 480 pixels. In a monochrome image each pixel
has its own brightness between the range of zero and one
where zero represents black and one represents white.
For example an eight-bit image can display 255 brightness
levels. In a color image the number of distinct colors that
can be represented by pixels depends on the number of
bits per pixel (bpp). Some standard values are:
• 8 bpp provides 256 colors.
• 16 bpp provides 65,536 colors referred to as Highcolor.
• 24 bpp provides 16,777,216 colors referred to as Truecolor.
In both full color LCD, plasma, OLED flat panels, and
CRT monitors, each of the pixels is constructed from three
sub-pixels for the three colors, and are spaced closely
together. A unique technology is the Sony Trinitron that
has three closely spaced stripes of red, green, and blue
(Figure 8-4c). Each of the sub-pixels has an intensity determined by its color RGB component values and due to
their close proximity they create an illusion of being one
specifically tinted pixel.
A recent technique for increasing the apparent resolution of a color display is referred to as “sub-pixel font
rendering.” This technique uses knowledge of the pixel
geometry to manipulate the three colors’ sub-pixels separately and works best on LCDs. It also eliminates much of
the anti-aliasing in some scenes and is used primarily to
improve the appearance of text. Microsoft’s ClearTypeTM
that is available in Windows XP is an example using this
technology.
The display resolution of a digital video monitor or
computer display is represented by the maximum number
of pixels that can be displayed on the screen, usually given
as a product of the number of columns horizontal (X ) and
the number of lines vertical (Y ). The horizontal number
is always stated first.
Common current computer display resolutions are
listed in Table 8-1.
The 640 × 480 resolution was introduced by IBM, and
has been in use from approximately 1990 to1997 in their
PS/2 VGA multicolor onboard graphics chips. This particular format was chosen partly due to its 4:3 ratio. The
800 × 600 array has been the standard resolution from
1998 to the present, but the 1024 × 768 is fast becoming
the standard resolution since it has not only the 4:3 ratio
but a higher resolution. Many websites and multimedia
products are designed for this resolution. Windows XP
is designed to run at 800 × 600 minimum although it
is also possible to run applications with the 640 × 480
format.
With 15- and 17-inch digital monitors in use, 1024 × 768
resolution is standard. For 19-inch monitors 1280 × 1024
is the recommended standard. Good 21-inch monitors are
capable of 1600 × 1200 resolution. There are also 24-inch
wide-screen monitors that can often display 1900+ pixels
horizontally.
SCREEN SIZES *
PIXEL
FORMAT
ASPECT
RATIO
LCD
PLASMA
QVGA
320 × 240
4:3
—
—
VGA
640 × 480
4:3
15, 17, 20
—
SVGA
800 × 600
4:3
20
—
XGA
1024 × 768
4:3
15
42, 43
XGA+
1152 × 864
4:3
—
—
SXGA+
1400 × 1050
4:3
—
—
WSXGA
1680 × 1050
16:10
22
—
WUXGA
1920 × 1200
16:10
23
—
QXGA
2048 × 1536
4:3
—
—
HDTV 1080i
1920 × 1080
16:9
42, 45
42, 50
HDTV 720p
1280 × 720
16:9
17, 23, 27, 30
42, 50
COMPUTER
STANDARD
* DIAGONAL (inch)
Table 8-1
Digital Video Monitor Display Formats
Analog Monitors and Digital Displays
then passes through the second sheet. The entire assembly
looks nearly transparent with a slight darkening caused by
light losses in the original polarizing sheet.
When an electric field is applied to the panel the
molecules in the liquid align themselves with the field
and inhibit the rotation of the polarized light. Since the
light impinges on the polarizing sheet perpendicularly to
the direction of polarization, all the light is absorbed in
the cell and it appears dark. Most visible wavelengths—all
colors—are rotated by LCDs in the same way. In a color
LCD each pixel triad is divided into three sections having one red, one green, and one blue filter to project the
individual colors. All colors are achieved by varying the
relative brightness of the three sections.
8.3.2 Liquid Crystal Display (LCD)
In 1968 a group at RCA demonstrated the first operational
LCD based on the dynamic scattering mode (DSM). In
1969 a former member of the RCA group at Kent State
University discovered the twisted nematic field effect in liquid crystals and in 1971 the ILIXCO Company produced
the first LCD based on this effect. This technology has
now superseded the DSM type and is now used in most
LCD displays.
The LCD is a thin lightweight panel consisting of an
electrically controlled light polarizing liquid sealed in cells
between two transparent polarizing sheets (Figure 8-6).
The two polarizing axes of the two sheets are aligned
and perpendicular to each other, and each cell is supplied with electrical contacts that allow electric fields to be
applied to the liquid inside. In operation, when no electric field is applied, light is polarized by one sheet, rotated
through the smooth twisting of the crystal molecules, and
8.3.2.1 Brightness
The brightness of a display is defined by a unit of luminance called the “nit” and is often used to quote the
brightness of a display. Typical displays have a luminance
LIGHT
LIGHT
VERTICAL
POLARIZER
LIGHT
SOURCE
POLARIZING
FILTER
NO
LIGHT
OUT
RANDOM
POLARIZATION
HORIZONTAL
POLARIZER
TWISTED
NEMATIC
CRYSTAL
DC VOLTAGE
POLARIZING
FILTER
LCD OFF
LCD ON
(LIGHT BLOCKED BY 2nd POLARIZER)
POLARIZATION OVERVIEW
TWISTED NEMATIC (TN) OPERATION
UPPER POLARIZER
UPPER GLASS
PATTERNED
TRANSPARENT
ELECTRODES
LIQUID CRYSTAL
LOWER GLASS
LOWER POLARIZER
CONSTRUCTION OF TN DISPLAY
FIGURE 8-6
Reflective twisted nematic LCD assembly
261
PATTERNED
TRANSPARENT
ELECTRODES
262
CCTV Surveillance
of 200–300 nits. Outdoor high-brightness displays can have
luminance in values in the range of 1000–1500 nits.
8.3.2.2 Liquid Crystal Display Modes of Operation
The LCD technology lends itself to several different modes
of operation. It can be operated in a transmissive or
reflective mode. A transmissive LCD is illuminated from
the backside and viewed from the front side. Activated
cells therefore appear dark while inactive cells appear
bright. The transmissive LCD technology is used in highbrightness indoor applications and for outdoor use. In this
mode of operation a lamp assembly is used to illuminate
the LCD panel that usually consumes more power than
the LCD panel itself (Figure 8-7).
The second LCD technology is the reflective type using
ambient light reflected off the display. It has a lower
contrast than the transmission type and is not generally
useful for video security applications except for batteryoperated systems where very low power operation is
required. It finds most application where small, handheld
monochrome displays are required.
Quality control issues in manufacturing LCD panels are
different from CRT monitors. Since the digital panels contain thousands of individual pixels, a defect in the panel
is visible whenever one or more of the pixels are not operating. However, the panels still may be useful if only a
limited number of pixels are not operating or if the defective pixels are not in the central part of the display. Several
criteria are used to grade the individual LCD panels and
determine whether they are suitable for security application. One criteria used for passing or failing LCD panels
was developed by IBM to quality-check their ThinkPad
laptop computers. If the panel had less than a specified
number of defective bright dots, or was not dark, it was
passed; if more, it was rejected (Figure 8-8).
The first generation LCDs used passive matrix technology. This technology uses a simple conductive grid to
deliver current to the liquid crystals in the target area. The
second-generation active matrix display uses a grid of transistors with the ability to hold a charge for limited time,
much like a capacitor. Because of the switching action of
the transistors, only the desired pixel receives a charge,
improving the image quality over a passive matrix. The
SINGLE
PIXEL
DATA LINE
TRANSISTORS
PIXEL
ELECTRODE
TFT
GATE LINE
FULL LCD ARRAY
THIN FILM TRANSISTOR (TFT) LIQUID CRYSTAL ARRAY
LAMP POWER SUPPLY
LAMP
LAMP
LCD PANEL
COMPOSITE VIDEO IN
RGB IN
VGA IN
DC POWER IN
VIDEO
PROCESSOR
LCD PANEL
DRIVE
ELECTRONICS
LCD ASSEMBLY
FIGURE 8-7
TFT Liquid crystal array—LCD assembly
Analog Monitors and Digital Displays
263
BLEMISH CRITERION USED BY
A MAJOR MANUFACTURER FOR
A 12" DIAGONAL LCD DISPLAY
RESOLUTION
BRIGHT
SPOTS
DARK
SPOTS
TOTAL
SPOTS
QXGA
15
16
16
UXGA
11
16
16
SXGA+
11
13
16
XGA
8
8
9
SVGA
5
5
9
12" LCD PANEL WITH 6 BLACK AND 3 WHITE SPOTS
FIGURE 8-8
Pass/fail quality control criteria for LCD panels
thin-film transistors hold a charge and therefore the pixel
remains active until the next refresh occurs.
The AMLED active matrix LCD is in widespread use
in the security industry and the only choice of notebook computer manufacturers. They are used due to their
lightweight, very good image quality, wide color range, and
fast time response. The display contains an active matrix
with polarizing sheets and cells of liquid crystal, and a
matrix of thin-film transistors (TFTs). These transistors
store the electrical state of each pixel in the display while
the other pixels are being updated. This method provides
a much brighter, sharper display than a passive matrix
LCD of the same size. An important specification for these
displays is the large viewing angle that they can accommodate. These displays have refresh rates of around 75 Hz.
8.3.3 Plasma
The plasma display is an emissive flat panel where light is
created by phosphors that have been excited by a plasma
discharge between two flat panels of glass that have a gas
discharge containing no mercury (contrary to the back
lights of the AMLCD panel). The plasma display uses a
mixture of the noble gases neon and xenon. The neon
and xenon gases in the plasma display are contained in
hundreds of thousands of tiny cells sandwiched between
the two plates of glass (Figure 8-9).
The control electrodes are also sandwiched between the
glass plates on both sides of cells. Electronics external to
the panel behind the cells address each of the pixels cells.
To ionize the gas in a particular cell the plasma display’s
computer charges the electrodes that intersect at that particular cell. When intersecting electrodes are charged with
a voltage difference between them and electric current
flows through the gas in the cell, it stimulates the gas and
causes it to release ultraviolet photons. The phosphors
in the plasma display give off color light when they are
excited by these ultraviolet photons. As with other displays
every pixel is made up of three separate sub-pixels with
different colored phosphor (RGB) to produce the overall
color of the pixel by varying the pulses of current flowing
through different cells.
Major attributes of the plasma display are that it is very
bright (1000 lux or higher), has a wide range of colors,
and can be produced in large sizes of up to 80 inches
diagonally. Another advantage of the plasma monitor over
others is its high contrast ratio, often advertised as high
as 4000:1. Since contrast is generally hard to define, an
absolute value for the contrast improvement over other
technologies is not yet available.
Plasma panels also have very high contrast ratios, creating a near-perfect black image—important when there is a
need to discern picture content in LLL scenes. The display
panel itself is about one-quarter inch thick while the total
thickness including electronics can be less than 4 inches.
Plasma displays use approximately the same power as a
CRT or AMLCD monitor. Plasma monitors still cost more
than all the other digital display technologies.
A main advantage of the plasma display technology
over others is that it is very scalable and very large,
and wide screens can be produced using extremely thin
materials. Plasma displays can have as many as 1024
shades, resulting in a high-quality image. Since each pixel
is lit individually, the image is very bright and has a very
wide viewing angle. The image quality is nearly as good
as the best CRT monitors.
Figure 8-9 shows examples of some standard LCD and
plasma monitors.
264
CCTV Surveillance
FIGURE 8-9
(A) 6.4" DIAGONAL LCD IN CASE
(B) 6.4" DIAGONAL LCD UNCASED
(C) 17" DIAGONAL LCD
(D) 42" DIAGONAL PLASMA DISPLAY
Standard LCD and plasma monitors
8.3.4 Organic LED (OLED)
An organic OLED is an LED made of a semiconducting
organic polymer (Figure 8-10).
These devices promise to be much cheaper to fabricate
than the inorganic LEDs used in other applications. The
OLEDs can be fabricated in small or large arrays by using
simple screen-printing methods to create the color display.
One of the greatest benefits of the OLED display over
traditional LCDs is that they do not require a backlight to
function. They draw far less power than LCDs and can be
used with small portable battery-operated devices which
in the past have been using monochrome low resolution
LCDs to conserve power. This also means that they will
be able to operate for long periods of time with the same
amount of battery charge. The first digital camera using
an OLED display was shown by the Kodak Company in
2003. This was the first OLED technology and is usually
referred to as small molecule OLED.
The second technology and improvement over the
first was developed by Cambridge Display Technologies
and is called “light emitting polymer” (LEP). Although
a latecomer, LEP is more promising because it uses a
more straightforward production technique (Figure 8-11).
The LEP materials can be applied to the substrate by a
technique derived from commercial inkjet printing. This
means the LEP displays can be made flexible and inexpensively. Organic LEDs operate on the principle of electroluminescence. An organic dye is the key to the operation
of an OLED. To create the electro-luminescence a thin
film of the dye is used and a current passed through it in
a special way.
The radically different manufacturing process of OLEDs
lends itself to many advantages over traditional flat panel
displays. Since they can be printed onto a substrate using
Analog Monitors and Digital Displays
265
LIGHT OUTPUT
TRANSPARENT ANODE
5–10 V
ORGANIC EMITTING STACK
TRANSPARANT CATHODE
GLASS/PLASTIC SUBSTRATE
(A) OLED DISPLAY
FIGURE 8-10
(B) OLED STRUCTURE
Organic LED (OLED) panel structure
(A) LEP APPLICATION USING INKJET TECHNOLOGY
(B) OLED STRUCTURE
MOVING INKJET CARTRIDGES
R,G,B LEP POLYMERS
2–10 VDC
METAL CATHODE
ROTATING
DRUM
ELECTRON
TRANSPORT
LAYER (ETL)
ITO ANOD
HOLE
INJECTION
LAYER (HIL)
3 OLED EMITTERS
ORGANIC STACK
GLASS SUBSTRATE
R
OLED FILM CARRIER
G
B
FILM/SUBSTRATE
DIRECTION
LIGHT OUTPUT
NOTE: LEP = LIGHT EMITTING POLYMER
ITO = INDIUM–TIN–OXIDE
FIGURE 8-11
OLED flat panel LEP technology
traditional ink jet technology they have a significantly
lower cost than LCDs or plasma displays. This scalable manufacturing process enables the possibility of
much larger displays. Unlike most security LCD monitors
employing backlighting, the OLED is capable of showing
true black (completely off). In this mode the OLED element produces no light, theoretically allowing for infinite
contrast ratio. The range of colors and brightness possible with OLEDs is greater than that of LCDs or plasma
displays. Needing no backlight, OLEDs require less than
half the power of LCDs and are well suited for mobile
applications where battery operation is necessary.
8.4 MONITOR DISPLAY FORMATS
There are several video formats used in different applications but the most predominant format remains the
4:3, which is used in almost all video security surveillance
applications. A new format representing HDTV has not
266
CCTV Surveillance
yet made any real impact in the security surveillance field
(Figure 8-12).
landscape scenes: parking lots, waterfronts, airport
runways and aircraft parking area, and public gathering
places.
High definition TV has many different pixel formats
and screen sizes that provide different resolutions and
recommended viewing distances (Table 8-2).
8.4.1 Standard 4:3
The standard 4 ×3 video format (horizontal × vertical) has
been in existence for many years and remains the predominant format at this time for the CRT and LCD. Liquid
crystal displays and OLED displays are also manufactured
in both the 4 × 3 and 16 × 9 formats.
8.4.3 Split-Screen Presentation
Equipment that combines video images can produce a significant reduction in the number of monitors required
in a security console room. While the monitors for these
displays are the same as those for single-image displays,
the image-combining electronics permit displaying multiple camera scenes on one monitor. Chapter 16 describes
the hardware to accomplish this function. The hardware
takes the form of electronic combining circuits and special
8.4.2 High Definition 16:9
The 16:9 HDTV format was introduced as a new widescreen display to satisfy the consumer and presentation
markets. This format has not yet found widespread use
in the video security sector but could offer advantages in
specific applications such as viewing wide-angle outdoor
WIDE-SCREEN
16
10
FORMAT
16
16:10
9
HORIZONTAL
VERTICAL
4
3
16:9
ASPECT RATIO
16 × 10
16 × 9
4×3
4:3
ARRAY SIZE:
MEGA PIXELS
320
240
76,800
VGA
640
480
307,000
SVGA
800
600
480,000
XGA
1024
768
786,000
XGA+
1152
864
995,000
SXGA
1280
1024
1,310,000
SXGA+
1400
1050
1,470,000
UXGA
1600
1200
1,920,000
HDTV
QVGA
FIGURE 8-12
STANDARD
1280
720
921,600
HDTV
1920
1080
2,073,000
WXGA
1280
768
983,000
WSXGA†
1680
1050
1,764,000
WUXGA
1920
1200
2,304,000
Monitor display formats
Analog Monitors and Digital Displays
16:9 DIAGONAL d
SCREEN SIZE (INCH)
MINIMUM VIEWING
DISTANCE D (ft)
267
MAXIMUM VIEWING
DISTANCE D (ft)
20
2.5
5.0
26
3.3
6.5
30
3.8
7.6
34
4.3
8.5
42
5.3
10.5
47
5.9
11.8
50
6.3
12.5
55
6.9
12.8
60
7.5
15
65
8.1
16.2
THE HDTV DISPLAY ASPECT RATIO IS 16:9 OR ABOUT 1.78:1.
STANDARD ANALOG VGA VIDEO IS 4:3 OR 1.333:1
DISPLAY
OBSERVER
d
HDTV IS APPLICABLE TO SURVEILLANCE APPLICATIONS
REQUIRING WIDE HORIZONTAL FIELD OF VIEW, HIGH
RESOLUTION AND LARGE SCREEN SIZE.
D
Table 8-2
HDTV Screen Sizes vs. Viewing Distance
applications or image-combining optics to produce multiple images—from 2 to 32 images—on one monitor
screen.
8.4.4 Screen Size, Resolution
Resolution specifications for monitors refer to a full camera image presented on the monitor. When a split-screen
presentation is used the resolution for each of the camera
scenes decreases proportionally to the decrease in horizontal width and vertical height. A four-camera (quad)
presentation on a monitor decreases the horizontal resolution by two and vertical resolution by two. Likewise,
nine camera scenes on a monitor decrease the horizontal resolution by three and vertical resolution by three.
When the screens are split to display 16 and 32 images,
the horizontal and vertical resolutions decrease proportionately.
8.4.5 Multistandard, Multi-Sync
Multistandard, multisync, and multivoltage television
monitor–receiver combinations are available that operate
on the US NTSC (525 TV lines) and the European CCIR
(625 TV lines) standards. Color systems operate on the
NTSC, PAL, and SECAM formats. The multisync monitors are used primarily in computer displays where the
computer monitor signal has a scan rate different from
the 60 Hz (or 50 Hz) rate and the monitor must synchronize to that other scan rate. Multivoltage monitors
operate from 90 to 270 volts AC, 50–60 Hz for worldwide use.
8.4.6 Monitor Magnification
The overall video system magnification depends on the
lens, camera, and monitor parameters. Section 4.2.2 analyzes the magnification as a function of the camera sensor
size (1/4", 1/3", etc.), the lens focal length, and the display monitor size (screen diagonal). Table 4-5 summarizes
the magnification of the overall video system for various
monitor, lens, and sensor sizes.
8.5 INTERFACING ANALOG SIGNAL TO DIGITAL
MONITOR
Connecting analog video signals to a digital flat-panel display requires special consideration. All flat-panel products
face a common problem. Though all of these displays generate the video image using digital techniques, many of
the video sources remain firmly entrenched in the analog world. There are numerous analog sources to contend
with. Two pertaining to video security are:
1. Computer video sources with component video and separate digital synchronizing signals (R, G, B, Y, C)
2. Composite video sources including NTSC and PAL signals and the S-video.
268
CCTV Surveillance
INPUT(S):
COMPUTER OR
COMPONENT
VIDEO
CCTV VIDEO
COMPOSITE
OR S-VIDEO
DIGITAL VIDEO
SIGNAL
OUTPUT(S):
RGB,
YCbCr
DISPLAY TECHNOLOGY:
DLP, LCD, LCoS, PLASMA
CCTV VIDEO
ENCODER
VIDEO
SCALER
DISPLAY FORMAT:
FRONT/REAR PROJECTION
FLAT PANEL
DVI/HDMI
DIGITAL
DVI = DIGITAL VIDEO INTERFACE (CONNECTOR)
DVI-D (DIGITAL)
DVI-A (ANALOG)
DVI-I (INTEGRATED DIGITAL/ANALOG)
HDMI = HIGH-DEFINITION MULTIMEDIA INTERFACE
FIGURE 8-13
DISPLAY
DRIVER
LCD = LIQUID CRYSTAL DISPLAY
DLP = DIGITAL LIGHT PROCESSING (PROJECTORS)
LCoS = LIQUID CRYSTAL on SILICON (PROJECTORS)
Analog to digital interface problem-block diagram
There are also numerous digital input signals that must
be interfaced to the flat-panel display.
The analog signals need to be converted into a digital
form so they can be scaled and optimized for the performance of the targeted digital display device. A typical
system block diagram is shown in Figure 8-13.
Ideally a single integrated circuit (IC) would be able to
receive all of the different input video signals, perform
the required scaling functions, and transmit the resulting
data to the digital display subsystem. So far the challenging and often conflicting requirements of such a device
have prevented development of a cost-effective single-chip
solution to this problem. In operation the device captures
RGB computer video at resolutions from VGA to UXGA
or YCbCr component video at resolutions from 480i to
1080i, including 720p. It can support resolutions as high
as 1600 × 1275 at 75 fps.
