Guide to GIS and Image Processing Volume 1 Release 2

Release 2
Guide to GIS and Image Processing
Volume 1
May 2001
J. Ronald Eastman
Clark Labs
Clark University
950 Main Street
Worcester, MA
01610-1477 USA
tel: +1-508-793-7526
fax: +1-508-793-8842
email: idrisi@clarku.edu
web: http://www.clarklabs.org
Idrisi Source Code
©1987-2001
J. Ronald Eastman
Idrisi Production
©1987-2001
Clark University
Manual Version 32.20
Table of Contents
Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .i
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Idrisi32.2 Release Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Exploring IDRISI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Keeping in Touch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Introduction to GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Components of a GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Spatial and Attribute Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Cartographic Display System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Map Digitizing System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Database Management System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Geographic Analysis System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Image Processing System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Statistical Analysis System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Decision Support System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Map Data Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Raster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Raster versus Vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Geographic Database Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Organization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Georeferencing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Analysis in GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Analytical Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Database Query . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Map Algebra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Distance Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Context Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Analytical Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Database Query . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Derivative Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Process Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
The Philosophy of GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Introduction to Remote Sensing and Image Processing . . . . . . . . . . . . . . . . . . . . . . . 17
Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Fundamental Considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Table of Contents
i
Energy Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Wavelength. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interaction Mechanisms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Spectral Response Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multispectral Remote Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hyperspectral Remote Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sensor/Platform Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Aerial Photography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Large Format Photography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Small Format Photography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Color Photography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Aerial Videography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Satellite-Based Scanning Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
LANDSAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SPOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IRS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
NOAA-AVHRR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RADARSAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
JERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AVIRIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
MODIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digital Image Processing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Restoration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Radiometric Restoration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Geometric Restoration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contrast Stretch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Composite Generation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digital Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Supervised Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Unsupervised Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Accuracy Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Vegetation Indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Principal Components Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Other Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
18
19
19
21
22
22
23
23
23
23
24
24
25
25
25
26
26
26
26
27
27
27
27
27
27
28
28
29
29
29
30
30
31
32
32
32
33
33
34
Idrisi System Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
System Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The IDRISI Application Window. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Menu System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Tool Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Status Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Program Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Paths and Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IDRISI Guide to GIS and Image Processing Volume 2
35
35
35
35
36
36
37
ii
The Project Working Folder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Project Resource Folders. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Setting the Data Paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Working with IDRISI Dialog Boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pick Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Output File Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overwrite Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Getting Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
37
37
37
37
38
38
39
40
Map Layers, Collections and Data Structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Map Layers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Layer Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Layer Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Layer Data Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Layer Collections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Group Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Collection Member Naming Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Collections and Pick Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Attribute Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Values Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Import/Export Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Layer File Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Raster Layers (.rst) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Raster Documentation Files (.rdc) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Vector Layers (.vct) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Common Features of All Vector Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Point Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Line Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Polygon Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Text Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Vector Documentation Files (.vdc). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Attribute Files (.mdb and .avl). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Tables (.mdb). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Values Files (.avl) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Attribute Documentation Files (.adc) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Other File Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Composition Files (.map) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Symbol and Palette Files (.sm0, .sm1, .sm2, .smt, .smp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reference System Parameter Files (.ref) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
41
41
41
42
42
42
43
43
43
44
44
44
44
44
44
47
49
50
50
50
50
51
51
52
52
52
53
55
55
55
55
Display System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Display Launcher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Palette and Symbol Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Autoscaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Table of Contents
57
57
57
57
iii
Automatic Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Launching Maps and Layers from IDRISI File Explorer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Windows and Layer Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Composer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Add Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Remove Layer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Layer Names, Types, and Visibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Layer Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Changing the Display Min / Display Max Saturation Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Legends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Georeferencing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Map Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
North Arrow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Scale Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Text Inset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Graphic Insets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Titles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Placemarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Feature Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Save Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Print Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Symbol Workshop
...........................................................................