Digital video interfacing (DVI) electronics is a form of
video connector (interface) made to maximize the display
quality of analog video on flat-panel LCD (and other) computer monitors and high-end video cards. It was developed
by an industry consortium, the Digital Display Working
Group (DDWG).
Existing EIA standards are analog as are the monitors they are connected to. However, multifunctional LCD
monitors and plasma screens internally use a digital signal.
Using VGA cabling results in the computer signal being
converted from the internal digital format to analog on
the VGA cable and then back to digital again in the monitor for display. This obviously reduces picture quality, and
a better solution is provided by DVI to simply supply the
original digital signal to the monitor directly.
The three types of DVI connections are:
1. DVI-D (digital)
2. DVI-A (analog)
3. DVI-I (integrated digital/analog).
One shortcoming of DVI is that it lacks USB passthroughs. The data format used by DVI is based on the
PanelLinkTM serial format devised by Silicon Image Inc.
A basic DVI-D link consists of four twisted pairs of wire
(R,G,B, and clock) to transmit 24 bits per pixel. DVI is the
only widespread standard that includes analog and digital
transmission options in the same connector.
8.6 MERGING VIDEO WITH PCs
Many CCTV security applications require combining the
CCTV image with computer text and/or graphics. To
accomplish this, the video and computer display signals
must be synchronized, combined, and sent to a monitor. Equipment is available to perform this integration
(Chapter 16).
As computer product technologies converge with video
products, many theories have been discussed about how
PCs and video should mesh, but the optimal blend has yet
to be found. Attempts to design a merged PC and video
Analog Monitors and Digital Displays
display are complicated by differences in the design of displays used by PC and video applications. Computer monitors have been optimized for reading 10 point–type text
and static images from a distance of 2 feet. Accordingly,
the physical display size is small and uses a low brightness
non-interlaced scan with fast screen refresh to produce a
crisp high-definition image that is easy on the eyes at short
viewing distances.
Video surveillance imagery on the other hand has been
optimized within the constraints of the relatively archaic
NTSC system. The system was designed to generate lowresolution high brightness images, using an interlaced
scan with low screen refresh rate (30 fps) for viewing moving images on a large display area at distances of 3–6 or
even 10 feet. Of particular interest are PC-graphics display
technologies that allow the monitor to replace the analog
video display without any degradation in picture quality.
To produce high fidelity pictures rivaling those of CRT
video displays, PCs must incorporate digital video processing technology that adapts the video stream to the characteristics of the PC monitor. This process must preserve the
inherent fidelity of the video source while simulating video
scan techniques on the PC display. This process requires
a combination of techniques to preserve the native resolution of the digital video stream, while simultaneously
handling the special de-interlacing and frame-rate conversion tasks necessary to produce high fidelity digital video
images.
The nature of the NTSC display format creates enormous challenges. Analog color video produces pictures at
a constant rate of 59.94 Hz. With an image size of 640 by
240 pixels, each field contains only half a full 480 line
picture vertical resolution. Each field scans only alternate
lines of the video display, with adjacent fields scanning
240 lines, each offseting from the other on the screen by
one half-line position. The first, or odd, field which scans
the display starts with a half-line scan while the second,
or even, field ends with a half-line. A full frame of NTSC
video actually contains 525 lines. Approximately 480 of
those 525 lines on the monitor are used for the video picture (the active lines), while the remainder makes up the
vertical-blanking interval. Therefore there are 262.5 lines
in each field. This sequence of field pairs—scanning every
other display line with a half-line off-set between fields—
creates an interlaced display scan. Each pair combines to
form a full 640 by 480 video picture frame.
Complicating matters, each of the fields within a frame
are separated in time by 1/60th of a second representing two discrete instances in time, 1/60th of the second
apart. To present a true video-like picture, the PC must
attempt to copy this display scanning technique as faithfully as possible. To accomplish this, the native resolution
of the digital video screen must be preserved from its
origin: typically an MPEG-2, or composite video decoder,
through to the digital to analog converters (DACs) of the
graphics device. Secondly the native video source must
269
be converted from an interlaced format to a display format that is suitable for the PC’s progressive scan display
mechanism without introducing any visible image glitches.
Finally the PC screen display refresh rate must be locked
to the field rate of the original video source to avoid display rate conversion artifacts. The conventional simplistic
approach to the interlaced to non-interlaced conversion
challenge is to capture both fields of the video frame in
local memory and read out both fields simultaneously to
the PC display. This interlacing technique ignores the fact
that the individual fields within a frame are temporally different. They occur 1/60th of a second apart and visually
represent two separate instances in time. A static object
such as a circle causes no problems with the simple store
and read de-interlacer, but if the object traverses horizontally across the screen a distorted motion occurs between
each field within the frame. In the most simplistic deinterlacing PC display, the two fields are displayed at the
same instant and feathering or inter-field motion-image
artifacts along the edges of horizontally moving objects
are easily noticed.
Another major problem that must be addressed is the
frame-rate conversion. The field rate of the analog video
source is fixed at approximately 60 fields per second
whereas PC displays are typically refreshed at a rate of 70
or 75 Hz (frames per second). The PC screen refresh rate
should match that of the original video source to truly
replicate the analog video behavior. Special digital display
frame-locking techniques must be used to ensure that the
screen refresh rate precisely tracks the actual field rate
of the original digital video source. The most effective
frame-locking technique is adaptive, digital frame locking,
rather than an analog phase locked loop (PPL) locking
technique.
8.7 SPECIAL FEATURES
There are several special features that may be incorporated
into analog and digital display monitors. These include:
• Interactive touch-screens allowing the monitor operator
to interact actively with the monitor for control and
communication functions. Technologies used include
resistance, infrared, and capacitance.
• Antiglare screens overlay the monitor display to provide
higher contrast of the video image when reflections and
glare from the surrounding environment are present.
• Sunlight-readable displays using high brightness flatpanel display technology.
8.7.1 Interactive Touch-Screen
The touch-screen system is particularly useful when guard
personnel must quickly and decisively react to an activity. At present the technique is not in widespread use,
270
CCTV Surveillance
but as system complexity and knowledge of its availability increase, more security systems will incorporate these
touch screens.
There are many ways to input data into the video security system. These range from keyboards to mouse to
voice etc. One method that is becoming more popular
is going directly to the source, using touch-screen technology. Allowing the user to input information directly
eliminates the need for a mouse or other pointing device,
thereby simplifying the input process.
Many monitors in security applications display the
outputs from video graphics and/or an alphanumeric
database generated by a computer. Some advanced systems operate with computer software and hardware that
permit interacting with the screen display by touching the
screen at specific locations and causing specific actions to
occur. The devices are called touch-screen templates and
are located at the front of the monitor. The touch screen
permits the operator to activate a program or hardware
change by touching a specific location on the screen.
Touch-screen interaction between the guard and the
hardware and video system has obvious advantages. It frees
the guard from a keyboard and provides a faster input
command response. Also, the guard does not have to
memorize keyboard commands and type the correct keys.
There is also less chance for error with the touch-screen
input, since the guard can point to a particular word, symbol, or location on the screen with better accuracy and
reliability. Different types of touch screens are available,
using different principles of operation.
8.7.1.1 Infrared
Infrared touch screens rely on the interruption of an IR
light grid in front of the display screen. This technique
uses a row of LEDs and photo-transistor detectors each
mounted on two opposite sides to create an invisible grid
of IR light in front of the monitor. When the IR beam
is interrupted by a finger or other stylus, causing one
or more of the photo-transistors to detect the absence
of light and transmit a signal with the X,Y coordinates,
a signal is returned to the computer electronics to perform a predetermined action. The space within the frame
attached to the front of the monitor forms the touch-active
area, and a microprocessor calculates where the person
has touched the screen. Figure 8-14a shows such a touch
screen installed on a monitor. Since there is no film or
plastic material placed in front of the monitor, there is
no change or reduction in optical clarity of the displayed
picture.
1 COMMAND (SINGLE CELL)
1 COMMAND (FOUR CELLS)
RETRO REFLECTORS
TYPICAL PROBE POINT
TYPICAL LIGHT PATH
CCD
DETECTOR
LED LIGHT
SOURCE
SENSING
MODULE
MIRROR
(A) LED—REFLECTORS—PHOTODETECTOR ARRAY
FIGURE 8-14
Monitor touch screens
(B) CONDUCTIVE POLYESTER
Analog Monitors and Digital Displays
The IR technology has no limitations in terms of objects
that can be used to touch the screen. The one disadvantage
is that the screen may react before it is physically touched.
8.7.1.2 Resistive
A second type of touch-screen technology is resistive,
which is in common use and inexpensive compared to
other methods. One shortcoming of resistive touch-screen
is that the indium tin oxide coating typically employed
is relatively fragile. The resistive touch panel consists of
a transparent, conductive polyester sheet over a rigid
acrylic back-plane; both are affixed to the front of the display to form a transparent switch matrix (Figure 8-14b).
The switch matrix assembly has 120 separate switch locations that can be labeled with words or symbols on the
underlying display, or a scene can be divided into 120
separate locations and interacted with by the operator.
Individual touch cells may be grouped together to form
larger touch keys via programming commands in the software. Typical light transmission for the resistive touchscreen is 65–75% so that not all the light pass through
the screen to the operator and the picture has lower
contrast.
Resistive touch screens combine a flexible top layer overlay with a rigid resistive bottom layer that is separated from
the top layer by insulated spacer dots. Pressing the flexible
top layer creates a contact with the resistive bottom layer
and control electronics identify the point at which the
contact is made on the screen. This technology provides
the benefits of high resolution and the fact that any type
of pointing device can be used. One shortcoming of the
resistive touch screen is its need for an overlay and spacer
dots and therefore it suffers from reduced brightness and
optical clarity, and the surface and the flexible top layer
can be prone to surface damage, scratches, and chemicals.
If a touch screen is required in an outdoor sunlit application one has to be especially aware that relatively inexpensive analog resistive models can cut light transmission
by as much as 20% and reduce the effectiveness of the
screen brightness.
8.7.1.3 Capacitive
A third type of interactive touch-screen accessory consists of an optically clear Mylar-polyester membrane that is
curved around the monitor’s front glass screen and transparent to the user. When the conductive surface of the
Mylar is pressed against the conductive surface of the glass
by the operator, a capacitance coupling draws the current
from each of the four electrodes to the touch point. Current drawn is proportional to the distance of the contact
point from each electrode, allowing the X,Y location of the
contact point to be determined. This change in voltage is
detected by the monitor electronics, which communicate
271
with the security system to indicate that the person has
touched the screen at a particular location.
The conductive coating over the surface of the display
screen is connected to electrodes at each of the edges.
This technology offers good resolution, fast response time,
and the ability to operate with surface contamination on
the face of the monitor but it is not suitable for gloved
hands. Capacitance touch screens also suffer from some
electronic drift, meaning that periodic recalibration is
required.
8.7.1.4 Projected Capacitance Technology (PCT)
Projected capacitance technology (PCT) uses embedded
micro-fine wires within a glass laminate composite. Each
wire has a diameter of approximately one-third of the
diameter of a human hair meaning that they become
nearly invisible to the human eye when viewed against
the powered up display. When a conducting stylus such
as finger touches the glass surface of the sensor, a change
in capacitance occurs, resulting in a measurable oscillation frequency change in the wires surrounding the contact point. The integrated controller calculates this new
capacitance value and these data are transferred to host
controller. Software is used to translate the sensor contact
point to an absolute screen position. The polyurethane
layer incorporating the touch-screen sensor array is sandwiched, and protected between the glass layers and is
therefore impervious to accidental and malicious damage, day-to-day wear and tear, and severe scratching. It
is able to accept input from bare and gloved hands and
needs no additional sealing to prevent the sensor from
being affected by moisture, rain, dust, grease, or cleaning
fluids.
8.7.2 Anti-Glare Screen
A common problem associated with television monitor
viewing is the glare coming from the screen when ambient lighting located above, behind, or to the side of the
monitor reflects off the front surface of the monitor. This
glare reduces the picture contrast and produces unwanted
reflections. In well-designed security console rooms where
the designer has taken monitor glare into consideration
at the outset, glare will not significantly reduce screen
intelligibility or cause viewer fatigue.
For best results, face the monitor in the direction of
a darkened area of the room. Keep lights away from the
direction in which the monitor is pointing that are either
behind the person looking at the monitor or from the
ceiling above. If there are windows to the outside of the
building where bright sunlight may come through, point
the monitors away from the outside windows and toward
the inside walls. When this cannot be accomplished and
272
CCTV Surveillance
WITH FILTER
(A) THIN FILM ANTI-GLARE FILTER
FIGURE 8-15
NO FILTER
(B) WITH AND WITHOUT FILTER
Monitor contrast enhancement using glare reduction filters
annoying glare would produce fatigue and reduce security,
any of the various anti-glare filters available should be
applied to the front of a monitor to reduce the glare and
increase the contrast of the picture. With a well-designed
anti-glare screen and proper installation, glare and reflection levels can be reduced significantly. Figure 8-15 is an
un-retouched photograph showing the contrast enhancement (glare reduction) provided by one of these filters.
These anti-glare optical filters are manufactured from
polycarbonate or acrylic plastic materials and are suitable for indoor applications. Polycarbonate filters used
in outdoor applications can withstand a wider temperature range and are therefore more suitable. The filters
come in a range of colors, with the most common being
neutral density (gray), green, yellow, or blue. The colored filters are used on graphic and data display computer
terminals, whereas the neutral-density types are used for
monochrome and color monitors. In the case of color
displays, the lighter gray filters should be used for glare
reduction.
8.7.3 Sunlight-Readable Display
Display brightness is measured and reported in nits. One
nit is roughly equal to the brightness of one standard candle. Marketing departments often like to use terminology
like ultra-bright and sun-bright, but they do not always
make the connection back into engineering units. The following levels of brightness define some of the capabilities
of these monitors:
• Bright (150–240 nits). This is the typical brightness of a
home or office computer and is suitable for most indoor
light conditions.
• High-Bright (250–340 nits). These are brighter than typical panel displays. They can be situated in brightly lit
rooms without reducing viewing capacity.
• Ultra-Bright (350–790 nits). These may be suitable for
some outdoor applications. They provide good visibility in highly illuminated environments where the light
source or bright reflections would not allow Bright or
High-Bright units to be easily read or images viewed.
• Sun-Bright (800 nits and up). Displays this bright are
suitable for outdoor sunlit applications and can be read
in direct sunlight.
Recent advances in LCD monitors have created new outdoor applications for sunlight viewable displays, especially
where CRTs are out of the question because of their bulky
size. Previously a display that provided 400 nits with a good
reflective surface was acceptable. Now the demand has
increased to 1500 nits with a minimum display size of about
15 inch diagonal. A 15 inch display can consume as much
as 50 W for a brightness of 1500 nits. Displays larger than
10 inch diagonal pose thermal problems when used in
a sun-loading environment. Manufacturers have typically
used a brute force method, increasing back-light power
to increase display brightness. This high power backlight in turn generates tremendous heat that causes an
AMLCD to go above its safe operating temperature. Heat
is detrimental to AMLCD survivability. To overcome the
heating problem massive heat sinks have been employed
in addition to expensive antireflective and IR face glass
Analog Monitors and Digital Displays
laminated onto the display to block the heat generated by
sunlight.
The display parameters required for sunlight readability
include:
• Brightness. The display must be bright enough to be legible under full sunlight. The required brightness ranges
between 400 and 1500 nits.
• Readability. The display image must be discernible by
the naked eye under all viewing conditions from full
sunlight to nighttime.
In addition to high brightness monitors, antireflective
coatings are often used to minimize reflections caused by
sunlight and an infrared coating is used to reject heat
caused by sun loading. The rejected heat spectrum covers
the near IR spectral region.
8.8 RECEIVER/MONITOR, VIEWFINDER, MOBILE
DISPLAY
Various manufacturers produce small lightweight CRT
and LCD television monitors or receiver/monitors to
accept base-band video signal inputs and/or VHF/UHF
commercial RF channels and are powered by 6, 9, or 12
volts DC (Figure 8-16).
These television monitors are particularly useful in
portable and mobile surveillance, law enforcement, and
for servicing applications. Often a portable surveillance
camera will be transmitting the video signal (perhaps also
audio) via an RF or UHF video transmitter operating on
one of the commercial channels or at 900 MHz or 2.4 GHz.
The small receiver-monitors with 1.5- to 5-inch-diagonal
CRT or LCD displays can receive and display the transmitted video signal and have an output to provide the
(A) HIGH RESOULUTION LCD MONITOR
FIGURE 8-16
Small flat screen receiver/monitor
273
base-band video signal for a VCR, DVR, or video printer
at the receiver site. These devices usually have medium
resolution (250–400 lines) often sufficient to provide useful security information and for camera installation and
testing.
Shock, vibration, and dirt are probably the most
common causes of failure for flat-panel displays in harsh
environments or used in mobile applications. The typical 15-inch TFT LCD monitor weighs about 13 lbs and
so an acceleration of 2 g makes it weigh 26 lbs. Standard office-style LCDs cannot stand up to the kind
of shock and vibration found that most mobile environments and to some degree industrial hardening is
required. At a minimum they should be shock rated at
1.5 g and have a vibration rating of 1 g. More severe
environments call for higher specifications. Aside from
shock and vibration, monitors will be subject high levels of grit, grime, water, dust, and oil than would be
expected in a normal office environment. This is where
the industrial, environmental and structural NEMA ratings
are helpful. A few key NEMA ratings for flat-panel displays used in mobile and harsh environments include the
following:
• NEMA 3 enclosures are suitable for outdoor applications, and repel falling dirt, rain, sleet, snow, and
windblown dust: enclosure contents are undamaged by
external ice formation.
• NEMA 4 adds protection from splashing and hosedirected water to NEMA 3 standard.
• NEMA 6 is similar to the NEMA 4 but the enclosure
prevents ingress of water during occasional temporary
submersion to a limited depth.
• NEMA 6P protects against water ingress after a prolonged submersion.
(B) TUNABLE 900 MHz TO 2.4 GHz
2.5" LCD RECEIVER/MONITOR
274
CCTV Surveillance
8.9 PROJECTION DISPLAY
The digital projector is an electro-optical device that converts a video image or computer graphics and data into a
bright image that is projected and imaged onto a distant
wall or screen using a lens or lens-mirror system.
The projector serves the following purposes:
• Visualization of video and stored computer data for
monitoring or presentation.
• Replaces the whiteboard and written documents.
• Provides the ability to view video images and other data
by many personnel at the same time.
• Provides the ability to playback images from a VCR,
DVR, and digital video disk onto a large screen.
Digital projection technologies include:
• High-intensity CRT
• LCD projectors using LCD light gates
• Texas Instruments DLP technology.
The current dominant technology at the high-end for
portable digital projectors is the Texas Instruments DLP
technology with LCD projectors dominating the lowend. Digital projectors take the form of a small tabletop portable projector using an external screen or a rear
projection screen forming a single unified display device.
The typical resolution for the portable projector is the
SVGA standard (800 × 600 pixels), with more expensive
devices supporting the XVGA (1024 × 768 pixels) format.
Projectors costs are determined in most part by their resolution and brightness. Higher requirements cost more. For
large conference rooms the brightness should be between
1000 and 4000 lumens. CRT devices are only suitable for
fixed installations because of their weight.
8.10 SUMMARY
There are several monitor types that can be used in the
security room. These include the standard analog CRT
monitor and the digital flat-screen LCD, plasma display,
and to a lesser extent the new organic light emitting diode
(OLED) displays. Where space permits, the standard CRT
monitor is a cost-effective solution and can provide a
bright, high-resolution monochrome or color image. In
more confined spaces or in any completely new installation the flat-panel display should be considered.
The quality of the image displayed is a function of
the number of TV lines the analog monitor can display,
and the number of horizontal and vertical pixel elements
available in the digital flat screen monitor. The standard
video format is 4 units wide by 3 units high and is almost
exclusively the format used in the video surveillance industry. A new format designed for high definition consumer
television monitors has a 16 by 9 format but has limited
use in the security industry. Video monitors are available
in multistandard, multisync configurations for use with all
available voltages and scan rates.
The use of digital flat screen monitors is increasing
rapidly, and interfacing digital computer systems with the
monitor has brought about problems which need solving. These include interfacing the analog signal provided
by most cameras to the new digital monitors. The use of
analog cameras and digital Internet cameras in the same
surveillance system further complicates the interface of
these cameras to the digital displays.
Video monitors are available having special features
such as interactive touch screens, anti-glare screens, and
sunlight-readable displays. In applications where the guard
must make quick, accurate decisions and many cameras
and security functions are involved, the touch screen can
serve a very important function for making the guard
more effective. There are several technologies available
for touch screen monitors including: infrared, resistive,
capacitive and projected capacitive technology (PCT).
Under difficult indoor and outdoor lighting situations
in which reflections are prevalent on the monitor screen,
anti-glare filters are available to reduce or eliminate this
problem. Under extreme sunlight conditions, new flatpanel displays are available with sunlight-readable displays
having very high brightness and high contrast.
There are many new applications in which the use
of automated video surveillance from remote sites or
mobile monitor stations is required. Using the digital
networks LAN, WAN, Wireless LAN (WLAN, WiFi) and
Internet cameras, laptop computers, PDAs, and other
portable devices using flat-screen display technology are
now available.
When video monitoring systems must be portable
or transportable or set up rapidly, small mobile, low
power flat-panel displays are available with monochrome
or color displays. These monitors have low electrical power requirements so that they can operate for
days using small rechargeable batteries. Using battery
power, these displays are suitable for rapid deployment
video surveillance systems and testing and maintenance
applications.
When the video scenes must be viewed by many personnel and a large-screen video image is required, video projectors are available for fixed installation or portable use.
Chapter 9
Analog, Digital Video Recorders
CONTENTS
9.1
9.2
9.3
Overview
9.1.1 Analog Video Cassette Recorder (VCR)
9.1.2 Digital Video Recorder (DVR)
9.1.2.1 DVR in a Box
9.1.2.2 Basic DVR
9.1.2.3 Multiplex DVR
9.1.2.4 Multi-channel DVR
9.1.2.5 Network Video Recorder (NVR)
Analog Video Recorder
9.2.1 Video Cassette Recorder
9.2.2 VCR Formats
9.2.2.1 VHS, VHS-C, S-VHS
9.2.2.2 8 mm, Hi-8 Sony
9.2.2.3 Magnetic Tape Types
9.2.3 Time-Lapse(TL) VCR
9.2.4 VCR Options
9.2.4.1 Camera Switching/Selecting
9.2.4.2 RS-232 Communications
9.2.4.3 Scrambling
9.2.4.4 On-Screen Annotating and Editing
Digital Video Recorder (DVR)
9.3.1 DVR Technology
9.3.1.1 Digital Hardware Advances
9.3.1.1.1 Hard Disk Drive storage
9.3.1.1.2 Video Motion Detection
(VMD)
9.3.1.1.3 Optical-Disk Image
Storage
9.3.1.1.4 Non-Erasable,
Write-Once Read-Many
(WORM) Disk
9.3.1.1.5 Erasable Optical Disk
9.3.1.1.6 Digital Audio Tape
(DAT)
9.3.1.2 Digital Storage Software Advances
9.3.1.3 Transmission Advances
9.3.1.4 Communication Control
9.3.2
9.3.3
9.3.4
9.3.5
9.3.6
9.3.7
9.3.8
9.3.9
DVR Generic Types
9.3.2.1 DVR in a Box
9.3.2.2 DVR Basic Plug-and-Play VCR
Replacement
9.3.2.3 Multiplex
9.3.2.4 Multi-Channel
9.3.2.4.1 Redundant Array of
Independent Disks
(RAID)
9.3.2.5 Network Video Recorder (NVR)
9.3.2.6 Hybrid NVR/DVR System
DVR Operating Systems (OS)
9.3.3.1 Windows 9X, NT, and 2000
Operating Systems
9.3.3.2 UNIX
Mobile DVR
Digital Compression, Encryption
9.3.5.1 JPEG
9.3.5.2 MPEG-X
9.3.5.3 Wavelet
9.3.5.4 SMICT
Image Quality
9.3.6.1 Resolution
9.3.6.2 Frame Rate
9.3.6.3 Bandwidth
Display Format—CIF
Network/DVR Security
9.3.8.1 Authentication
9.3.8.2 Watermark
9.3.8.3 Virtual Private Network (VPN)
9.3.8.3.1 Trusted VPNs
9.3.8.3.2 Secure VPNs
9.3.8.3.3 Hybrid VPNs
9.3.8.4 Windows Operating System
VCR/DVR Hardware/Software Protection
9.3.9.1 Uninterrupted Power Supply (UPS)
9.3.9.2 Grounding
275
276
CCTV Surveillance
9.3.9.3
9.4
9.5
9.6
Analog/Digital Hardware
Precautions
9.3.9.4 Maintenance
Video Recorder Comparison: Pros, Cons
9.4.1 VCR Pros and Cons
9.4.2 DVR Pros and Cons
Checklist and Guidelines
9.5.1 Checklist
9.5.2 Guidelines
Summary
9.1 OVERVIEW
9.1.1 Analog Video Cassette Recorder (VCR)
Prior to 1970s real-time video recording systems used magnetic reel-to-reel tape media and required manually changing of the magnetic tape reels. Operation was cumbersome
and unreliable, and the tape was prone to damage or
accidental erasing. The recorder required the operator to
manually thread the tape from the tape reel through the
recorder onto an empty take-up reel, similar to threading
8 mm and 16 mm film projectors. Not very much video
security recording was done in this era.
In the 1970s the first analog VCR was introduced to
the security industry. The arrival of the VCR permitted
easy loading and unloading of the tape cassette without
the user contacting the tape. The VCR machines provided
real-time 30 fps recording. VCRs found widespread use in
security when the VHS tape format became the dominant
consumer VCR format. However, most security applications require 24/7/365 day operation without stopping.
The consumer VCRs were not designed for continuous
use and did not operate in a TL mode, and were not a reliable choice for security applications. The security industry
had the manufacturers design industrial versions that have
served the security industry for many years. These TL VCRs
were especially designed to withstand the additional burden of long-term continuous recording and the start–stop
of TL recording.
The VCR was the only viable technology until the late
1990s, and still is a convenient method for recording security video images. During this period specialized functions
were added that further enhanced their usefulness which
included: (1) alarm activation (2) T-160 24 -hour real-time
recording, and 40 day extended TL recording (960 hours)
on a single cassette.
The recorded video images are used for general surveillance of premises, to apprehend and prosecute thieves and
offenders, and to train and correct personnel procedures
of security personnel. A played-back camera image from a
high-quality recording system can be as valuable as a live
observation. The video image on a monitor however is
fleeting, but the recorded image can be played back over
and over again and a hard copy printed for later use. The
original VHS tape can easily be given to the police or other
law enforcement agency for investigation or prosecution.
The TL VCR records single pictures at closely spaced
time intervals longer than the real-time 1/30-second frame
time. This means that the TL mode conserves tape, while
permitting the real-time recording of significant security
events. When a security event of significance occurs, a
VMD or alarm input signal causes the VCR to switch from
TL to real-time recording. These real-time events are the
ones the security guard would consider important and
normally view and act on at a command monitor. TL
recording permits the efficient use of a recorder so that
changing of VCR tapes is minimized.
Video cassette recorders have some shortcomings however. They can generate grainy and poor quality images
during playback because videotapes are frequently reused,
and record/playback heads are worn or misaligned,
degrading the quality of the video image. VCR tapes are
changed manually which leaves room for human error,
whereby the tape may not be changed at the required time
or a recorded tape is inserted into the machine, erasing
previously recorded images. Clean space for storing the
videotapes can also become a problem particularly true in
high-usage casino applications.
The DVR eliminates all these problems.
9.1.2 Digital Video Recorder (DVR)
The movement from tape-based real-time and TL video
recording to today’s DVRs has been a vast improvement,
and a quantum jump forward in technology. There is a new
generation of video cameras, with digital signal processing
and IP cameras and other digital devices to interface to
the DVRs, CRTs and flat screen LCD and plasma digital
monitors. Today’s DVRs do not represent the end of the
technological advancement, but are rather the beginning
of intelligent recording devices that are more user-friendly
and economical.
The DVR was first introduced to the video industry in
the early 1990s and it recorded the video image on a HD
drive in digital form. Why the change to DVR technology? The VCR has been in use for years but has been a
weak link in the overall video security system. One important reason is the VCR’s requirement for excessive tape
maintenance, deterioration of the tapes over time and
use, inability to reproduce the high resolution and image
quality produced by the digital cameras, and the excessive
manpower required to review the tapes. Another disadvantage of the VCR technology is the sheer volume of
storage space required to archive the tapes in mid- to
large-size systems at casinos, etc. This prompted many a
dealers, systems integrators and end users to switch over
to DVR equipment.
Analog, Digital Video Recorders
Another advantage is the significantly better image quality on playback. One reason for this is that the analog
VCR only records a single field of the video image. The
DVR records a full frame of information. The VCR reduces
the detail of the video image during playback by one half,
thereby rendering a poorer image than the original live
image. DVRs, on the other hand, record the full video
image on the HD and do not introduce picture noise,
and provide high stability and higher quality video image.
The DVR also eliminates the need for head and tape
replacement thereby significantly reducing maintenance
over its lifetime.
The DVR converts the incoming video camera signal
into a recorded magnetic form on a magnetic HD. The
recorder later reconstructs the video signal into a form
suitable for display on a video monitor or to be printed
on a hard-copy video printer or for transmission to a
remote site.
Most DVRs have more than one internal HD drive.
The software controlling them automatically moves the
recorded images internally, so that if there is a failure,
only a portion of the data is lost. The average 80 GByte
HD drive can store approximately 100 hours of data, while
the VCR can store just 8 hours of data. Images on HD
drives do not degrade and can be retrieved, copied, and
reused hundreds of times without compromising the picture quality. Important security images can also be stored
permanently (archived) on HD drives, recorded on digital audio tape (DAT) recorders, or burned into DVDs for
future use.
Digital video recorders provide higher quality images
than VCRs particularly during picture pause in which the
DVR exhibits no distortion and no picture tearing during
pause, single frame advance, or rewind and fast-forward
modes. This is true from the first viewing, after many viewings of the same image, regardless of the number of times
the digital image is viewed or copied. Switching from VCR
to DVR machines eliminates all tape hassles: tapes will not
have to be changed, no prying the tape out of the VCR
slot, and no cleaning or replacing tape heads again.
Unlike the VCR, the DVR permits programming the
picture resolution as required by the application. It can be
programmed locally or remotely to permit recording in
real-time or TL modes depending on the programming,
and can respond to video motion alerts or alarm inputs.
The video image quality remains the same regardless of
how many times the images are stored or re-recorded.
The DVR hardware is available in several configurations:
DVR in a box, DVR Basic (plug-and-play), DVR multiplex, and DVR multi-channel. Large systems networking
to remote sites use network video recorders (NVR).
9.1.2.1 DVR in a Box
The DVR in a box is created by adding a printed circuit
(PC) card to a PC computer and converting it into a DVR.
277
This solution might be expedient but it does have some
limitations: it has very few user-friendly features. It is used
in low-cost, small video systems.
9.1.2.2 Basic DVR
The single channel DVR is the basic replacement for the
real-time or TL VCR. The single channel DVR looks, feels,
and has the controls similar to a VCR, and can be set
by the operator from a local or remote site. It functions
like a VCR but has many advantages over the VCR. DVRs
produce sharp images over long periods of time and after
many copies have been made. The DVR is especially well
suited to perform video motion detection and can be activated by external alarms. A primary market for the basic
DVR is as a VCR replacement in a legacy analog installation. In this application the video cabling infrastructure
is already in place for transmission of video images from
the cameras to the recorder, and the most cost-effective
solution is the basic DVR. A single channel DVR is a costeffective replacement for a VCR and provides long-term
maintenance-free recordings that can be viewed locally,
remotely, and transmitted anywhere over wired or wireless networks. Making the switch from the traditional VCR
to the DVR is now here at an affordable price and can
replace the VCR with no changes to the rest of the system.
9.1.2.3 Multiplex DVR
Midsize DVRs using multiplexer technology provide highquality recording capabilities for 4–16 cameras. For midsize surveillance systems, DVRs with built-in multiplexers
operate like traditional multiplexers connected to VCRs.
The equipment offers on-screen menus and require only
simple keystrokes to find images or events by alarm input,
time, date, camera number, or other identifiers. These
recorded images can be stored for days, weeks, or months.
DVR multiplexers are available with Ethernet connections
to provide high-quality remote transmission to other sites
on the network. This makes the video images available at
local monitoring sites or at a central monitoring station,
thereby providing instant access to critical recordings. The
images can be retrieved using standard IP addresses and
PCs to make the remote monitoring and recordings available to authorized personnel anywhere on the network.
Multiplexed systems are used in midsize systems to
record multiple cameras. While recording multiple cameras, the multiplexed DVR cannot record all cameras
in real-time but rather time-share and record the cameras in sequence. DVR time-sharing operates in the
same way as a standalone video multiplexer. Multiplexed
recorders usually have a maximum storage capacity of
480–600 GByte representing 600–750 hours of real-time
recording (depends on resolution). The combination of
multiplexer and DVR has the advantage that the interface
278
CCTV Surveillance
between the multiplexer and VCR is already accomplished
by the manufacturer.
9.1.2.4 Multi-channel DVR
The multi-channel DVR technology requires significantly
more HD drives and error correcting codes to reliably
store and manipulate the images. The result is a system
where all cameras are recorded at all times without any loss
of video image information. Large enterprise systems using
multi-channel machines can record images from hundreds
of cameras for months or more, storing high resolution
video image scenes in real-time, near real-time or TL. To
store a large number of images the video image files are
compressed by removing redundant data in the image
file. The number of bytes in each image file is reduced
significantly so that the files can be stored efficiently on
the HD.
The multi-channel DVR records each camera on individual channels (not multiplexed) and is designed for
applications with a large number of cameras as in an
Enterprise system. This system can be expanded to nearly
an unlimited number of video channels by adding additional HD memory and appropriate software control. The
multi-channel DVR permits storing 60 images per second
per camera whereas the multiplexed unit divides the 60
images by the number of cameras. The multiplexed scanning often causes a jerky motion of the image during
playback and is especially noticeable when the image per
second (IPS) for the cameras falls below 15 IPS.
Most DVRs are triplex rather than duplex in design.
This means that they can: (1) display and record live video,
(2) display the recorded video locally, and (3) display the
recorded video remotely all simultaneously. In the case of
remote viewing, a standard Web browser connected to
the Internet via an ISP is all that is needed to view the
video images.
Digital video recorders allow fast searching of recorded
information based on time, date, video motion detection,
alarm input, or other external events. This permits fast
retrieval of video images and avoids wading through countless frames of video information as required with standard
VCR technology. The operator can view the information
of interest in a matter of seconds. This is a primary advantage of DVR technology over VCR.
9.1.2.5 Network Video Recorder (NVR)
The NVR records video and audio data streams received
over Ethernet networks using the TCP/IP protocol. The
NVR receives compressed video data streams from the
transmission channel and transfers the streams to an internal HD for storage in the DVR.
The NVR technique uses the Ethernet networks already
in place in most buildings, and features such as motion
detection, scene analysis, and alarm notification are
employed. These features have added to the growing popularity of network surveillance. All digital video sources or
analog cameras connected to video servers feed the digital
data streams into the network. A computer with sufficient
storage capacity serves as the DVR. The DVR accesses the
video data streams of the remote network cameras and
video servers and stores them on the HD.
9.2 ANALOG VIDEO RECORDER
9.2.1 Video Cassette Recorder
The innovation of video cassettes and the VCR resulted
in wide acceptance of this recording medium for over
25 years. VCRs used the Victor Home System (VHS) video
cassette as the recording medium. The newer Sony 8 mm
format gained some popularity in the security market
because of its small compact size while the VHS-C format
found limited use. Present real-time VCR systems record
2, 4, or 6 hours of real-time monochrome or color video
with about 300 lines of resolution on one VHS or 8 mm cassette. TL recorders can have total elapsed recording times
of up to 960 hours. Most TL recorders have alarm input
contacts that switch the recorder to real-time recording
when an alarm condition occurs.
The VCR has always been the weakest link in the video
security system with respect to image quality and reliability. Both the VHS and 8 mm recorders fall short in that
they do not record high resolution camera images. The
main reason for this is that both recorders record a field
rather than a frame, thereby losing half the camera resolution. The VHS and 8 mm formats called S-VHS and
Hi-8 increased resolution and picture quality but do not
meet the resolution capabilities of most monochrome and
color cameras. TL recording makes maximum use of the
space available on the video cassette by recording individual images at a slow pre-selected rate, thereby slowing the
recording rate. Instead of recording at a normal 30 fps
the TL VCR records one picture every fraction of a second or every few seconds. Prior to the use of DVRs using
computer HD drives, all security installations recorded the
video images using VCRs.
The TL tape machine may take about two to three minutes to search from the beginning to the end of the tape
since the tape in the TL recorder is advanced and reversed
linearly by mechanical devices and motors.
9.2.2 VCR Formats
Almost all security VCRs use the 1/2-inch wide magnetic
tape format. A compact cassette called VHS-C and sold
by JVC Company found little use in the security industry.
In the 1990s Sony developed the compact tape formats
using an 8 mm (1/4-inch)-wide tape cartridge. While most
Analog, Digital Video Recorders
security video recorders use the standard VHS cassette
format, many portable systems use the more compact Sony
8 mm and Hi-8 cassette.
279
VHS electronics and encoding scheme. The VHS-C cartridge is played back on a standard VHS machine with a
VHS-C-to-VHS cartridge adapter.
9.2.2.2 8 mm, Hi-8 Sony
9.2.2.1 VHS, VHS-C, S-VHS
Standard real-time continuous recording times for VHS
tapes are 2, 4, and 6 hours. When these cassettes are used
in TL mode, where a single image or a selected sequence
of images are recorded, 8, 24, 40, and up to 960 hours can
be recorded on a single 2-hour VHS cassette. VCRs record
the video scene on magnetic tape using the same laws of
physics as used in audiotape recorders (Figure 9-1).
The challenging aspect of recording a video picture on
a magnetic tape is that the standard US NTSC video signal
has a wide bandwidth and includes frequencies above
4 MHz (4 million cycles per second) and down to 30 Hz, as
compared with an audio signal with frequencies between
20 and 20,000 Hz. To record the high video frequencies
the tape must slide over the recording head at a speed
of approximately 6 meters per second or faster. All VCRs
have a helical-scan design, in which the magnetic tape
wraps around the revolving drum about half a turn and
is pulled slowly around a rapidly rotating drum having
magnetic record, playback, and erase heads (Figure 9-2).
This design reduces the cassette tape speed an order of
magnitude (one-tenth) slower than the linear recording
head tape speed. The audio is recorded conventionally
along one edge of the tape as a single (monaural) or a dual
(stereo) channel. Along the other tape edge is the control
track, normally a 30 Hz square-wave signal (NTSC system)
that synchronizes the VCR to the monitor during playback.
Some VCR machines have a full-track erase head on the
drum to erase any prerecorded material on the tape.
The VHS-C tape format makes use of a small tape cartridge slightly larger than the Sony 8 mm and uses the
Sony developed the smaller 8 mm and Hi-8 format video
technology (Figure 9-3).
The format and cassette are significantly smaller than
the VHS but maintain image quality and system capability similar to that of the larger format. Cassette running
times are 1/2 hour, 1 hour, and 2 hours. The 8 mm configuration is particularly suitable for covert applications
requiring a small, light-weight recorder.
The resolution obtained with standard color VHS and
8 mm VCRs, whether operating in real-time or TL mode,
is between 230 and 240 TV lines. This is not sufficient
for many security applications. Monochrome TL recorders
provide 350 TV-line resolution. The new color S-VHS and
Hi-8 format real-time and TL recorders increase the horizontal resolution to more than 400 TV lines, suitable for
facial identification and other security applications.
As with the standard VHS and 8 mm systems there is
no compatibility between the S-VHS and Hi-8 formats.
There is some compatibility between VHS and S-VHS,
and between 8 mm and Hi-8. Some important differences
between and features of the standard VHS and 8 mm, and
the S-VHS and Hi-8 formats are as follows:
• S-VHS and Hi-8 recordings cannot be played back on
conventional VHS or 8 mm machines.
• S-VHS and Hi-8 video cassettes require high coercivity,
fine-grain cobalt-ferric-oxide and metal tapes to record
the high-frequency, high-bandwidth signals.
• All S-VHS and Hi-8 recorders can record and playback
in standard mode. The cassettes have a special sensing
notch that automatically triggers the VCR to switch to
the correct mode.
TAPE HEAD
ROTATING
CAPSTAN
E
ON IELD
F
V
T
1
AUDIO
TRACK
1/2"
TAPE
WIDTH
TAPE
GUIDE
2
TAPE
GUIDE
HEA D 1
HEA D 2
TAPE TRAVEL
FIGURE 9-1
VHS video cassette recorder geometry and format
HEA D 2
HEA D 1
VHS TAPE FORMAT
CONTROL TRACK
280
CCTV Surveillance
HELICAL SCANWHEEL
ROTATION
INERTIA
IDLER
1/2" VHS
TAPE PATH
INERTIA
IDLER
VIDEO
HEADS
ERASE
HEAD
AUDIO AND
CONTROL HEADS
CASSETTE
CAPSTAN
SUPPLY REEL
FIGURE 9-2
TAKE UP REEL
VHS recorder technology
HELICAL TAPE PATH DESIGN
TAPE WRAP CONFIGURATION
ROTATING
DRUM
ROTATING
HEADS
TAPE TRAVEL
TAPE
TRAVEL
TAPE
GUIDES
8 mm TAPE FORMAT
HEAD A
8 mm
TAPE WIDTH
FM AUDIO AND
VIDEO SIGNAL
HEAD TRAVEL
DIRECTION
DIGITAL AUDIO
HEAD B
FIGURE 9-3
Sony 8 mm video cassette recorder and format
TAPE TRAVEL
Analog, Digital Video Recorders
9.2.2.3 Magnetic Tape Types
The magnetic tape grade plays a critical role in determining the final quality of the video picture and life
of the recorder heads. Manufacturers have improved
tape materials resulting in significant improvements in
picture quality, and maintaining “clean” pictures (low signal dropout and noise) over long periods of time and after
many tape replays. For security applications it is important
to choose a high-quality tape with matched characteristics
for the VCR equipment and format used.
Most security videotape formats when grouped by size
fall into two categories: 8 mm and VHS (Table 9-1).
Video cassette recorders record multiple 2:1 interlaced
cameras best by synchronizing them sequentially using
a 2:1 sync generator. This technique provides stability,
enhances picture quality, and prevents picture roll, jitter,
tearing, or other disturbances and artifacts. If randominterlace cameras are used, they should be externally
synchronized. Table 9-2 summarizes the physical parameters and record/play times of the VHS and 8 mm tape
cassettes.
The real-time or TL VCR provides a means for recording consecutive video images over a period of time ranging from seconds to many hours and recording thousands of individual video pictures on magnetic tape. A
2-hour VHS cassette records 216,000 images of video
(2 hr @ 30 frames/sec = 216,000 frames). If camera information (ID) and time and date are coded on the tape,
equipment is available for the operator to enter the camera number, time, and date to retrieve the corresponding
images on the tape. However, to locate a specific frame,
many minutes may be needed to shuttle the tape, playback, and display the image. Locating a specific frame or
time on the tape is a lengthy process since the video cassette tape is a serial medium and retrieval time is related
to the location of the picture on the tape. The random
access nature of the DVR or optical HD performs this task
easily, resulting in quick retrieval of any image anywhere
on the disk (Section 9.3.1).
9.2.3 Time-Lapse (TL) VCR
The TL recorder is a real-time VCR that pauses to record
a single video field (or frame) every fraction of a second
or number of seconds, based on a predetermined time
interval (Figure 9-4).
Standard VCRs record the video scene in real-time: the
fields or frames displayed by the camera are sequentially
recorded on the tape and then played back in real-time,
slow-motion, or a frame at a time. In the TL mode, the
VCR records only selected fields (or frames) a fraction of
the time. Time-lapse recorders have the ability to record
in both real-time and a variety of TL ratios, which are
operator-selected either manually or automatically. The
automatic switchover from TL mode to real-time mode
is triggered by an auxiliary input to the VCR. When the
signal from an alarm device or VMD is applied to the VCR
input, it records in real-time mode for a predetermined
length of time after an alarm is received, and then returns
to the TL mode until another alarm is received.
TAPE FORMAT
LUMINANCE (Y)
BANDWIDTH
(MHz)
VHS
3.4–4.4
(1.0)
629
240
VHS-C
3.4–4.4
(1.0)
629
240
S-VHS
5.4–7.0
(1.6)
629
400
8 mm
4.2–5.4
(1.2)
743
270
Hi-8
5.7–7.7
(2.0)
743
430
DIGITAL 8
13.5 *
DV
MINI DV
CHROMINANCE (C)
CENTER FREQUENCY
(KHz)
RESOLUTION
(TV LINES)
3.375 **
500
13.5
3.375
500
13.5
3.375
500
* SAMPLING RATE
** SAMPLING RATE FOR UV CHROMINANCE SIGNALS
Table 9-1
281
VHS, S-VHS, 8 mm, and Hi-8 Parameter Comparison
282
CCTV Surveillance
PLAYING TIME (HRS)
MAXIMUM
RESOLUTION
(TV LINES)
TAPE
WIDTH
mm (inches)
CASSETTE SIZE
L × W × H (mm) *
240
12.7 (1/2)
T-60
240
T-120
TAPE FORMAT
STANDARD
(SP)
LONG
(LP)
EXTENDED
(EP)
188 × 104 × 25
0.33
0.66
1.0
12.7 (1/2)
188 × 104 × 25
1
2
3
240
12.7 (1/2)
188 × 104 × 25
2
4
6
400
12.7 (1/2)
188 × 104 × 25
2
4
6
P6-60
270
8.0 (0.31)
95 × 62.5 × 15
1
—
—
P6-120
270
8.0 (0.31)
95 × 62.5 × 15
2
—
—
P6-60
430
8.0 (0.31)
95 × 62.5 × 15
1
2
—
P6-120
430
8.0 (0.31)
95 × 62.5 × 15
2
4
—
DIGITAL 8
500
8.0 (0.31)
95 × 62.5 × 15
2
—
—
DV
500
6.35 (0.25)
125 × 78 × 14.6
3
—
—
*** MINI DV
500
6.35 (0.25)
66 × 48 × 12
1
1.5
—
VHS-C
VHS:
S-VHS
* 8 mm
** Hi-8
* TAPES AVAILABLE—15, 30, 90, 120 MINUTES
** TAPES AVAILABLE—30, 60, 80 MINUTES (SP MODE)
*** TAPES AVAILABLE—45, 90, 120 MINUTES (LP MODE)
Table 9-2
Video Cassette Recorder Tape Physical Parameters and Formats
STANDARD REAL- TIME VIDEO RECORDING: 30 FRAMES/SEC, 60 FIELDS/SEC
FRAME 1
FIELD 1
FRAME 2
2
5
4
3
FRAME 4
FRAME 3
6
8
7
9
10
11
12
13
14
T=0
T = 1/60 sec
T = 1/30 sec
TIME LAPSE
VIDEO RECORDING
TIME LAPSE RECORDER
FIELD 1
FIELD 7
FIELD 14
T=0
T = 6/60 = 1/10 sec
T = 12/60 = 1/5 sec
TIME INTERVAL
PROGRAMMED
BY OPERATOR
FIGURE 9-4
Time lapse (TL) video cassette recorder (VCR)
Time-lapse video recording consists of selecting specific
video images to be recorded at a slower rate than they are
being generated by the camera. The video camera generates 30 frames (60 fields) per second. One TV frame
consists of the interlaced combination of all the evennumbered lines in one field and all the odd-numbered
lines in the second field. Each field is essentially a complete picture of the scene but viewed with only half the
Analog, Digital Video Recorders
TOTAL RECORDING
PERIOD *
HOURS
2 **
TIME-LAPSE
RATIO
DAYS
RECORDING/PLAYBACK SPEED
(RECORDING INTERVAL)
1 FIELD
PER ___ SEC
1 FRAME
PER ___SEC
RECORDING/PLAYBACK
(PICTURES/SECOND)
FIELDS
FRAMES
.083
1:1
0.017
0.034
60
30
.50
6:1
0.1
0.2
10
5
0.2
0.4
5
2.5
12
24
1
12:1
48
2
24:1
0.4
0.8
2.5
1.25
72
3
36:1
0.6
1.2
1.7
0.85
60:1
1.0
2.0
1.0
0.5
120
5
180
7.5
90:1
1.5
3.0
0.66
0.33
10
120:1
2.0
4.0
0.50
0.25
360
15
180:1
3.0
6.0
0.33
0.17
480
20
240:1
4.0
8.0
0.25
0.13
600
25
300:1
5.0
10.0
0.20
0.10
720
30
360:1
6.0
12.0
0.16
0.08
960
40
480:1
8.0
16.0
0.2
0.11
240
283
* TAPE CASSETTE: T-120
** STANDARD REAL-TIME VIDEO
Table 9-3
Time-Lapse Recording Times vs. Playback Speeds
vertical resolution (262 1/2 horizontal lines). Therefore,
by selecting individual fields—as most TL VCRs do—and
recording them at a rate slower than 60 per second, the
TL VCR records less resolution than available from the
camera. When the TL recorded tape is played back and
viewed on the monitor at the same speed at which it was
recorded, the pictures on the monitor will appear as a
series of animated still scenes. Table 9-3 presents a comparison of TL modes as a function of TL ratio, total recording
period, recording interval fields, and fields per second
recorded.
It is apparent that the larger the TL ratio, the fewer
the pictures recorded over any period of time. For example, for a TL ratio of 6:1, the recorder captures 10 images
(fields) per second, whereas in real-time (or 1:1) it captures 60. Although the recorder is only recording individual fields spaced out in time, if nothing significant is
occurring during these times, no information is lost.
The choice of the particular TL ratio for an application
depends on various factors including the following:
• Length of time the VCR will record on a 2-, 4-, or 6-hour
video cassette
• Type, number and duration of significant alarm events
likely to occur
• Elapsed time period before the cassette can be replaced
or reused
• TL ratios available on the VCR.
To minimize tape usage and maximize information
recorded, select the lowest TL ratio consistent with the
requirement. By carefully analyzing operating conditions
and requirements and using TL, it is possible to record
events without sacrificing important information—and at
substantially less tape cost than real-time recording.
In the TL recording mode the videotape speed is much
slower than in real-time since the video pictures are being
recorded intermittently. This maximizes the use of tape
storage space and eliminates the inconvenience of having
to change the cassette every few hours. To review the tape
it is scanned at faster than normal playback speed. When
more careful examination of a particular series of images is
required, the playback speed is slowed or paused (stopped)
and a more careful scrutiny of the tape is made.
9.2.4 VCR Options
The following are some VCR options:
• Built-in camera switcher
• Time/date generator
• Sequence or interval recording of multiple cameras on
one VCR
• Interface with other devices: cash register, ATM, etc.
• Remote control via RS-232
• 12-volt DC power operation for portable use.
9.2.4.1 Camera Switching/Selecting
Time-lapse VCRs allow recording of multiple cameras
and selected playback of numerically coded cameras.
284
CCTV Surveillance
CAMERA 8
CAMERA 5
VCR OR DVR
(RECORD MODE)
G
1 2 3 4 5 6 7 8
16
CAMERA VIDEO
ENCODER FOR
16 CAMERAS
CAMERA 1
SCENE FROM CAMERA (5) STORED ON VIDEO TAPE EVERY 8TH FRAME
o
o o
3
4
G
5
6
7
8
1
G
3
2
4
G
5
6
o
o
o
G
INTERVAL =
1/30 sec × 8 =
0.266 sec
CAMERA VIDEO
DECODER FOR
16 CAMERAS
VCR OR DVR
CAMERA 5
SELECTED
(PLAYBACK MODE)
FIGURE 9-5
05
G
SCENE 5 DISPLAYED
CONTINUOUSLY AND
UPDATED EVERY
0. 266 sec
Multiplexing multiple cameras onto one time-lapse VCR
Figure 9-5 shows the technique for a 16-camera input system using 8 cameras.
The VCR multiplexes up to 16 cameras onto one videotape, reducing equipment cost by eliminating the need
for one VCR per camera input. The VCR separates the
recordings from each camera and displays the fields from
one camera. When many cameras are recorded on one
VCR, rather than sorting through scenes from all the cameras when only one is of interest, the operator can select
a particular camera for viewing. To locate a specific video
image, the operator shuttles the tape, advance or backup,
one image at a time. In operation during real-time or TL
video recording, the VCR electronics inserts a binary synchronizing code on the video signal for every image with
each camera uniquely identified. During real-time playback the scenes from any one of the 16 cameras can be
chosen for display. In Figure 9-5 camera scene 5 has been
chosen for presentation where the pictures are updated
every 0.266 seconds.
9.2.4.2 RS-232 Communications
Video cassette recorders interface and communicate twoway data with computer systems via a RS-232 port enabling
the computer to communicate with the VCR and control
it. The RS-232 port permits the recorder to communicate
via digital networks, telephone lines, dedicated two-wire
or wireless channels. The computer becomes a command
post for remote control of recorder functions: real-time
or TL mode, TL speeds, stop, play, record, fast rewind,
fast-forward, scan, reverse-scan, pause, and advance. These
remote functions are put at the fingertips of the security
operator whether in the console room, at the computer
across the street, or at a distant location.
9.2.4.3 Scrambling
Video recordings are often made that contain inherently
highly sensitive security information and there is a need for
scrambling the video signal on the recording. Equipment
Analog, Digital Video Recorders
BASEBAND
VIDEO
TRANSMISSION
OR
SCRAMBLED
VIDEO
DESCRAMBLED
VIDEO
RECORDING
285
BASEBAND
VIDEO
CAMERA
WIRED *
SCRAMBLER
RECORDED
SCRAMBLED VIDEO
ON VCR OR DVR
VIDEO
TRANSMITTER
WIRELESS *
DESCRAMBLER
VIDEO
RECEIVER
DESCRAMBLER
DESCRAMBLER
RECORDED
SCRAMBLED VIDEO
ON VCR OR DVR
*
VCR/
DVR
DESCRAMBLER
* MODES OF OPERATION:
(1) WIRELESS TRANSMISSION
(2) WIRED TRANSMISSION
(3) VIDEO CASSETTE RECORDING
FIGURE 9-6
VIDEO
MONITOR
Video recording scrambling
is available to scramble the videotape signal to prevent
unauthorized viewing of video recordings, making it very
difficult to reconstruct an intelligible picture (Figure 9-6).
The scrambling technology safeguards the video signal
as it is recorded or transmitted, producing a secure signal
unusable in its scrambled form. The scrambling code is
changed constantly and automatically and renders frameby-frame decoding fruitless. It is password-protected so
that only personnel entitled to view the tape can gain
access to it. Time access codes can be programmed to
restrict descrambling to a scheduled time interval. The
complete system consists of an encoder connected in the
camera output and decoder connected at the monitor
location.
control the on-screen editing of the videotape whenever
changes are required.
These are typical text entered by a security operator:
• Superimposed listing of cash register transactions
• Personal identification number (PIN) of an individual
using an ID card
• Verification of the authenticity of ID cards used at
remote ATMs, gas pumps, retail stores, and cashdispensing machines
• Record of action taken by security personnel reacting
to alarms or video camera activity.
9.3 DIGITAL VIDEO RECORDER (DVR)
9.2.4.4 On-Screen Annotating and Editing
Video cassette recorders have built-in alpha-numeric character generators to annotate tape with: time, date, day of
week, recording speed, alarm input, camera identifier, and
time on/off status information. In retail applications the
recorder can annotate the video image with the cash register dollar amount to check a cashier’s performance. In a
bank ATM application a video image is annotated with the
transaction number to identify the person performing the
transaction. The RS-232 interface permits the operator to
There are several key differences between DVRs and VCRs
that result in significant advantages for DVR users. The
most notable difference between the DVR and VCR is the
medium used for recording the video images. VCRs record
images on magnetic tapes, while digital systems use HD
drives, DATs or DVDs. This differentiation has significant
implications in terms of the video image quality, speed
of information retrieval, image transmission speed, and
remote monitoring capabilities. Digital video systems using
DVRs can be accessed over LAN, intranets, and the Internet. This permits security personnel to monitor remote
286
CCTV Surveillance
sites across the street, town, or locations hundreds or thousands miles away. Using an Internet browser or other application software on any PC or laptop allows security with
personnel or corporate management to view recorded digital video images at a secure IP (Internet protocol) address
from anywhere in the world.
Security systems using DVRs can play a major role in
alarm verification. Having the ability to perform video
assessment from remote locations means the system can
be used to prevent false alarm responses by security and
police personnel. The ability to remotely and instantly
view the alarm site means that if a review of the video
images indicates there are no intruders, a false alarm can
be declared and no law enforcement personnel need be
notified.
The digital video images are stored on HD drives similar
to those used in the PC industry and have storage capacities measured in hundreds of megabytes or gigabytes,
providing a low cost storage media for the compressed
video files. Small and medium-size systems use several HD
drives, while large enterprise systems use a large number of HD drives. These HD drives are synchronized and
shared to store images from many video cameras reliably,
and available for rapid access by the user. The DVR has
high reliability as compared to the VCR recorder. The
DVD provides higher image quality as compared to its
VCR predecessor.
The DVR video images can be downloaded to an external medium. The Zip file/disk or an email over the Internet is the easiest for the DVR.
9.3.1 DVR Technology
The technology difference between analog and digital
recording is that the analog tape recorder incorporates
a magnetic field to align the magnetic particles on the
surface of the VHS tape to correspond to the video signal
image. In contrast, the DVR converts the analog signal into
a digital signal of ones and zeros, compresses this digital
signal, and then stores it on the magnetic DVR HD drive,
DVD, or DAT.
The combination of affordable image compression technologies and large capacity HD drives has made the development of the DVR a reality. Although HD DVR recording
like VHS still uses a magnetic recording medium, the digital nature of the data insures that all retrieved footage is
an identical copy of the originally recorded signals.
Standard DVRs have some shortcomings when used in
mid-range and large size Enterprise systems. Since video
inputs are local to the DVR, and a camera source has
to be wired at the location of the DVR, this results in a
significant investment in cable.
DVRs are rapidly replacing the VCR as the preferred
storage/retrieval medium for video security systems. An
obvious difference between the two technologies is that
VCRs use standard VHS format magnetic tape, while digital DVRs store images on the DVR HD, DVD, DAT, or any
combination of these media.
The operators of DVR can search for recorded information based on time, date, video image, camera input,
alarm, or video motion. Operators achieve much faster
retrieval times as compared with VCRs. Rather than wading through countless frames of video information, the
DVR operator can locate the desired images in a fraction
of a second.
The DVR is superior to the VCR in image quality. The
VCR records only every other field of the video image,
while the DVR records a full frame (two fields per frame)
producing twice the resolution. The DVR digital image
does not deteriorate on playback or re-recording whereas
with the VCR there is deterioration of the image after each
new copy is made. The DVR requires far less servicing as
compared to the VCR with all its mechanical drives and
the VCR magnetic tape is prone to tape failure. DVRs offer
additional features such as remote video retrieval, combining the multiplexer with the DVR, pre- and post-image
recording, retrieval on alarm, and networking capabilities.
The basic block diagram of a DVR is shown in Figure 9-7.
The analog video signal from the camera is converted
into a digital signal via the analog-to-digital converter
(A/D) at the front end of the DVR. Following the A/D
converter is the digital compression electronics with its
programmed compression algorithm. The amount of compression (compression ratio) is based on the compression
algorithm chosen: JPEG (Joint Photographic Engineers
Group), MPEG-4 (Motion Picture Engineers Group),
(Wavelet, H.263, H.264, etc.). The compression algorithm
chosen is based on the DVR storage capacity and the image
rate and quality of the images required. Following the
compression electronics is the authentication electronics
that imbeds a security code into each image. The digitized video signal is then ready for storage in the HD
drive.
The HD drive stores the compressed video image and
other data on a magnetic coating on the HD. A magnetic
head is held by an actuator alarm and is used to write and
read the data. The disk rotates with constant rpm and data
is organized on the disk in cylinders and tracks. The tracks
are divided into sectors (Figure 9-8).
Hard Disk storage capacity is measured in hundreds of megabytes, gigabytes (1000 MByte), or terabytes
(1000 GByte). Video image retrieval is fast but not
instantaneous. There is a delay between the time an operator inputs a command to retrieve an image and when
the image is displayed on the monitor screen. With DVR
systems this time is a small fraction of a second. With VCRs
it is seconds to minutes.
Image retention time refers to how long the DVR can
record before it begins to write over its oldest images.
The amount of recording time required depends on the
Analog, Digital Video Recorders
INPUT:
ANALOG VIDEO
VIDEO *
COMPRESSION
ANALOG/DIGITAL
CONVERTER
287
SECURITY
AUTHENTICATION
DIGITAL **
STORAGE
OUTPUT:
DIGITAL
DATA RETREIVAL
(ACCESS)
VIDEO
DE-COMPRESSION
DIGITAL/ANALOG
CONVERTER
ANALOG
* COMPRESSION RATIO BASED ON STORAGE CAPACITY
AS WELL AS NUMBER AND QUALITY OF IMAGES STORED
** STORAGE MEDIA: MAGNETIC HARD DISK
FIGURE 9-7
Digital video recorder (DVR) basic block diagram
application: local codes, regulations, or business classification. Mandated storage times generally range from a
week to a month or months. To record for longer periods of time or to archive, a compact disk (CD), additional
HD drives, or DAT recorders are used. Since real-time
recording and high image resolutions consume significant
HD space quickly, the new compression schemes using
MPEG-4, H.264, and others are needed to offset these
higher image per second (IPS) recording rates and image
quality requirements.
In summary, DVRs offer these advantages over analog
recorders: (1) better picture quality, (2) less maintenance,
(3) random access search, (4) pre- and post-alarm recording, (5) built-in or optional multiplexer, (6) expandable
storage for longer recording time, (7) real-time and TL
event recording modes, (8) network interface through
LAN, WAN, and Internet, (9) motion detection, and
(10) password protection.
9.3.1.1 Digital Hardware Advances
9.3.1.1.1 Hard Disk Drive storage
Hard disk drive storage capacity and speed have increased
dramatically in the past years and the trend will continue.
DVRs can include built-in 40-, 80-, 160-, or 250 GByte
HD drives that can provide storage of high-resolution
monochrome or color images for days, weeks, or months.
To achieve reliability, the older small computer system
interface (SCSI) drives were previously specified as the
choice for DVR applications. Now the integrated drive
electronics (IDE) drives offer similar performance and
reliability at a much lower cost. These IDE HD drives
found in mid-range and enterprise recorders provide storage in the terabyte (1000 GByte) range. These enterprise
class recorders can have almost unlimited storage using
external HD including configurations that can tolerate
a failed drive without losing any video recorded video
images. The IDE HD drives have narrowed the gap in
speed and reliability compared to the relatively expensive
SCSI HD drives, making IDE ATA100 thermally compensated drives a popular storage media for DVRs. Figure 9-9
summarizes Digital video storage media.
9.3.1.1.2 Video Motion Detection (VMD)
Every video scene at some time has video motion caused by
a person moving, an object moving, or some other motion
activity. Many DVRs have VMD built into them. Digital signal processing (DSP) is used to detect the motion in the
video image and cause some form of alarm or video representation on the monitor screen. This feature enables
the DVR to remain in an inactive or TL operational mode
until activity occurs, and increases the recording rate and
displays the alarm on-screen in the form of an outlined
object of the person moving or other activity in the camera field of view. This technology increases overall video
288
CCTV Surveillance
CAMERA
ANALOG TO
DIGITAL
CONVERTER
COMPRESSION
MAGNETIC
HARD DISK
MEDIA
PLAYBACK
(RETRIEVE)
HEAD
DE - COMPRESSION
DIGITAL TO
ANALOG
CONVERTER
SCENE
DIGITIZED TRACKS
OF VIDEO
MAGNETIC RECORDING
(RETRIEVING) HEADS
DIGITAL VIDEO SIGNAL
DISK
ROTATION
MONITOR
VIDEO
SIGNAL
T
0
1/60
1 FIELD
2/60 = 1/30
3/60
(SEC)
1 FIELD
1 FRAME
3 FIELDS
FIGURE 9-8
Magnetic hard disk (HD) digital video storage
image storage time since the DVR does not have to record
non-events or records them at a slower rate. When activity
occurs it becomes visible on the video screen or causes
some other alarm notification. One should be aware that
for prosecutorial applications images acquired through a
motion detection DVR may be inadmissible if there is no
recording made prior to and after the time of the event.
The ability to respond to alarm inputs—whether individual contact closures or software-generated procedures—
are a major feature of DVRs and these important capabilities should be included in the design. The ability to
incorporate immediate automatic recording on alarm is
one of the features that puts the basic DVR a step above
the off-the-shelf PCs equipped with video capture cards
and base level software for setting parameters.
Digital video recorders with internal VMD create a
searchable audit-trail by camera every time there is motion.
Unlike when using the VCR, security personnel can quickly
find the video images of interest on the DVR by date, time,
image motion activation, or alarm input.
9.3.1.1.3 Optical-Disk Image Storage
For very long-term video image recording and archiving, an
optical-disk medium is chosen. Optical storage media are
durable, removable disks that store video images in digital
format. There are two generic systems available: (1) nonerasable write-once read-many (WORM) and (2) erasable.
These two electro-optical storage systems are described in
the following sections. The optical disk recorder stores
the video image on a disk using optical recording media
rotating at high speed. The picture is stored and identified
by coding the signal to the specific camera and the time and
date at which it was put on disk. At a later time the stored
picture can be retrieved in random access at high speed.
Most optical disks used in security applications are WORM
disks since these are admissible in law enforcement investigation and prosecution cases.
9.3.1.1.4 Non-Erasable, Write-Once Read-Many (WORM)
Disk
The WORM optical-disk recording system provides a
compact means to store large volumes of video images.
The drive uses a 800-megabyte, double-sided, removable
diskette, which is rugged and reliable. In security applications, a WORM drive has a significant advantage over
magnetic recording media because the optical image cannot be overwritten, eliminating the risk of accidental or
intentional removal or deletion of video pictures. This is
Analog, Digital Video Recorders
ANALOG TO
DIGITAL
CONVERTER
COMPRESSION
RECORDING
HEAD
MAGNETIC
HARD DISK
MEDIA
PLAYBACK
(RETRIEVE)
HEAD
DE-COMPRESSION
289
DIGITAL
TO ANALOG
CONVERTER
CRT
MONITOR
(ANALOG)
ANALOG
CAMERA
IP DIGITAL
CAMERA
RECORDING MEDIA
TYPE/FORMAT
STORAGE
RECORDING
TIME (HRS)
NUMBER OF
CAMERAS
FRAME
RATE (fps)
COMPUTER
HARD DRIVE (HD)
60, 120 GB INTERNAL
160, 500 GB EXTERNAL
HOURS TO
MONTHS
4,8.16.32
EXPANDABLE
UP TO
120
650 MB
BACKUP OF
HARD DRIVE
—
—
60, 120 GB INTERNAL
160, 500 GB EXTERNAL
333, 666,
1555, 3444
4
4
30
30
250 MB
DEPENDS
ON DATA
—
—
CD-ROM
CD-R
CD-RW
MOBILE
HARD DRIVE (HD)
REMOVABLE
ZIP
FIGURE 9-9
FLAT PANEL DISPLAY
(DIGITAL)
Digital video storage media
important in law enforcement applications. The WORM
disk containing the video images is removable and can
therefore be secured under lock and key, stored in a vault
when the terminal is shut down or the system is turned
off, or sent to another location or person. Reliability is
extremely high, with manufacturers quoting indefinite life
for the disk and a minimum mean time between failure
(MTBF) of greater than 10 years. The reason for this
longevity is that nothing touches the disk itself except a
light beam used to write onto and read from the disk.
9.3.1.1.5 Erasable Optical Disk
Erasable optical-disk media is now available that can be
erased (as on present magnetic media) and overwritten
with new images (Figure 9-10). Each image stored on the
optical HD is uniquely identified and may be retrieved in
random access in less than 1 second. Optical disks store
huge amounts of data; approximately 31 reels of data
tape are equivalent to one single 5¼-inch-diameter optical
disk—the size of an ordinary compact disc. Standard optical disks can store many terabytes of information. While
most optical disks used in security are WORM, erasable
optical disks are also in use. Erasable disks use the principle of magneto-optics to record the video information
onto the disk in digital form. The video image data or
other information is erasable, allowing the same disk to
be reused many times, just like magnetic HD. Reading,
writing, and erasing the information on the optical disk
is done using light energy and not magnetic heads that
touch or skim across the recording material. Therefore,
magneto-optical disks have a much longer life and a higher
reliability than magnetic disks. They are immune to wear
and head crashes as occasionally occur in magnetic HD
drives. This catastrophic event occurs when a sudden vibration or dust particle cause the mechanical head in the
drive to bump into the recording material, thereby damaging it. In the case of the optical disk, the opto-magnetic
layer storing the information is imbedded within a layer
of plastic or glass, protecting it from dust and wear. The
optical disk is an excellent medium when large amounts
of high-resolution video images need to be stored and
retrieved for later use.
9.3.1.1.6 Digital Audio Tape (DAT)
Digital audio tape is a format for storing or backing up
video data (originally for music) on magnetic tape. It was
co-developed in the mid-1980s by the Sony and Philips
Corporations. DAT uses a rotary-head (or helical scan)
format where the read/write head spins diagonally across
the tape like a VCR. It uses a small 4 mm-wide tape having a
signal quality that can surpass that of a CD and can record
data (video images) at a rate of 5 MBytes/minute. The
DAT storage capacity is 6 GBytes on a standard 120-minute
cartridge. DAT decks have both analog and digital inputs
and outputs.
290
CCTV Surveillance
VIDEO IN
MODULATED LASER
LENS
OPTICAL
POLARIZER
OPTICAL
ANALYZER
LENS
VIDEO OUT
OPTICAL
TRACKS
BIT
DIGITAL
VIDEO
DETECTOR
MAGNETO - OPTIC
LAYER
BONDING
AGENT
OPTICAL DISK
BIAS
COIL
FIGURE 9-10
Erasable optical disk recording
9.3.1.2 Digital Storage Software Advances
Digital technology, faster microprocessors, high density
inexpensive solid state memory, and the availability of
larger and cheaper HD drives have made DVRs affordable
in security video applications. Adding the combination
of affordable image compression technologies and large
capacity HD drives has made the development of the DVR
a reality. Although HD DVR recording like VHS still uses a
magnetic recording medium, the digital nature of the HD
data permits transmitting the video images over networks
to remote sites and insures that all retrieved images are
identical copies of the original images.
Digital video images can be stored on a HD but several
things must be considered since around-the-clock recording of pictures requires a vast amount of storage space. To
overcome this, when there is motion in the scene, a startand-stop recording mode can be implemented. Alternatively only a few frames per second can be stored much as the
TL recorder in the analog régime. In Table 9-4 examples of
the image sizes and storage requirements for five different
image resolutions and image recording rates are given.
The formula for calculating these storage requirements
and image rates is:
DVR storage capacity =
HD drive storage capacity
image size × pictures per day
= storage time in days or hours
Example: Calculate the storage capacity of a 250 GByte
DVR to record an image frame rate of 15 images per
second with a picture quality set to standard = 18 KByte:
DVR storage capacity =
250 GByte
18 KByte × 15 ips × 86400 sec
= 10.7 days
= 10 days and seventeen hours
This calculation is based on the DVR recording continuously at the selected recording speed for a 24-hour period
(86,400 sec). This is a worst case scenario, since the DVR
can be programmed to record only if motion is present or
at selected times of the day. Both of these settings will dramatically increase the unit storage potential and eliminate
the storage of unneeded or useless video images.
Analog, Digital Video Recorders
PICTURE*
QUALITY
DVR RECORDING SPEED (IMAGES/SEC)
†
(250 GByte HARD DRIVE)
30
15
10
7.5
5
3
1
2D/6H
4D/9H
6D/14H
8D/9H
13D/4H
22D/0H
66D/2H
60
291
HIGHEST
1D/3H
HIGH
1D/17H
3D/3H
6D/14H
9D/21H
13D/4D
19D/19H
33D/0H
99D/4H
STANDARD
2D/16H
5D/1H
10D/17H
15D/9H
20D/12H
30D/19H
51D/9H
154D/7H
BASIC
4D/0H
7D/16H
15D/9H
24D/0H
30D/19H
46D/4H
77D/2H
231D/9H
LOW
8D/0H
15D/10H
30D/19H
46D/4H
61D/16H
92D/12H
154D/7H
462D/20H
IMAGES/DAY = 60 pps—5.184 MBytes
30 ips—2.6 MBytes
15 ips MODE—1.296 MBytes
7.5 ips MODE—648 KBytes
5.0 ips MODE—259.2 KBytes
3.0 ips MODE—86.4 KBytes
1.0 ips MODE—86.4 KBytes
*IMAGE SIZES—HIGHEST = 42 KByte
HIGH = 28 KByte
STANDARD = 18 KByte
BASIC =12 KByte
LOW = 6 KByte
**
ALL RECORDING TIMES BASED ON 250 GByte HARD DRIVE
†
RECORDING TIME: 1D/3H = 1 DAY AND 3 HOURS
FORMULA FOR CALCULATING NUMBER OF IMAGES STORED ON 250 GByte HARD DISK MEMORY
RECORDING TIME =
HARD DISK STORAGE
IMAGE SIZE × PICTURES/DAY
= STORAGE TIME IN DAYS AND HOURS
EXAMPLE: HARD DISK STORAGE = 250 GByte
IMAGE SIZE = 12 KByte
IMAGES/SEC = 10 ips
SECONDS IN A DAY = 86,400 Sec
PICTURES/DAY = 10 ips × 86,400 Sec
RECORDING TIME =
250 GByte
12 KByte × 10 ips × 86,400 Sec
= 24.1 DAYS
Table 9-4 Digital Storage Requirements and Images Per Second (IPS) for Five Different Image
Resolutions on a 250 GByte hard drive
9.3.1.3 Transmission Advances
A fast-growing application for DVRs and digital storage
systems is for the remote video retrieval via modem or network using LAN, WAN, and wireless (WiFi). Transmission
speeds are increasing, compression algorithms are improving, and remote video solutions implementing automated video surveillance (AVS) at remote sites are being
installed. New software that allows viewing of multiple IP
addressable digital recorders from a central location is
increasing and will become a must-have feature. Many
companies are implementing digital recording for remote
viewing in video systems using LANs, WANs, and Webbased systems. A major advantage of an IP-addressed network is its ability to receive video signals anywhere using
equipment ranging from a simple Internet browser to
special client-based application software. Using these networks eliminates the need to run new cabling and provides
an easy solution for future system expansion.
Cellular transmission is the slowest transmission method
for video transmission and is not widely used in the video
security market. However, in areas that offer no other service it is the only way to offer remote surveillance. The
transmission speed of the cellular system is 9.6 Kbps (bits
per second) and is increasing with time. Dial-up or pub-
lic switched telephone network (PSTN) is the most common method of the available DVR transmission methods,
but since it was designed for the human voice and not
high-speed video transmission, it does not provide high
bandwidth or speed of transmission. This type transmission mode has a maximum speed of 56 Kbps but in spite
of the relatively slow service, its cost and availability are
the major factors for its continued use.
Integrated systems digital network (ISDN) is a digital
phone line with two 64 Kbps channels. Competition from
cable and digital subscriber line (DSL) service has reduced
the pricing to acceptable levels for the video security market. DSL technology has sufficient bandwidth for highspeed access to the Internet and live video monitoring.
This digital broadband link directly connects a premise
to the Internet via existing copper telephone lines. The
DSL speed is listed as nearly 1.5 Mbps but depends on
the routing, the distance from the network hub, and the
number of people on the network.
A very high-speed, expensive digital system using dedicated lines is the AT&T T1 network transmitting up to
1.544 Mbps. The T3 lines have almost 30 times the capacity
of T1 lines and can handle 44.736 Mbps of data.
The widest transmission network is achieved using a
fiber-optic optical carrier (OC) transmission channel. The
292
CCTV Surveillance
TRANSMISSION
TYPE
TYPICAL
DOWNLOAD
SPEED
TRANSMISSION
TIME FOR 25 KByte
IMAGE (sec)
MAX. FRAME
RATE FOR
25 KByte IMAGE
CONNECTION
MODE
PSTN
45 Kbps
6
10 Frames/min
DIAL UP
ISDN
120 Kbps
2
0.5 Frames/sec
DIAL UP
IDSL
150 Kbps
2
0.06
ADSL—LOW END
640 Kbps
0.3
3
ADSL—HIGH END
5 Mbps
0.05
20
HDSL
1.5 Mbps
0.2
6
VDSL
20 Mbps
0.01
80
CABLE MODEM
750 Kbps
0.3
3
T1
1.5 Mbps
0.2
6
T3
45 Mbps
0.007
180
10BaseT
5 Mbps
0.05
20
100BaseT
50 Mbps
0.005
200
1000BaseT
500 Mbps
0.0005
2000
OC3
155 Mbps
0.0019
620
OC12
622 Mbps
0.0005
2500
FIREWIRE *
400 Mpbs
0.0008
1600 Frames/sec
DIRECT CONNECTION
DIRECT CONNECTION
* APPLE COMPUTERS VERSION OF IEEE STANDARD 1394
IDSL: ISDN DSL
HDSL: HIGH BIT-RATE DSL
ADSL: ASYNCHRONOUS DSL
VDSL: VERY HIGH DATA RATE DSL
Table 9-5
Parameters of Digital Transmission Channels for DVR Use
OC is used to specify the speed of fiber-optic networks conforming to synchronous optical network with synchronous
optical network (SONET) standards. SONET is a physical layer network technology designed to carry large volumes of traffic over relatively long distances on fiber-optic
cabling. SONET was originally designed by the American
National Standards Institute (ANSI) for the public telephone network in the mid-1980s.
FireWire is an ultra high-speed serial data connection
developed by Apple Computer. The technology provides
a high-speed serial input/output bus for computer peripherals that can transfer data at speeds of 400 Mbps. It
is especially well suited for transferring very large DVR
video image files for viewing or archiving. Table 9-5 summarizes the parameters of available digital transmission
channels.
9.3.1.4 Communication Control
All the functions available on the VCR and DVR machines
can be controlled remotely using communications via
the RS232 port(s) on the devices, and transmitted bidirectionally over the network to remote locations. Like-
wise camera functions (zoom, focus, iris, presets, etc.),
alarms, pan/tilt, and any internal DVR programming can
be done remotely.
9.3.2 DVR Generic Types
The DVDs can be divided into four groups or hardware
implementations:
•
•
•
•
DVR
DVR
DVR
DVR
in a Box-PC Card and a PC
Basic Plug-and-Play VCR Replacement
Multiplex
Multi-channel.
9.3.2.1 DVR in a Box
The DVR in a box is implemented by adding a PC Board
card to the standard PC computer that instantly turns
the PC into a DVR. The PC card has four video inputs
providing a four-channel DVR. It seems simple to do but
it does have limitations. The DVR should be a dedicated
system operating alone. Mixing and matching the DVR
Analog, Digital Video Recorders
with other software programs can cause the total system
to crash. Another shortcoming of many DVR cards is that
they do not supply alarm inputs or outputs thus creating
a very limited application machine.
293
available are the ability to connect and perform remote
video retrieval via: modem, wired LAN, WAN, Internet, or
wireless WiFi.
9.3.2.4 Multi-Channel
9.3.2.2 DVR Basic Plug-and-Play VCR Replacement
The DVR basic Plug-and-Play differs from the DVR in a
box in that it is a separate component designed and built
specifically to be a DVR. The DVR basic is a self-contained
unit having all the front panel controls that the standard
industrial real-time/TL VCR has. These DVRs generally
have a single- or four-channel video input capability and
offer a minimum of setup parameters to permit the user
to customize the picture quality, pictures size, or alarming features to meet the particular application. This DVR
has been designed as a drop-in replacement for an existing
analog VCR.
9.3.2.3 Multiplex
The multiplex DVR is the largest of the four groups of
DVR types used for video recording. The machine combines an 8- or 16-channel multiplexer with the DVR unit.
This multiplex DVR shares the video input in the same way
as the standalone video multiplexer. The combined DVR
and multiplexer has the advantage that the installer no
longer has to worry about the interface wiring and compatibility of setup programs between the two devices. Some of
the features that have been included in the multiplex DVR
are: (1) motion or activity detection, (2) remote video
retrieval by a modem over a digital channel, (3) alarm
inputs-contact closures or software generated, and (4) ability to adjust the IPS recorded. Recorders equipped with
a multiplexing capability allow users to watch live and
recorded images on one monitor while the multiplexing
DVR continues to record.
Multiplex DVR technology should be capable of multitasking, duplexing, and triplexing by performing the
record and playback and live viewing functions simultaneously. VCRs cannot do that but most DVRs can.
The operator using a multiplex DVR with triplex
functionality can simultaneously review and archive the
video images without interrupting the recording process.
Uninterrupted recording ensures that no event will go
unrecorded or missed.
One shortcoming of the multiplexed video recorder is
that it does not record all camera images from each camera
connected to the system simultaneously. It incorporates a
time-share system to record multiple camera inputs one at
a time.
The multiplex DVR technology allows a single unit to
replace not only the recorder but also all the accessory
items needed to run a VCR-based video system. There is no
need for separate multiplexers, switchers, or any devices
other than the camera, lens, and monitor. Other features
Both the multiplexed and multi-channel DVR systems use
a system called redundant array of independent disks
(RAID) to control the multiple HD drives and provide
management and distribution of the data across the system. Different RAID levels are used depending on the
application to optimize fault tolerance, the speed of access,
or the size of the files being stored. RAID Levels 1 and 5
are the most commonly used in video security applications.
The multi-channel DVR is designed for high-end applications having many cameras and monitors. Applications
using these systems require multiple, month-long storage
times, real-time video recording, and a very large number (hundreds) of video inputs. Multi-channel DVRs allow
cameras to be recorded at 60 IPS, whereas in the multiplex unit the cameras are time-shared between the images
displayed.
The primary difference between a multiplexed DVR
and a multi-channel DVR is that the multiplex recorder
uses only one display while the multi-channel DVR has
multiple displays, either split screen or multiple monitors.
Instead of time-sharing the recorded information, the
multi-channel unit records all camera images at 30 IPS
simultaneously. The system offers the highest performance
and playback in a multiple camera system.
The multi-channel DVR units have large HD drives
with capability to store an excess of 480 GByte data and
expanded storage derived from additional HD external
memory and DAT and jukebox storage systems controlled
by RAID controllers.
Multi-channel DVRs using many HD drives require coordination and control. In order to store and protect as
much information as possible, the RAID must control the
HD drive or a DAT jukebox system. The RAID capability controls and protects the HD drive data and provides
immediate online access to data despite a single disk failure. Some RAID storage systems can withstand two concurrent disk failures. RAID capability also provides online
reconstruction of the contents of the failed disk to a
replacement disk.
9.3.2.4.1 Redundant Array of Independent Disks (RAID)
A redundant array of independent disks is a system using
multiple HD drives to: (1) share or replace data among
the drives and/or (2) improve performance over using
a drive singularly. Originally RAID was used to connect
inexpensive disks to take advantage of the ability to combine multiple low-cost devices using older technology into
an array that together offered greater capacity, reliability, and/or speed than was affordable in a singular device
294
CCTV Surveillance
using the newest technology. At its simplest level, RAID
is a way of combining multiple hard drives into a single
logical unit. In this way the operating system sees only
one storage device. For the purposes of video security
applications, any system that employs the basic concept of
recombining physical disk space for purposes of reliability
or performance is a RAID system.
This system was first patented by IBM in 1978. In 1988,
RAID Levels 1 through 5 were formally defined in a paper
by Patterson, Gibson, and Katz. The original RAID specification suggested a number of prototype RAID levels or
combinations of disks. Out of the many combinations and
levels only two are in general use in security systems: RAID
Level 1 and Level 5. RAID Level 1 array creates an exact
copy (or mirror) of all data on two or more disks. This
is useful for systems where redundancy is more important
than using the maximum storage capacity of the disk. An
ideal RAID Level 1 set contains two disks which increases
reliability by a factor of two over a single disk. RAID Level
1 implementation can also provide enhanced read performance (playback of the video image), since many implementations can read from one disk while the other is busy.
The RAID Level 5 array uses block-level striping with
parity data distributed across all member disks. RAID Level
5 is one of the most popular RAID levels and is frequently
used in both hardware and software implementations. Virtually all storage arrays offer RAID Level 5.
Summarizing the two most common RAID formats
found in DVR video security systems:
1. RAID Level 1 is the fastest, fault-tolerant RAID configuration and probably the most commonly used. RAID
Level 1 is the only choice in a two drive system. In this
two drive system the mirrored pair mirror each other
and looks like one drive to the operating system. The
increased reliability in this configuration is exhibited in
that if one drive fails the video image data is available
from the other drive.
2. RAID Level 5 provides data striping at the byte level
and also uses stripe error correction information.
This results in excellent performance and good faulttolerance. Level 5 provides better storage efficiency
than Level 1 but performs a little slower.
9.3.2.5 Network Video Recorder (NVR)
The future of digital video recording will be based on current information technology (IT) infrastructure, namely
networking. By employing automatic network replenishment technology, the NVR can cope with network downtimes without sacrificing recording integrity. The concept of a virtual HD eliminates the concern of HD sizes.
Figure 9-11 shows a block diagram of the NVR system.
In a new installation where the designer has free reign
to design the solution a NVR serving the storage requirements of an entire Enterprise is a choice to consider. One
issue to consider if a NVR is used is that video images
require large storage data files, and for the NVR installation a separate dedicated network for security may be
necessary. Another consideration is that, even to simply
receive video, knowledge of setup parameters for the individual camera is necessary, and the NVR would need to
be programmed accordingly. Moving to a digital recording solution, whether the DVR, NVR, or a hybrid system
combination of DVR, NVR requires careful planning and
design.
The NVR solution is a system that uses digital or analog cameras converted to IP cameras using a network
server. This digital data is delivered to a network in accordance with the TCP/IP transport protocol and recorded
by a NVR.
The HD is usually controlled by a RAID Level 5 controller which can be expanded to other HD drives for
increased storage capacity. To overcome storage shortcomings in the midsize and larger systems the NVR
is used.
A DVR’s capacity is based on the number of HD drives
and the storage capacity of each HD. For large numbers of cameras and long archiving times separate DVR
units are required. Image retrieval across separate units
becomes impractical since most multiplex DVRs are of
a one channel design. In order to accommodate 4, 9,
16 or more video inputs, internal or external multiplexers are used. The requirement to time-share the cameras
means the IPS usually drops to a few. Dedicated DVRs do
not take advantage of common IT principles like RAID
storage.
An NVR is basically a standard networked PC with a
software application that controls the flow of digital video
data. Thanks to the availability of network interface, a concept called virtual HD drive can be realized. The virtual
memory concept is commonplace in today’s computer systems. The central processing unit (CPU) is made to accept
the larger virtual size because of a logic unit—the memory management unit (MMU)—which is responsible for
loading and unloading just a section of memory that the
CPU currently needs. The concept is used for digital video
recording. The data that has been successfully copied over
the network may then be erased from the local HD which
frees capacity on the local HD drive. The net effect is that
the local HD will never fill up as long as the network storage device can accept the data. The virtual HD makes the
retrieval of recorded video footage especially convenient:
instead of searching over several physical disk volumes the
user always sees a single disk of sufficient capacity.
The most important question that must be considered
before attempting remote video surveillance is whether the
network available has sufficient bandwidth for video transmission to the remote site. Bandwidth requirements for
quality video transmission range from 256 Kbps to 1 Mbps
depending on the video compression method used, the
image quality required, and image refresh rate (IPS) used
Analog, Digital Video Recorders
295
SITE 2
SITE 1
ANALOG
CAMERA(S)
ANALOG
CAMERA(S)
SERVER
SERVER
ROUTER
*
ROUTER
DIGITAL IP
CAMERA(S)
*
DIGITAL IP
CAMERA(S)
INTERNET
LAN/WAN
NETWORK **
VIDEO
WIFI
RECORDER
SITE 3
* DIGITAL COMPRESSED VIDEO
(MJPEG, MPEG-2, MPEG-4).
*
** SUFFICIENT STORAGE TO SUPPORT ALL SITES
ANALOG
CAMERA(S)
SERVER
SECURITY AUTHENTICATION.
ROUTER
RAID LEVEL 5 CONTROLLER FOR
EXPANDED STORAGE CAPACITY.
FIGURE 9-11
DIGITAL IP
CAMERA(S)
Network video recorder (NVR) system block diagram
by the application. Most LAN or WAN systems can operate
successfully using industry standard 10BaseT or 100BaseT
Ethernet supported by standard computer operating systems. If the remote viewing system does not use the Web
for its connection it is called an intranet. The intranet
IP-address assignments and network parameters are controlled by the in-house network manager.
9.3.2.6 Hybrid NVR/DVR System
The hybrid DVR/NVR system incorporates elements of
both the DVR and NVR. This type of system uses
distributed architecture with analog cameras connected
to IP video servers and IP cameras connected directly to
the network. The IP video from the IP cameras and the
IP video servers are both stored on a server connected
to the network. The NVR solution may be the most costeffective if installed on an existing shared network, but
it is highly debatable whether many facilities would allow
the increased network traffic created by the video images.
Hybrid DVR/NVR solutions open up exciting possibilities
in that they can use legacy analog cameras and existing
video cabling as well as IP cameras. The hybrid solution
permits the centralization of the system configuration,
leading to greater flexibility in locating the equipment to
where it is most convenient.
9.3.3 DVR Operating Systems (OS)
The terms operating system (OS) and platform are familiar
terms associated with computer systems. DVRs are computers designed to record video information with other
features specifically tailored to the security application.
Fundamentally an OS does several things: (1) manages the
hardware of the computer system defined as the CPU processor, memory, and disk space, (2) manages software and
other housekeeping functions, and (3) provides a consistent way for applications to interface with the hardware
without having to know all the details about that hardware.
Today’s OS take on several forms including the Microsoft
Windows family and Linux embedded formats.
Manufacturers of DVR products have built their systems on a variety of proprietary OS platforms. Windows
296
CCTV Surveillance
systems include the Windows 9X, XP family, Windows
NT, and Windows 2000. Embedded proprietary OS platforms like UNIX are specifically designed to run on a
unique DVR product. By nature, this proprietary OS is
used by a single manufacturer unless licensed to other
competitors. One concern regarding these proprietary
platforms is that their distribution is limited and the
user should question how extensively the product has
been field-tested against a wide variety of software security
threats.
9.3.3.1 Windows 9X, NT, XP, and 2000 Operating
Systems
The Microsoft Windows 9X, 16 bit families of OS namely
Windows 95, 98, and ME have been the primary platforms
used by DVR manufacturers. Several reasons why this family of OS has been so popular with the creators of DVR
products are that the product is relatively stable, familiar to
most users, and less expensive than its client–server counterparts. Since no major changes in the OS have occurred
over the lifetime of this family of products, developers of
DVR software have been able to avoid making costly software rewrites. There is a downside, however, to the security
aspect of the Microsoft 9X family of OS. The Windows 9X
family has a fundamental flaw that has been recognized
by Microsoft and by the manufacturers of DVR products
(see Section 9.3.8).
Significant improvements in Windows 2000 were modifications to the OS core to prevent crashes, to enhance
dynamic system configuration, and to increase system
uptime, and providing a unique method for self repair.
The Windows 2000 has built-in tools that make the OS
applications easier to deploy, manage, and support. Centralized management utilities, troubleshooting tools, and
support for self-healing applications make the management of an IT infrastructure easier. Windows 2000 also
offers use of HD drives and enhanced level of hardware
support. These improvements include the largest database
of third-party drivers and support for the latest hardware
standards. This makes it easier for HD drive users to easily
upgrade to the latest versions of software released by the
HD drive manufacturer.
9.3.3.2 UNIX
The UNIX OS is designed to be used by many people at
the same time. The UNIX is a multi-user and multi-tasking
OS developed at the Bell Telephone Laboratories, NJ, by
Ken Thompson and Dennis Ritchie in 1969. It is widely
used as the master control program in workstations and
servers, and as an embedded OS in proprietary DVRs. It
has the Internet TCP/IP protocol built-in and is the most
common OS for servers on the Internet.
9.3.4 Mobile DVR
Mobile DVRs are embedded systems designed specifically
for use in vehicles and rapid deployment systems (RDS)
and for short-term security installations (see Chapter 21).
These mobile DVRs are small and rugged, vibration
and shock resistant, and the choice for vehicle security,
RDS, and surveillance applications are when portability is
necessary.
The mobile DVR system can connect from 1 to 4 video
cameras and record and display at a full 30 IPS with audio
recording as an option. LAN and Internet interface connections are available and vehicle status and speed can be
displayed on the recording. The camera images can be
displayed individually at full screen or in a quad format.
Auto switching from camera to camera is supported and
the dwell time per camera can be set by the user. Four
input alarm contacts allow each camera to be recorded
on receipt of an alarm signal. Video can be recorded continuously, on motion, on alarm or by schedule. The VMD
zones can be set in each camera, or the whole field of view
of each camera can be used as the motion criterion. Video
files can be searched by date, time, and alarm state in a single or quad configuration. ID and password protection are
provided at different levels from system manager to system
operator. A standard Internet browser for remote connection with user ID protection can monitor the mobile
DVR site through the Internet, LAN, WAN, or wireless
WiFi.
These mobile DVR systems provide far superior performance over traditional analog VHS VCRs which are
prone to hardware failure due to humid and dusty environments and to shock and vibration. These DVRs contain rugged 30 GByte HD drives with input/output RS-232
control ports. DVRs are available having a microcontroller that translates its pushbutton commands into the
Sony/Odetics control protocol for full configuration and
control through the RS-232 control port. A serial connector allows Windows Control Software to be used for
PC-based control and configuration. Figure 9-12 shows
examples of fixed mobile DVRs.
9.3.5 Digital Compression, Encryption
Video compression is the science of eliminating as much
digital data in the video signal as possible without it being
evident to the observer viewing the image. Today’s systems have compression ratios ranging from 10:1 to 2400:1
making it possible to transmit or record huge amounts of
video data. Basic video compression methods can be classified into two major groups: lossy and lossless. Lossy techniques reduce data both through complex mathematical
algorithms and through selective removal of visual information that our eyes and brain usually ignore. Lossless
Analog, Digital Video Recorders
(A) FIXED DVR
297
(B) SMALL MOBILE DVR
(C) HARDENED MOBILE DVR
FIGURE 9-12
Compact PC-based fixed and mobile DVRs
compression, by contrast, discards only redundant information making it possible to reconstruct the exact original
video image signal.
The need for recording and storing days of video image
scenes requires that the signals be compressed to reduce
the file size. There are several different compression algorithms utilized in digital VCRs that are mostly derived from
the JPEG, MPEG, Wavelet, H.263, and H.264 algorithms.
Both JPEG and MPEG are both based on the discrete
cosine transform (DCT) in which blocks of 8 by 8 pixels
are grouped and then transformed into the frequency
domain. The Wavelet algorithm transforms the entire picture into the frequency domain, resulting in relatively small
file sizes as compared to the DCT-based algorithms. Likewise, the H.263 and H.264 are designed for low bit rate
systems.
In a typical video signal one image is similar to the
next, and it is possible to make a good prediction of what
the next frame or field in the sequence will look like. It
is also possible to bi-directionally interpolate images based
on those that came before and after. The method is to
compare the most recent image with the previous image
and determine if there was a change, and then decide
whether to store or not to store those frames if there has
been a change.
There are many techniques used to compress the video
image for storage in a DVR. One method is redundancy
reduction and is accomplished by removing duplication
from the signal source before it is compressed and stored.
Three forms of redundant reduction are:
1. Spatial: Correlation is between neighboring pixel values
2. Spectral: Correlation is between different color planes
or bands
3. Temporal: Correlation between adjacent frames in
the sequence.
A second form of reduction is called irrelevancy reduction. This method omits parts of the signal that will not
be noticed by the observer. Two areas described by the
Human Visual System (HVS) organization are in the lowfrequency visual response and color recognition areas.
298
CCTV Surveillance
9.3.5.1 JPEG
The JPEG compression algorithm was introduced in 1974
by the Joint Photographic Expert Group and uses the DCT
compression algorithm based on a video stream of 8 × 8
pixels. This is the algorithm primarily used to download
images over the Internet and can achieve compression
ratios up to 27:1.
It is designed to exploit known limitations of the human
eye, notably the fact that small color changes are perceived less accurately than small changes in brightness.
Using all of these compression methods reduces in the file
storage size while maintaining a high-quality stored video
image.
9.3.5.2 MPEG-X
The MPEG compression algorithm uses the same DCT
compression found in JPEG. The difference is that MPEG
compression is based on motion-compensated block-based
transform coding techniques. The primary technique used
in this algorithm is conditional refresh where only changes
in the image scene are compressed and stored which
reduces the amount of storage required. This is called
inter-frame (I) compression. MPEG uses the same algorithms as JPEG to create one I-frame and then removes
the redundancy from successive frames by predicting them
from the I-frame, and in coding only the differences from
its predictions (P-frames). B-frames are bi-directionally
interpolated. MPEG compression allows for three types
of frames:
1. I-frame (compress entirely within a frame).
2. P-frame (based on predictions from a previous frame).
3. B-frames (bi-directionally interpolated from previous
and succeeding frames).
MPEG-1, MPEG-2, MPEG-4 and the latest MPEG-4 AVC
(H.264) are the four basic MPEG forms used in video
compression. Each has a different compression ratio:
1.
2.
3.
4.
MPEG-1
MPEG-2
MPEG-4
MPEG-4
= 25 to 100:1
= 30 to 100:1
= 50 to 100:1.
AVC = 50 to 200:1 (or more)
MPEG compression forms have large file sizes and therefore many of today’s DVR manufacturers have modified
this standard to meet the needs of the video security industry. These modified standards called H.263 and H.264 are
designed for low bit rate communications. H.263 is better
than MPEG-1 or MPEG-2 for low-resolution and low bit
rate images. MPEG-4 AVC (H.264) is now considered to
be the best video compression standard.
9.3.5.3 Wavelet
Wavelet compression technology is based on full-frame
information and is based on frequency not on 8 × 8 pixel
blocks. It compresses the entire image—both the high and
low frequencies—and repeats the procedure several times.
Wavelet compression can provide compression ratios up
to 350:1.
9.3.5.4 SMICT
Super motion image compression technology (SMICT) is
a proprietary video compression technology that produces
a small file size for high resolution image reproduction.
The OS is Windows 2000 or Windows XP. SMICT can
provide compression ratios from 40:1 to up to 2400:1. Typical single image file sizes range up to 2500 Bytes. The
SMICT compression algorithm includes video authentication. One manufacturers’ system can record 16 cameras at
the rate of 3 IPS at 220 horizontal TV lines and 320 × 240
format size onto a single 75 GByte HD for between 30 and
60 days. Recording time is higher if there is less activity in
the video image. Using a PSTN connection with a 56 KByte
modem, four cameras can be viewed at 2 IPS, each in
quad mode.
9.3.6 Image Quality
Video image quality from a DVR is dependent on the resolution in the camera image, the compression algorithm
and ratio, and the IPS displayed.
9.3.6.1 Resolution
The term resolution is often misused and misunderstood
in the security industry. In analog systems the recorded
image almost always fills the entire monitor screen. The
resolution for analog video cameras and VHS recorders is
defined as (1) number of TV lines in a horizontal width of
the screen equal to the height of the screen or (2) the
total number of horizontal TV lines across the width of
the monitor. Digital resolution refers to spatial resolution
(number of pixels per line) and the number of lines or
rows per image, again defined in pixels. Digital resolution
is also defined as the total number of pixels on the screen.
This definition not only affects the overall resolution of
the system but also the overall size of the displayed image.
The common digital monitor image sizes are defined as
1/4 CIF, CIF, and 4 CIF (see Section 9.3.7).
9.3.6.2 Frame Rate
Most video security applications require that at least 2–5
IPS per camera be recorded to ensure that enough images
are captured to clearly identify a person, an object, or
activity. A recording rate of 15 IPS is perceived as nearly
real-time. This is the minimum rate when all motions and
activities are required to be recorded as in locations such
as casinos, retail stores, banks, etc. where fast motion and
Analog, Digital Video Recorders
sleight of hand must be detected. When basic DVRs or the
multiplex DVRs are used, the number of camera inputs
will affect the IPS recording rate for each camera. Today’s
midsize recorders can record at rates from approximately
60 to 480 IPS. Dividing the IPS rate by the number of cameras in the multiplex systems calculates the average IPS
per camera recording speed. In large Enterprise systems a
multi-channel DVR or NVR recording system is required.
These systems can record a large number of cameras simultaneously so that a rate of 5 IPS or higher can be achieved.
The ability to change the number of recorded IPS per
video input is important since the main purpose of any
DVR and multiplexer is to provide a simple and costeffective method to monitor live and recorded images via
a multi-screen display. This form of TL recording eliminates
the gaps between video scenes created by conventional
sequential switchers.
9.3.6.3 Bandwidth
When video images from DVRs are transmitted to remote
locations, the image frame rate and resolution are directly
affected by the bandwidth of the transmission network.
As a rule of thumb, the wider the network bandwidth,
the more the IPS, and the better the resolution (more
pixels). Bandwidth requirements for quality video transmission range from 256 Kbps to 1 Mbps depending on
the video compression method used, the image quality
required, and IPS used by the application.
Cellular phone is the slowest transmission method and
not widely used in the video security industry. Its bandwidth is 3000 Hz and has a 9.6 Kbps data rate. Dial-up or
PSTN with a modem is the most common transmission
method and has a maximum data rate of 56 Kbps. While
relatively slow, its low cost and availability contribute to
its continued use. ISDN is a digital phone line with two
64 Kbps channels. This costs more than the PSTN but with
competition from the cable network and digital subscriber
line (DSL), pricing for these are acceptable to the video
security industry. Typical speeds for the DSL network is
1.544 Mbps but depends on cable routing, distance, and
number of other clients using the same line. Much wider
bandwidth choices include the AT&T T1 and T3 lines,
and the OC3–OC12 optical fiber networks.
299
els) of the captured image. The larger the picture size, the
larger the storage required on the hard drive.
The abbreviation CIF has two definitions. The first is
Common Intermediate Format (CIF), a standard developed
by the International Telecommunications Union (ITU)
for video teleconferencing, and is the standard in current
use throughout the digital video security industry.
Table 9-6 defines this CIF pixel format, aspect ratio, and
bit rate for NTSC and PAL systems. The original CIF is
also known as Full CIF (FCIF). Quarter CIF is designated
as QCIF and Four CIF as 4CIF.
The 4CIF image improves the resolution by a factor of
four over the 1CIF image by doubling the number of pixels
in both the vertical and horizontal axis. 4CIF uses all the
camera pixels and reproduces the best image quality from
a high resolution camera.
The ability to identify persons, objects, and activities
greatly affects the required stored image format and consequently the resolution of the image. The 1CIF image can
be used to identify faces, license plates, and other detail
only under favorable conditions. The 4CIF display is the
format of choice and uses one of the MPEG-4, H.464, or
other high compression standards.
Also shown in Table 9-6, but not to be confused with
Common Intermediate Format is the Common Image Format
also abbreviated CIF, which is the standard frame size for
digital video based on Sony’s D1 format that defines the
two standard SDTV frames.
9.3.8 Network/DVR Security
9.3.8.1 Authentication
9.3.7 Display Format—CIF
An important requirement in any local or remote video
monitoring system is the need to keep the video information secure and error-free. DVR image degradation can
be caused by equipment failure or produced by manmade activity (hackers, viruses). The security provider
must be diligent and make all efforts to ensure that the
information is accurate and the system tamperproof. Stateof-the-art image authentication software has increased the
reliability of digital video monitoring by preventing the
tampering of the signal. The safeguards can be incorporated with either special compression methods using
date/time stamping or the summation of pixels changes,
all of which will insure the acceptance of the digital video
record in a court of law.
Some standards of authentication include:
Digital images from DVRs or other sources can be displayed on monitors in full size or a fraction of the monitor
screen size. Compression technology is critical and a significant factor in determining the storage required and
the final resolution obtained in the digital video image.
The CIF image size determines the size (number of pix-
• Images must be from the original VCR tape or DVR
hard drive
• Images should be recorded in a WORM drive
• Images should have a check sum error checking methodology
• Images should have a date digital signature.
300
CCTV Surveillance
COMMON INTERMEDIATE FORMAT*
(CIF)
CIF
QCIF (QUARTER CIF)
SQCIF (SUB QUARTER CIF)
BIT RATE AT
30 FRAMES/SEC
(Mbps)
SCREEN AREA
PIXEL FORMAT
ASPECT
RATIO
FULL
352 × 288 (PAL)
352 × 240 (NTSC)
1.222
36.5
1/4
176 × 144 (PAL)
176 × 120 (NTSC)
1.222
9.1
128 × 96 (PAL)
1.333
(4 × 3)
4.4
704 × 240 (NTSC)
704 × 288 (PAL)
1.222
18.3
—
1/2 CIF
FULL
4CIF (4 × CIF)
FULL
704 × 576 (PAL)
704 × 480 (NTSC)
1.222
146.0
16CIF (16 × CIF)
FULL
1408 × 1152 (PAL)
1408 × 960 (NTSC)
1.222
583.9
*
COMMON INTERMEDIATE FORMAT (CIF) DEVELOPED BY the INTERNATIONAL TELECOMMUNICATIONS UNION (ITU) IN
STANDARD H.261 FOR VIDEO TELECONFERENCING. THIS FORMAT IS IN CURRENT USE THROUGHOUT THE DIGITAL VIDEO
SECUIRTY INDUSTRY.
MPEG COMPRESSION STANDARDS ARE BASED ON THE CIF FORMATS.
COMMON IMAGE FORMAT **
SCREEN AREA
PIXEL FORMAT
FULL
640 × 480
1/4 VGA
1/4
320 × 240
1/16 VGA
1/16
160 × 120
D1 (SONY FORMAT)
FULL
720 × 480 (NTSC)
720 × 576 (PAL)
VGA
**
NOT TO BE CONFUSED WITH CIF ABOVE, COMMON IMAGE FORMAT IS A STANDARD
FRAME SIZE FOR DIGITAL VIDEO BASED ON SONY’S D1 FORMAT.
Table 9-6
Common Intermediate and Common Image Format (CIF) Parameters
The WORM format allows the operator to review video
images as often as required but the images can never be
altered. The check sum is a method which records the number of levels and pixels per recorded line and stores this
information in the recorder’s program. On review this
sum is checked and if the two are not equal an alarm or
visual cue notifies the operator that a change has occurred.
Authentication should also include a date/time stamping or digital signature inserted on all recorded video
images.
A network authentication protocol called Kerberos
is designed to provide strong authentication for
client/server applications by using secret-key cryptography. This protocol was developed by the Massachusetts
Institute of Technology (MIT) in the mid-1980s, and is
free and has been implemented into and available in many
commercial products. It was created by MIT as a solution
to network security problems. The Kerberos authentication system uses a series of encrypted messages to prove
to a verifier that a client is running online on behalf of
a particular user. It uses strong cryptography so that a
client can prove its identity to a server (and vice versa)
across an insecure network connection. Kerberos requires
a trusted path through which passwords are entered. If the
user enters a password in a program that has already been
modified by an attacker (Trojan horse), then an attacker
may obtain sufficient information to impersonate the user.
After a client and server have used Kerberos to prove their
identity, they can also encrypt all of their communications
to assure privacy and data integrity as they go about their
business. In 1989 Version 5 of the protocol was designed
and is in use in many systems today.
Analog, Digital Video Recorders
9.3.8.2 Watermark
A digital watermark is a digital signal or pattern inserted
into a digital image. It is inserted into each unaltered copy
of the original image. The digital watermark may also serve
as a digital signature for the copies. For law enforcement
and prosecution purposes it is critical that digital tapes
and disks be watermarked, since digital information may
easily be altered and modified through software manipulation. The law in most countries requires that information
recorded by DVRs not be altered or modified. An example
of such watermarking techniques is used in the Panasonic
digital disk recorder utilizing a proprietary algorithm to
detect if the image has been altered or modified. If the
image has been changed in any way, when it is played back
the word altered appears on the monitor indicating that
the original image is not being viewed.
9.3.8.3 Virtual Private Network (VPN)
The security of digitally transmitted information has
existed in the IT world for many years. With the
rapid increase in the use of digital video hardware and
transmission networks, the security industry looks to the
IT community for additional technologies to make video
transmission more secure and safe from external attack.
The data security requirements have changed significantly
in the past ten years as the Internet has grown, and
vastly more companies have come to rely on the Internet
for communications and hence the security solutions are
necessary.
A VPN is a private data network that makes use of
the public telecommunications infrastructure, maintaining privacy and providing security through the use of a
tunnel protocol and security procedures.
The VPN provides an encrypted connection between
user’s distributed sites over a public network such as the
Internet. By contrast, a private network uses dedicated
circuits and possibly encryption. The VPN is in contrast
with the system of home or leased lines that can only be
used by one company. The primary purpose of a VPN is
to give the company the same capabilities as private leased
lines but at a much lower cost. By using the shared public
infrastructure, companies today are looking at using VPNs
for both extranets and wide area intranets.
There are three basic classifications of VPN technologies: (1) trusted VPN, (2) secure VPN, and (3) Hybrid
VPN.
9.3.8.3.1 Trusted VPNs
Before the Internet became nearly universal, a VPN consisted of one or more communication circuits leased from
a communications provider where each leased circuit
acted like a single wire that was controlled by the customer.
The basic idea was that a customer could use these leased
circuits in the same way that they use physical cables in
their local network. The privacy afforded by these legacy
301
VPNs was only that the communications provider assured
the customer that no one else would use the same circuit.
The VPN customer trusted the VPN provider to maintain
integrity of circuits and to use the best available business
practices to avoid snooping of the network traffic. This
methodology really offers no real security.
9.3.8.3.2 Secure VPNs
Networks that are constructed using encryption are called
secure VPNs. Vendors created protocols that would allow
traffic to be encrypted at the edge of one network or at
the originating computer, move over the Internet like any
other data, and then be decrypted when it reached the corporate network or a receiving computer. This encrypted
traffic acted like a tunnel between the two networks. Even
if an attacker could see the traffic it could not be read
to make a change in the traffic or make use of the data,
without the changes being seen by the receiving party who
would therefore reject the data. The encrypted tunnel provides a secure path for network applications and requires
no changes to the application.
9.3.8.3.3 Hybrid VPNs
The hybrid VPN uses a secure VPN that is run as part of
a trusted VPN, creating a third type of VPN. The secure
parts of the hybrid VPN can be controlled by the customer
or the same provider that provides the trusted part of the
hybrid VPN. Sometimes an entire hybrid VPN is secured
with the secure VPN, but more commonly only a part of
a hybrid VPN is secure.
9.3.8.4 Windows Operating System
Within a year of Windows 95 release, Microsoft identified
a major security problem that could not be fixed without a complete software rewrite. Microsoft then embarked
on the development of a completely new platform which
resulted in Windows NT that was built on the concept of
creating a high level network security OS. However, the
majority of DVR manufacturers continue to use Windows
95 and Windows 98 rather than take the costly route of
rewriting their software with the more secure Windows NT.
The newer Windows 2000, the 32-bit OS that was built
on the Windows NT technology provides the users of HD
drives many comprehensive security features that protect
sensitive video and other security data. These enhanced
security features provide local protection in addition to
securing information as it is transmitted over a LAN, WAN,
WiFi, phone line, or the Internet. With Windows 2000, the
system administrator and authorized users can select from
multiple levels of security. For advanced users, Windows
2000 also supports standard Internet security features such
as IP security, Kerberos authentication, Layer 2 Tunneling
Protocol, and VPNs. Many large companies have migrated
to Windows 2000 to take advantage of this secure OS.
302
CCTV Surveillance
9.3.9 VCR/DVR Hardware/Software Protection
Both VCRs and DVRs require various types of protection
and handling to avoid hardware and software failures.
They also require periodic maintenance. In particular,
the VCR recorder and the VHS magnetic tape cassettes
require special care because of their complex mechanical
tape handling mechanism and the vulnerable videotape
cassette.
9.3.9.1 Uninterrupted Power Supply (UPS)
Hardware main power protection via power conditioning is a must for VCRs and DVRs. As with computer
systems, voltage surge and power line filtration must be
included in any recorder installation. Installations in areas
prone to lightning or other electrical disturbances require
extra precautions. Power protection must not be treated
as “just another box” that must be included with the
system. Unfortunately, it is usually after a major failure
that most people realize that this protection is critical.
Appropriate protection includes a UPS and surge protector (Chapter 23).
9.3.9.2 Grounding
Like other electronic equipment, DVRs require proper
electrical grounding to insure that transient voltages on
the power line are safely directed to ground. This ground
connection will greatly reduce the possibility of damage to
the recorder and its internal HD drive. Such grounding
is important in high-risk areas that experience lightning
storms and applications where electromagnetic interference (EMI) or radio frequency interference (RFI) may be
present. The grounding wire on a three-pronged power
cord is sufficient for grounding the recorder, but a check
should be made that the AC power socket into which
it is plugged is connected to earth ground. This can be
tested by using an ohmmeter and measuring the resistance
between this location and the earth ground location. It
should measure near zero ohms.
9.3.9.3 Analog/Digital Hardware Precautions
Both the analog and digital video recorders contain
mechanical moving parts. As such they should be treated
with care when installing, moving, or relocating them. Do
not rough handle them.
The VCRs and DATs have many mechanical parts which
can become misaligned or damaged if the machine is
dropped or mishandled, or if tape insertion/removal
is performed carelessly. This can render the machine
inoperative.
Digital video recorders have one or more HD for storage
of the video image. Do not unnecessarily jar or drop the
machine as this could damage or reduce the lifetime of
the HD. After powering down the DVR, it should not be
moved for a few minutes after power shutoff to insure that
the HD platter has come to a complete stop and that the
HD head that reads and writes information from the disk
has come to a parked position.
9.3.9.4 Maintenance
Since the VCR, VCR time lapse, DAT, and DVR recorders
are used over a long periods of time, preventative maintenance for these devices is important. To ensure reliable
operation this is especially true for the VCR video heads.
The video heads rotate at 1800 rpm and the video head
gradually wears out and head-to-tape contact is reduced,
resulting in a noisy picture. VCRs must be operated in a
dust-free, controlled humidity and temperature environment to ensure reliable operation. If the VCR tape fails or
the cassette jams, retrieve the cassette and carefully manually remove the broken tape, and then splice the tape to
salvage the remaining information recorded.
In the case of the DVR, the PC-based operating system
(OS) may crash and therefore it is wise to back up the
recorded information onto external backup storage or use
a RAID-configured HD drive system. Short of these measures, it is a challenge to retrieve the information from the
HD. If DAT recorders are used for backup, head clogging
is often difficult to detect because of the powerful error
correcting built into these machines. They operate even
with only one head operating. To test for head clogging,
turnoff the error correction and read the error rate of
the unit and see whether it is within the manufacturer’s
specifications. If not, clean or replace the heads.
9.4 VIDEO RECORDER COMPARISON: PROS,
CONS
Although the VCR has served the video surveillance industry well for several decades, the VCR technology has several
shortcomings. These have been brought into the limelight
with the introduction of the DVR in the late 1990s. The
following is a list of many pros and cons for the DVR
and VCR.
9.4.1 VCR Pros and Cons
The criteria used to assess the analog VCR and digital
DVR cross several boundaries including cost, size of system, hardware already in place, availability of recording
media, manpower to administer, and maintenance. The
analog VCR has served the security industry well over the
last decades but the digital DVR will clearly replace it
swiftly.
Analog, Digital Video Recorders
VCR Pros
• Low cost proven technology with long history of service
• Easy to copy and provide as evidence to law enforcement
• Difficult to alter video images (as compared to digital recorders).
VCR Cons
• Tape heads need regular maintenance and eventually
wear out and need replacement
• Tape handling mechanism has many precision mechanical parts that can go out of alignment or fail
• Tape needs to be changed on a regular basis daily
depending on application requiring manpower
• Tape is sensitive to humidity, dust, chemicals, and high
level magnetic fields.
9.4.2 DVR Pros and Cons
DVR Pros
• Produces a permanently clear and crisp record on a
HD drive
• Serves as long-term backup device and requires no additional data management costs
• Reproduces the original picture quality after many
copies are made
• Provides multi-channel recording in real-time when
required
• Simultaneous recording, viewing, and transmitting to
remote site
• Intelligent motion detection acts as alarm sensor
• Multiplexer can be integrated with DVR
• Remote control Pan/tilt, camera zoom/focus/iris
• Nonstop recording limited only by the HD drive storage
space available
• Remote access by a LAN, WAN, WiFi, ISBN, DSL,
modem, PSTN.
303
guideline lists some of the factors to consider when choosing a video recording system.
9.5.1 Checklist
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
How many cameras must be recorded?
How many IPS per camera?
What quality (resolution) of images is required?
What is the size of image: 1/4 CIF, CIF, or 4 CIF?
What length of storage is required?
What is the location for monitoring?
How many sites?
Does the recording system permit limiting the amount
of bandwidth required to transmit video across the network?
Does the remote viewing software allow viewing cameras from multiple systems on the same screen at the
same time?
Does the remote viewing system software allow searching for recorded video and playing back from multiple
systems at the same time, on the same screen?
Can the system be administrated remotely?
How much training will the staff require to use it efficiently?
Can the system record at different frame rates, quality
settings, and video settings for individual cameras?
Does the system record video prior to the beginning of
the event (pre-alarm recording)?
Upon alarm condition can the system send an email
notification?
Can different recording schedules be programmed for
each hour or day?
Can the system send video to a remote location for
automatic video display upon alarm condition?
Can multi-camera views be created and then automatically sequenced between them on the video monitor?
Can the system automatically archive video data to a
network storage device?
Can pan-tilt-zoom cameras be controlled from both the
system and the remote software?
DVR Cons
• Eventual HD failure
• OS and/or application program crash
• Digital data more easily altered unless water-marking or
other high level security is built in.
9.5 CHECKLIST AND GUIDELINES
There is a large variety of VCR and DVR hardware to
choose from to record the video image. Prior to the late
1990s the VCR was the only technology choice. The DVR
is now the major technology choice. This checklist and
9.5.2 Guidelines
• Initially install DVRs in highly sensitive areas to improve
image quality, image retrieval, and searching time.
• Enable remote video monitoring for authorized personnel. This can cut travel costs, improve operational
efficiency, and make the DVR investment more costeffective.
• Choose a basic DVR or multiplexed DVR for small- to
medium-size installations.
• Choose a multi-channel DVR or NVR for large systems.
• Choose a security level that matches the security
requirement.
304
CCTV Surveillance
9.6 SUMMARY
The VCR or DVR records video images to establish an
audit trail for the video surveillance activity. It can be
viewed at a convenient time by security, law enforcement,
or corporate personnel to identify a person, determine
the activity that occurred, or assess the responses of security personnel. The video recording provides a permanent medium with which to establish credible evidence
for prosecuting a person involved in criminal activity or
suspected thereof, and for use in a criminal trial, civil litigation, or dismissal. The video recording provides a basis of
comparison with an earlier recording to establish if there
was a change in condition at a particular location, such
as moved or removed equipment or personnel patterns,
including times of arrival and departure.
Video cassette recorders and digital video recorders
are excellent tools for training and evaluation of personnel performance. They serve as a source of feedback
when evaluating employee performance. By reviewing the
recording, management can determine which employees
are working efficiently and which employees are not performing up to standards, without on-site supervision.
Magnetic HD DVRs and optical HD recorders have a
clear advantage over VCRs when video images must be
retrieved quickly from a large database of stored images.
Retrieved video images can be printed on thermal, inkjet,
or laser printers when: (1) a hard copy audit trail of video
images are required for dismissal, court-room, or insurance purposes, (2) a guard needs a hard copy printout
when dispatched to apprehend an individual at a suspected crime scene, (3) to produce a permanent hard
copy of an activity or accident for insurance purposes,
etc. The video record offers the ability to instantly replay
a video image and print it. This feature is important in
real-time pursuit and apprehension scenarios.
The Internet has and will continue to change how video
images will be recorded and distributed locally and to
remote sites. The hardware, software, and transmission
channels already exist to provide security personnel, corporate management, and government organizations to
perform automated video security (AVS).
Chapter 10
Hard Copy Video Printers
CONTENTS
10.1
10.2
10.3
Overview
Background
Printer Technology
10.3.1 Thermal
10.3.1.1 Monochrome
10.3.1.2 Wax Transfer
10.3.1.3 Color-Dye Diffusion
10.3.2 Ink Jet, Bubble Jet
10.3.3 Laser, LED
10.3.4 Dye-Diffusion, Wax-Transfer
10.3.5 Film
10.4 Printer Comparison
10.4.1 Resolution and Speed
10.4.2 Hardware, Ink Cartridge Cost Factors
10.4.3 Paper Types
10.5 Summary
10.1 OVERVIEW
Hard-copy printout from a live video monitor and
VCR/DVR recorder or other transmitted surveillance
images is a necessity to the video security system.
Monochrome and color printers permit good to excellentquality reproduction of the scene image on hard-copy
printout. The printed hard-copy image is used by security personnel for apprehending offenders, responding to
security violations and for a permanent record of a scene,
activity, object, or person.
The video printer is a device that accepts: (1) an analog video signal from a camera or VCR or (2) a digital
signal from a computer, a DVR, or an IP camera, and
transfers the information to paper (or film). The information can be text, graphics, and video, and can be printed
in either color or monochrome depending on the data
content. Printers vary greatly in terms of their technology,
sophistication, speed, and cost.
10.2 BACKGROUND
The three most popular video printer technologies for
video applications are thermal, ink jet, and laser.
Thermal. Early models of thermal hard-copy printers
produced crude facsimiles of the monitor picture with
low resolution and poor gray-scale rendition. Today’s
advanced technology enables printers to produce excellent monochrome or color image prints with resolution
approaching that of a high-quality camera. Of the several
monochrome and color printout technologies available
for the security industry, the monochrome thermal printer
is the most popular because of its low hardware and paper
costs. They need no toner or ink, only a special paper.
Ink Jet, Bubble Jet. The present ink-jet printer was built
on the progress made by many earlier versions and has had
a long history of development. Among the contributors
to the evolution have been the Hewlett Packard (HP)
and Canon Companies, claiming a substantial share of
credit for the development of the modern ink jet. In 1979
Canon invented and developed the drop-on-demand inkjet method where ink drops are ejected from a nozzle by
the fast growth of an ink vapor bubble on the top surface
of a small heater. Canon named this bubble jet technology.
In 1984 Hewlett-Packard (HP) commercialized the ink-jet
printer and it was the first low-cost ink-jet printer based
on the bubble jet principle. HP named the technology
thermal inkjet. Since then, HP and Canon have continuously
improved on the technology and currently thermal ink-jet
printer dominates the major segment of the color printer
market. The four major manufacturers now accounting
for the majority of ink-jet printer sales are Canon, HP,
305
306
CCTV Surveillance
Epson, and Lexmark. Ink-jet printers are a common type
of computer printer used for video security applications.
Laser, LED. Laser and LED (light emitting diode) printers provide an alternative to the ink-jet and bubble jet
printers for producing hard-copy video printouts. Laser
and LED printers rely on technology similar to a type
of dry process photocopier that was first introduced by
Xerox Corp. This process known as electro-photography
was invented in 1938 and later developed for their copier
machines by Xerox and Canon in the late 1980s.
The first laser printer was created by Xerox researcher
Gary Starkweather by modifying a Xerox copier in 1971
and was offered as a product as the Xerox Star 8010 in
1977. The first successful laser printer was the HP LaserJet, an 8-page per minute (ppm) model, released in 1984.
The 8010 used a Canon printing engine controlled by
HP-developed software. The laser printer uses a rotating
mirror to form the image on the drum. The HP LaserJet
printer was quickly followed by laser printers from Brothers Industries, IBM, and others.
The Okidata Company developed and has been producing a printer using LED technology instead of a laser
for many years. Okidata and Panasonic now produce LED
printers using an array of small LEDs to form the latent
image on the drum, and no mirror scanner is required.
This LED technology offers some potential advantages
over the laser system.
Other. Two other technologies used to produce highquality color images are: (1) thermal transfer printer
(TTP) using thermal plastic wax and (2) thermal sublimation printer (TSP) using dye diffusion. Both techniques produce brilliant colors and excellent resolution.
The printer cost is high and the ink cartridges expensive,
and therefore these printers are not in high use in security applications. Color laser printers are not used in the
video security industry because of their high equipment
cost and ink cartridge replacement cost, as compared to
other technologies now available.
In addition to the standard monochrome laser printers
that use a single toner, there also exist color laser printers
that use four toners to print in full color. Color laser
printers tend to be about five to ten times as expensive as
monochrome.
Polaroid film technology has been used in the video
industry for many years and is still used for special applications. It has lost its popularity because of the new
thermal, ink-jet, laser, and LED technologies that have
become available.
Dot matrix printers are not suitable for monochrome
or color video image printing because of their lower resolution and slower speed and high noise levels.
10.3 PRINTER TECHNOLOGY
Most video image printouts are still done with
monochrome thermal printers. The reason for this is the
significantly lower cost of the printer hardware and the
lower cost of the hard-copy printout, since no ink cartridge
head or ink is required for the monochrome video printer.
However, the overwhelming use of color video cameras
in security monitoring systems has motivated manufacturers to provide cost-effective solutions for printing color
images. In a color video system, the lens receives the
color picture information and through the color camera
converts the light image into three electrical signals corresponding to the red, green, and blue (R, G, B) color
components in the scene. These three signals presented
to an RGB monitor produce a color image on the monitor. In a color printer, the three primary colors in the
video signal, R, G, and B, must be reversed to obtain their
complementary colors: cyan, magenta, and yellow.
10.3.1 Thermal
Three thermal technologies for producing hard-copy
printout are: (1) monochrome, (2) wax transfer, and
(3) color-dye diffusion.
10.3.1.1 Monochrome
The monochrome thermal video printer is the most popular type used in security industry. The primary reason is
that it can produce resolution comparable to the resolution of the cameras and sufficient printing speed required
for video security applications. Another reason for their
popularity is that the cost for the hardware, printout paper,
and printer head are less than those of other printer technologies. Figure 10-1 shows a monochrome thermal video
printer and hard-copy printout.
Thermal monochrome printers create an image by
selectively heating coated paper as the paper passes over
the thermal printer head (Figure 10-2). The coating turns
FIGURE 10-1
Thermal video printer
Hard Copy Video Printers
307
THERMAL PRINTING STEPS
VIDEO SIGNAL
THERMAL
HEAD
CONVERSION ELECTRONICS
VIDEO
PRINT
RANDOM ACCESS MEMORY
(FREEZE FRAME)
THERMAL HEAD
SUPPLY
ROLL
PLATEN
THERMAL
PRINT PAPER
PLASTIC WAX COATED
INK PAPER
PRINT PAPER
FIGURE 10-2
Thermal printer block diagram
black in the areas where it is heated, creating the image.
Care must be taken with the handling and storage of the
thermal paper, as it suffers from sensitivity to heat and
abrasion which can cause darkening of the paper or fading
due to light.
The thermal printer converts the video signal from the
camera into a digital signal and stores it in a random
access memory (RAM) or other storage device. The video
freeze-frame module captures and “freezes” the image as a
snapshot of a moving video scene. This temporary storage
allows the printer to operate at a much slower speed than
the actual real-time video frame rate. After the video image
has been captured, it is converted to an electrical drive
signal for the thermal head located adjacent to the paper.
Depending on the video drive signal level, the paper is
locally heated, causing the wax on the paper to melt and
turn black (or another color). Depending on the amount
of heat applied, a larger or smaller dot is produced providing a gray-scale level to the image. As the video information is scanned across the slowly moving paper, the image
is “burned in,” thereby creating a facsimile of the video
image. Scanning an entire monochrome video image one
pixel at a time takes approximately 8 seconds. Since the
video image is stored in the printer until a new frame is
captured, multiple copies can be made. The printed video
image is recorded on a treated paper that resists fading
from sunlight and physical tearing. Figure 10-3 shows a
monochrome thermal video printer hard-copy printout.
10.3.1.2 Wax Transfer
In the color TTP, a plastic-wax, single-color-coated ribbon
(the width of the paper roll) is inserted between the thermal print head and the paper (Figure 10-4). The ribbon
is heated locally from behind causing the wax-based ink
coating to melt, and the image to transfer to the paper.
The full-color prints are produced in the thermal plastic
color printer through the multiple passes of three ribbons
having the colors cyan, magenta, and yellow. The inking
paper used is divided into three sections with differentcolored ink; these three sections pass the thermal printer
platen in sequence. As each color passes over the thermal
head, an electrical signal proportional to the amount of
the respective color in the video signal heats the head
so that the ink of the required color is deposited on the
paper. Depending on the amount of heat applied, a larger
or smaller amount of ink from the paper will be transferred from the base film to the print paper. The first time
the paper passes the head, yellow is deposited on it, then
magenta, then cyan. By printing these three colors, so that
they are superimposed exactly on each other, the printer
is able to produce a high-resolution print with excellent
308
CCTV Surveillance
(A) SURVEILLANCE
FIGURE 10-3
(B) FACIAL IDENTIFICATION
Thermal printer quality: surveillance, facial identification
PRINCIPLES OF COLOR PRINTING
THERMAL
HEAD
LENS
COLOR
CCTV
SCENE
RED
PRINT
PAPER
YELLOW
CAMERA
CCD SENSOR
GREEN
ELECTRONICS
BLUE
MAGENTA
CYAN
VIDEO SIGNAL:
COLOR ADDITION
PRINTING SIGNAL—INK
COLOR SUBTRACTION
BASE FILM
INK LAYER
THERMAL
HEAD
INK
PAPER
TRANSFERRED
INK
SUPPLY
ROLL
THERMAL
HEAD
PRINT PAPER
PLATTEN
CYAN
MAGENTA
YELLOW
C
COLOR
SHEET
M
Y
C
PRINT
PAPER
PLATTEN
(ROLLER)
FIGURE 10-4
TAKEUP
ROLL
Plastic wax thermal transfer printer (TTP) block diagram
REQUIRES 3 PASSES
OF INKED RIBBON
INK: YELLOW, MAGENTA, CYAN
Hard Copy Video Printers
color rendition. By this principle, each dot on the final
print copy is transferred from the base film ink layer to
the print paper. Reproducing a satisfactory color image
requires precisely engineered mechanical components so
that the absolute registration between the three colors is
printed. It also requires precise electronic technology to
accurately combine the timing, signal, and video fidelity
to ensure a faithful video image.
10.3.1.3 Color-Dye Diffusion
The TSP dye-diffusion printing media uses three ink dye
papers (Figure 10-5). The TSP printer operates through
the use of a polyester-based substrate (donor element)
that contains a dye and binder layer, which when heated
from the back side of the polyester sublimates (becomes
gaseous) and transfers to the paper where the dye then
diffuses into the paper itself. The ink paper consists of a
cartridge containing three-color sequential printing inks
(cyan, magenta, and yellow).
10.3.2 Ink Jet, Bubble Jet
There are two different types of ink-jet printers:
the continuous-jet printer and the drop-on-demand
BLOCKS OF
COLOR DYE
printer (Figure 10-6). The continuous-jet printer uses a
steady stream of ink droplets emanating from print nozzles
under pressure. An electric charge is selectively applied
to the droplets, causing some to be deflected toward the
print paper and others away from the paper. The printout
is the composite of all the individual dots in the image
produced in this manner.
The drop-on-demand printer is a simpler and more popular ink-jet printer. This printer forms droplets of ink in
the nozzle and ejects them through appropriate timing of
electronic signals, thereby producing the desired image
on the paper. The majority of ink jet printers produce a
single dot size for each dot. Higher resolution types use a
technology called dithering to increase the resolution and
smooth jagged edges in text and lines in graphs and video
images. Ink jet printers have found a significant market
in the surveillance field and have good resolution, color
rendition, and speed per copy.
Most current ink jets work by having a print cartridge
with a series of tiny electrically heated chambers constructed using photolithography technology. The printer
produces an image by driving a pulse of current through
the heating elements. A steam explosion in the chamber
forms a bubble which propels the droplets of ink onto
the paper. Canon named it the Bubble Jet. When the
YELLOW (Y)
COLOR TRANSFER
RIBBON
MAGENTA (M)
DYE EVAPORATES
WHEN HEATED
CYAN (C)
COLORED
IMAGE
C
HEATER ARRAY
C
M
COATED PAPER
•
•
•
•
FIGURE 10-5
309
DYES: CYAN, MAGENTA, YELLOW, BLACK (OPTIONAL)
THERMAL DYE SUBLIMATION (SOLID STATE TO GASEOUS)
REQUIRES MULTIPLE PASSES OF PAPER
256 TEMPERATURE LEVELS—NEAR CONTINUOUS TONE
Dye-diffusion thermal sublimation printer (TSP) block diagram
310
CCTV Surveillance
PIEZO-ELECTRIC TECHNOLOGY
THERMAL TECHNOLOGY
INK SUPPLY
HEATING ELEMENT
(RESISTOR)
INK CHAMBER
CAVITY
PIEZO DISK
FIRING
CHAMBER
INK
DROPLET
MOVING INKJET CARTRIDGES
(CYAN, MAGENTA, YELLOW, BLACK)
PRINTHEAD
NOZZLE
NI
EJK
ROTATING
DRUM
det
:
:
inte
TO OM ET pr
FR NKJ
I
dc
c
ypo
INK
DROPLET
T
nirp
opy
PAPER
DIRECTION
FIGURE 10-6
Ink jet, bubble jet printer technology
bubble condenses, surplus ink is sucked back up from the
printing surface. The ink’s surface tension pumps another
charge of ink into the chamber through a narrow channel
attached to an ink reservoir. Epson’s micro-piezo technology uses a piezo-crystal in each nozzle instead of a heating
element. When current is applied, the crystal bends, forcing a droplet of ink from the nozzle.
The greatest advantages of ink jet printers are quiet
operation, capability to produce color images with near
photographic quality, and low printer prices. One downside of the ink jet printer is that although they are generally cheaper to buy than the lasers, they are far more
expensive to operate when it comes to comparing the cost
per page. Ink cartridges used in ink jet printers make them
many times more expensive than laser printers to produce
the print.
10.3.3 Laser, LED
Laser printers provide an alternative to the ink jet and bubble jet printers for producing hard-copy video printouts.
The laser printer can produce high-quality monochrome
images with excellent resolution—300 dots per inch (dpi)
and grayscale (halftone) rendition.
Laser and LED printers rely on one and the same
technology used in the first photocopying machines. This
process is known as electro-photography and was invented
in 1938 and later developed by Xerox and Canon in the
late 1980s. The electro-photographic process used in laser
printers involves six basic steps:
1. A photosensitive surface (photo-conductor) is uniformly charged with static electricity by a corona discharge.
2. Then the charged photo-conductor is exposed to an
optical image through light to discharge it selectively
and form a latent, invisible image.
3. The latent image development is done by spreading
toner, a fine powder, over the surface which adheres
only to the charged areas, thereby making the latent
image visible.
4. At the next step an electrostatic field transfers the developed image from the photosensitive surface to the sheet
of paper.
5. Then the transferred image is fixed permanently to the
paper by fusing the toner with pressure and heat.
Hard Copy Video Printers
6. The final step in the process occurs when all excess
toner and electrostatic charges are removed from the
photoconductor to make it ready for the next printing cycle.
In operation the laser printer uses a laser beam to produce an image on a drum (Figure 10-7). Because an entire
page is transmitted to a drum before the toner is applied,
laser printers are sometimes called page printers. Figure 10-8
shows the schematic diagram of the laser and page printer
and Figure 10-9 the rotating mirror scanning mechanism.
Laser printing is accomplished by first projecting an
electric charge onto a revolving drum by a primary electrically charged roller. The drum has a surface of a special
plastic or garnet. Electronics drives a system that writes
light onto the drum. The light causes the electrostatic
charge to leak from the exposed parts of the drum. The
light from the laser alters the electrical charge on the drum
wherever it strikes. The surface of the drum then passes
through a bath of very fine particles of dry plastic powder
or toner. The charged parts of the drum electrostatically
attract the particles of powder. The drum then deposits
the powder onto the sheet of paper. The paper passes
through a fuser, which with heat and pressure bonds the
plastic powder to the paper.
Each of these steps has numerous technical choices.
One of the more interesting choices is that some “laser”
printers actually use a linear array of LEDs to write the
light on the drum instead of using a laser. The toner is
essentially ink and also includes either wax or plastic. The
chemical composition of the toner is plastic-based or waxbased so that when the paper passes through the fuser
assembly the particles of toner will melt. The fuser can
be an infrared oven, a heated roller, or in some very fast
expensive printers, a xenon strobe light.
The laser printer relies on the laser beam and scanner
assembly to form a latent image on the photo conductor
bit by bit. The scanning process is similar to electron-beam
scanning used in a CRT monitor. The laser beam modulated by electrical signals from the printer’s controller is
directed through a collimating lens onto the rotating polygon mirror that reflects the laser beam onto the drum.
The laser beam then passes through a scanning lens system
which makes some corrections to it and scans the beam
onto the photo-conductor on the drum. This complex
technology is the major key for insuring high precision in
the laser spot at the focal plane. Accurate dot generation
at a uniform pitch (spacing) ensures the best printer resolution. Figure 10-9 shows the light path through the laser
printer from the laser source to the photoconductor on
the drum.
BASIC ELECTRO-STATIC COPYING PROCESS
1
CHARGING
IMAGE EXPOSURE:
LASER
2
6
PHOTO
SENSITIVE
DRUM
DRUM CLEANING
5
3
DEVELOPING
FIXING
4
TRANSFER
NEGATIVE CHARGE
POSITIVE TONER
LIGHT SOURCE:
LASER
LASER SCANNING
OPTICS
ORIGINAL DOCUMENT
IMAGE EXPOSURE
PRIMARY
CHARGE
CLEANING
DEVELOPER
ROTATING
DRUM
TONER
FIXING
PAPER
PAPER
TRANSFER/SEPARATE
FIGURE 10-7
Laser page printer schematic diagram
311
312
CCTV Surveillance
ROTATING
POLYGON
MIRROR
FOCUSING
LENSES
SCANNING
LASER BEAM
SOLID STATE
LASER LIGHT
SOURCE
MIRROR
ESA
irp R
n
t
c de
ypo
L
ROTATING
DRUM
ed
:
rint
OM
Rp
SE
LA
FR
TO
:
TONER
cop
y
PAPER
DIRECTION
FIGURE 10-8
Laser printer rotating mirror scanning mechanism
A second type of page printer falls under the category
of laser printers even though it does not use lasers at all. It
uses the radiation from a linear array of LEDs to expose
the image onto the drum (Figure 10-10). Once the drum
is charged, however, the LED printer operates like the
laser printer.
The LED printers developed by Okidata and Panasonic
use an array of small LEDs instead of using a laser to form
the latent image on the drum. In this technology a light
source controlled by the printer’s CPU illuminates a lightsensitive drum creating an attractive charge on the drum.
No mirror scanner is required using this LED technology.
The drum rotates past a toner attracting the toner particles where the drum has been illuminated. The drum
rotates the paper to the toner, is transferred, making the
image that is fused onto the paper (Figure 10-11).
The LED array consists of thousands of individual digital
LED light sources, spanning the width of the image drum
directing light through focusing lenses directly onto the
drum surface. This methodology can have an advantage
over the laser light source system. In the case of the laser,
a single light source and a complex system of fixed lenses
and mirrors and a rotating mirror deflects the laser beam
across the drum as it rotates. Complex timing is used to
ensure that the laser produces a linear horizontal track
across the drum surface. Careful parallax correction must
be employed since the edges of the drum are farther from
the laser than the center of the drum. The LED array
technology eliminates any possibility of parallax errors or
timing errors since they are arranged across the entire
page width and are fixed.
The resolution obtained with the laser and solid state
LED implementations result in approximately the same
resolution although the LED seems to have a slight edge.
Laser heads can produce dot sizes of 60 micrometers (m)
whereas LED technology can produce dot sizes as small as
34 m. Inherently the LED light source should be more
reliable than the laser system since it has no moving parts.
These LED machines are guaranteed for five full years.
The LED design inherently has a higher speed than
the laser design since it has no moving parts. There is
a limit to how fast the drum in the laser system can be
rotated and still maintain horizontal scanning integrity. In
the LED technology there is no scanning or moving parts
and therefore it can print faster at higher resolutions than
the laser design. As shown in Figure 10-12 the resolution
of the LED design at 600 or 1200 dpi remains constant
independent of the page print speed whereas in the case
of the laser design the resolution drops when the print
speed is increased. Another advantage of the LED design
over the laser is that the LED has a straight line paper path
that is less susceptible to jams.
One of the chief attributes of these laser printers is resolution. Laser printers can print between 300 and 1200 dpi.
Laser printers produce very high-quality print and are
capable of printing an almost unlimited variety of fonts.
Most laser printers come with a basic set of fonts called
Hard Copy Video Printers
313
PHOTOCONDUCTOR SURFACE
ROTATING DRUM
SCANNING LASER
BEAM
FIXED SCANNING
AND FOCUSING LENS
MIRROR
ROTATING
POLYGON
MIRROR
COLLIMATING
OBJECTIVE
LENS
SOLID STATE
LASER LIGHT SOURCE
FIGURE 10-9
Laser printer light path from laser source to drum photoconductor
internal or resident fonts, but additional fonts can be added
in one of two ways:
1. Laser printers have slots to insert font cartridges
utilizing read-only memory (ROM). Fonts have been
pre-recorded onto these cartridges. The advantage of
font cartridges is that none of the printer’s memory
is used.
2. All laser printers come with a certain amount of RAM
that can be expanded upon by using memory boards in
the printer’s expansion slots. Fonts can then be copied
from a disk to the printer’s RAM. This is called downloading fonts, and these fonts are often referred to as
soft font s, to distinguish them from the hard fonts available on font cartridges. The more RAM a printer has,
the more fonts that can be downloaded at one time.
Laser printers can print text, graphics, and video images.
Significant amounts of memory are required in the printer
to print high-resolution graphics and images. For example,
to print a full-page graphic/image at 300 dpi requires at
least 1 MByte of printer RAM. For a 600 dpi image at least
4 MByte RAM is required.
Laser and LED printers are non-impact type and are
therefore very quiet. The speed of laser printers ranges
from about 4 to 20 text pages per minute (ppm). If a
typical rate of 6 ppm is used, this is equivalent to about
40 characters per second for text printing. Laser printers
are controlled through page description languages (PDL)
with the two de facto standards for PDLs being:
1. Printer Control Language (PCL) developed by HP
2. PostScript developed by Apple Computer for the
Macintosh computer.
PostScript has become the de facto standard for Apple
Macintosh printers and for most desktop publishing systems. Most software can print using either of these PDLs.
PostScript has some features that PCL lacks. Some printers
support both PCL and PostScript.
For video applications, in particular, there is an
increased demand for print quality (image resolution,
sharpness, and color rendition), and printer manufacturers have devoted considerable amounts of time and
money on technology advancements. In particular, they
have focused on those that eliminate smear, steps, or
314
CCTV Surveillance
GALLIUM ARSENIDE (GaAs)
LED ARRAY LIGHT SOURCE
FOCUSING LENS
GaAs LIGHT PULSES
ROTATING
DRUM
LE
DP
rint
ed
cop
y
TONER
co
:
ed
OM
rint
DP
LE
TO
FR
:
py
PAPER
DIRECTION
FIGURE 10-10
LED printer schematic diagram with fixed LED page illumination
BASIC ELECTROSTATIC COPYING PROCESS
1
CHARGING
2. IMAGE EXPOSURE
FIXED LINEAR ARRAY
6
DRUM CLEANING
5
PHOTO
SENSITIVE
DRUM
3
DEVELOPING
FIXING
TRANSFER
NEGATIVE CHARGE
POSITIVE TONER
LIGHT SOURCE:
FIXED LED ARRAY
ORIGINAL DOCUMENT
FOCUSING OPTICS
IMAGE EXPOSURE
PRIMARY
CHARGE
DEVELOPER
CLEANING
ROTATING
DRUM
TONER
FIXING
PAPER
PAPER
TRANSFER/SEPARATE
FIGURE 10-11
Light emitting diode (LED) printer block diagram
Hard Copy Video Printers
315
RESOLUTION
(dpi)
10000
LA
SE
R
LED
1200 dpi
1000
LED
600 dpi
100
10
20
30
50
100
PRINT SPEED
(ppm)
NOTE: LED RESOLUTION REMAINS CONSTANT OVER RANGE OF PRINT SPEED
LASER RESOLUTION DECREASES AS PRINT SPEED INCREASES
FIGURE 10-12
Resolution vs. printing speed for the LED and laser printers
other jagged edges on straight lines in the video image
or graphics. The laser and ink jet technologies both place
dots of ink on the paper. In order to smooth out these
dots along edges of text, graphics and images, they have
implemented technologies to change the size and placement of the dots to fill in and smooth out the boundaries of letters, and straight lines and curves in the images
(Figure 10-13).
In one technology, as many as four different-sized dots
are produced and grouped in various combinations along
the edges of boundaries to smooth out these images. The
result is a crisper better-looking image with sharper edges,
smoother curves and none of the jagged edges. The technology changes the size and placement of the dots to fill-in
and smooth-out the boundaries.
Both laser and LED printers offer an excellent solution
for video image printing to produce high-quality images
at high speeds. Table 10.1 compares the laser printer and
LED printer specifications.
images. Typical systems have a resolution of 500–600 pixels, have 64 levels of gray scale, and require 60–80 seconds
to print out. These printers carry a very high price tag and
are normally used for printing still images and therefore
have not found their way into the surveillance field.
Thermal wax transfer monochrome and color printers
function by adhering a waxed-based ink onto the paper.
As the paper and ribbons travel in unison beneath the
thermal printer head, the wax-based ink from the transfer ribbon melts onto the paper. When cool, the wax is
permanent. This type of thermal printer uses a full size
panel of ribbon for each page to be printed regardless
of the contents of the page. Monochrome printers have a
black panel for each page to be printed, while color printers have three (CMY) or four (CMYK) colored panels for
each page. Unlike dye sublimation printers these printers
cannot vary the dot intensity, which means that the image
must be dithered. These printers are not in widespread
use in video security applications.
10.3.4 Dye-Diffusion, Wax-Transfer
10.3.5 Film
The high-resolution thermal laser printer uses an entirely
different and more complex principle to produce
extremely high resolution continuous-tone laser-printed
Hard-copy video images can be printed on black-andwhite or color photographic film such as the instant prints
developed by Polaroid Corp. The image is first captured
316
CCTV Surveillance
THE SMALLER DOTS AND DITHERING
ALLOW THE DOTS TO FILL IN AND
SMOOTH OUT THE BOUNDARIES OF
LETTERS, GRAPHICS AND PHOTOS
AREA OF ENLARGEMENT
WITH SMOOTHING
TECHNOLOGY
WITHOUT SMOOTHING
TECHNOLOGY
FIGURE 10-13
LED and laser printer smoothing
PRINTER TYPE
LASER PRINTER
LED PRINTER
TECHNOLOGY
(a) ELECTROPHOTOGRAPHY
(b) SCANNING LASER BEAM
(c) INK TONER IN CARTRIDGE
(a) ELECTROPHOTOGRAPHY
(b) LINEAR LED ARRAY
(c) INK TONER IN CARTRIDGE
PRINT SPEED—
ppm (PAGES/MINUTE)
STANDARD: 4–50
INDUSTRIAL: UP TO 1000
10–26
RESOLUTION—
dpi (DOTS/INCH)
300–2400
MOVING PARTS
PAPER HANDLING, DRUM,
ROTATING MIRROR SCAN WHEEL
PAPER HANDLING, DRUM,
(STATIONARY LED ARRAY)
NOISE LEVEL
LOW
VERY QUIET
PRINTER COST
$200– 8000
$250– 8000
PRINT COST/PAGE
$0.03– 0.09 COLOR
$0.01– 0.03 BLACK/WHITE
$0.03– 0.09 COLOR
$0.01– 0.03 BLACK/WHITE
Table 10-1
Comparison of Laser Printer and LED Printer
300–1200, MAINTAINS
RESOLUTION AT HIGH SPEEDS
Hard Copy Video Printers
in a freeze-frame image storage device. Then the film is
exposed and developed with the Polaroid film back. While
the resolution and rendition of the image is quite good,
Polaroid film is more expensive and more difficult to work
with than thermal paper.
10.4 PRINTER COMPARISON
There are several criteria that should be considered by one
choosing a video printer. These include resolution, speed,
initial cost of equipment, cost of paper, toner or cartridge,
and of course the quality of the final printed hard copy.
10.4.1 Resolution and Speed
The thermal printer is in widespread use and can print
with a resolution of 250–500 TV lines. This printer is probably the best choice for reproducing monochrome images
with reasonable continuous-tone printing.
Monochrome thermal printers provide a fast means—
8 seconds per print—for obtaining a hard-copy printout
from any video signal. Operating these printers is relatively inexpensive.
Ink jet printers are capable of producing high-quality
print approaching that produced by laser printers. Typically models provide a resolution of 300 dpi but there are
models offering higher resolutions.
The laser printer can produce high-quality monochrome images with excellent resolution—300 dpi and
halftone (grayscale) rendition. The cost for operating
the laser printer depends on a combination of costs:
paper usage, toner replacement, drum replacement, and
other consumables such as the fuser assembly and transfer
assembly. The laser and LED printers can print from a
low resolution of 300 dpi to a high resolution of 1200 dpi.
By comparison, offset printing usually prints at 1200 or
2400 dpi. Some laser printers achieve higher resolutions
using special techniques.
Resolution for thermal dye-diffusion and wax-color video
printers is typically 500 dots horizontal, and printout time
approximately 80 seconds per print. Since each point
(pixel) in a picture or resolution element in the color
video image is composed of three separate colors, the actual
detail resolution of the image is one-third the number of
dots, or typically less than 200 TV-line resolution for the
printed color image. While this is significantly less than the
500 or 600 TV-line resolution in the monochrome image,
the addition of color to the print adds useful information.
The print paper roll produces 3- by 4-inch pictures.
10.4.2 Hardware, Ink Cartridge Cost Factors
The thermal printer enjoys popular demand for printing monochrome and color video images because
317
of its ruggedness, convenience, and reasonable price.
Monochrome thermal printers cost from $1100 to $1600.
The typical video thermal printer (Figure 10-1) holds a roll
of plastic wax-coated thermal paper sufficient to produce
one-hundred-and-twenty 3 × 5-inch video pictures.
There are two main design philosophies in ink jet head
design. Each has strengths and weaknesses. Fixed head
philosophy uses a built-in print head that is designed to
last for the entire life of the printer. Consumable ink cartridge costs are typically lower in this design. If, however,
the head is damaged it is usually necessary to replace the
entire printer. Epson has traditionally used fixed print
heads. In fact, disposable heads have proven to be equally
good and are used in the HP and other popular manufacturers’ machines.
The disposable head philosophy uses a print head that
is part of the replaceable ink head cartridge. Every time
the printer runs out of ink the entire cartridge is replaced
with a new one. This adds substantially to the cost of consumables but it also means that a damaged or empty print
head is only a minor problem and the user can simply
buy a new cartridge. HP has traditionally favored the disposable print head as did the Canon in its early models.
Canon now uses replaceable print heads in most models
that are designed to last the life of the printer, but can
be replaced at anytime by the user if they should become
clogged or inoperative for some reason. The ink tanks are
separate fo