Media Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interactive Display Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Move and Resize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cursor Inquiry Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Feature Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pan and Zoom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Collection Linked Zoom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Placemarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
GPS Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Saving Routes and Waypoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
How It Works. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interactive Screen Digitizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Layer Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Automatic Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ID or Value / Index of First Feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Mouse Button Functions When Digitizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Deleting Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digitizing Additional Features Within a Single File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Saving The Vector Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digitizing Multiple Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A Special Note About Digitizing Point Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A Special Note About Digitizing Polygon Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A Final Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IDRISI Guide to GIS and Image Processing Volume 2
58
58
58
59
60
60
60
61
61
62
63
63
63
63
63
64
64
64
64
64
64
64
65
65
65
66
67
67
67
67
68
68
69
69
69
70
70
71
71
71
71
71
71
72
72
72
72
72
72
iv
Database Workshop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Working with Linked-Table Collections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Launching Database Workshop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Displaying Layers from Database Workshop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Database Query using an SQL Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Mapping the Filtered Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Toggling the Filter on the Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Removing the Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Query by Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Other Database Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Calculating Field Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Finding Specific Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sorting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Entering or Modifying Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Modifying the Table Structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Assigning Data To and Extracting Data From Raster Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Export and Import. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
73
73
73
73
74
74
74
74
75
75
75
75
75
75
76
76
Idrisi Modules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
The File Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
General Conversion Tools Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Government/Data Provider Formats Submenu (Import Only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Desktop Publishing Formats Submenus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Software-Specific Formats Submenus (Import and Export) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
The Display Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
The GIS Analysis Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
The Database Query Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Performing Database Query in IDRISI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
The Mathematical Operators Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
The Distance Operators Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
The Context Operators Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
The Statistics Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
The Decision Support Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
The Change / Time Series Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
The Surface Analysis Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Interpolation Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Geostatistics Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Topographic Variables Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Feature Extraction Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
The Image Processing Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Restoration Submenu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Enhancement Submenu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Transformation Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Fourier Analysis Submenu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Signature Development Submenu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Hard Classifiers Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Soft Classifiers / Mixture Analysis Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Table of Contents
v
Hardeners Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hyperspectral Image Analysis Submenu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Accuracy Assessment Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Reformat Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Raster / Vector Conversion Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Data Entry Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Window List. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Help Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ArcView/ArcInfo Quick Start Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
102
102
103
103
104
105
106
106
106
Idrisi Modeling Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Idrisi Macro Modeler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Model Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DynaGroups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DynaLinks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Using DynaGroups, DynaLinks and Submodels Together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Calculator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Command Line Macros. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Macro File Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The IDRISI32 API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
107
107
109
109
110
110
111
112
112
Database Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Collecting Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Find Data In Digital Format and Import It . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Where To Find Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Physical Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Find Data In Hard Copy Format and Digitize It . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Collect Data Yourself In The Field Then Enter It. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Substitute An Existing Data Layer As a Surrogate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Derive New Data From Existing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Georeferencing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General Import Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tools for Import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Importing Satellite Imagery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Importing GIS Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Integration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Requirements for Data Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tools for Data Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IDRISI Guide to GIS and Image Processing Volume 2
115
115
115
115
116
116
116
117
118
118
118
118
119
119
119
119
119
120
121
121
121
122
122
vi
Georeferencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Geodesy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Geoid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reference Ellipsoids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Geodetic Datums. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Datums and Geodetic Coordinates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cartographic Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Projection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Grid Referencing Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Georeferencing in IDRISI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reference System Parameter Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Where .ref Files Are Stored . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Creating New .ref Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The .ref File Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Projection and Datum Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Algorithms Used by PROJECT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
123
123
124
124
125
125
125
126
126
128
128
128
129
131
131
131
Appendix 1: Ellipsoid Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Appendix 2: Datum Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Appendix 3: Supplied Reference System Parameter Files . . . . . . . . . . . . . . . . . . . . 145
Geodetic (Latitude/Longitude). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Universal Transverse Mercator (UTM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
US State Plane Coordinate System 1927 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
US State Plane Coordinate System 1983 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Gauss-Kruger. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
145
145
145
149
153
154
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Table of Contents
vii
IDRISI Guide to GIS and Image Processing Volume 2
viii
Introduction
IDRISI1 is a geographic information and image processing software system developed by the Clark Labs, a not-for-profit
organization within the Graduate School of Geography at Clark University. It is designed to provide professional-level
geographic research tools on a low-cost not-for-profit basis. Since its introduction in 1987, IDRISI has grown to become
the largest raster-based microcomputer GIS and image processing system on the market. It is used in over 130 countries
around the world by a wide range of research, educational, government, local planning, and resource management institutions. IDRISI is supported by a dedicated full-time staff of development, technical support and customer service professionals. The Clark Labs also develops the digitizing and vector editing package CartaLinx.
During its early development, partial support was provided by the United Nations Environment Programme Global
Resource Information Database (UNEP/GRID), the United Nations Institute for Training and Research (UNITAR),
and the United States Agency for International Development (USAID). Today, all support comes through software sales.
However, close relations are maintained with these and many other international development agencies in the attempt to
provide equitable access to geographic analysis tools.
IDRISI is the industry leader in raster analytical functionality, covering the full spectrum of GIS and remote sensing needs
from database query, to spatial modeling, to image enhancement and classification. Special facilities are included for environmental monitoring and natural resource management, including change and time series analysis, multi-criteria and
multi-objective decision support, uncertainty analysis (including Bayesian, Dempster-Shafer, and Fuzzy Set analysis), simulation modeling (including force modeling and anisotropic friction analysis) and surface interpolation and statistical characterization. Yet, despite the highly sophisticated nature of these capabilities, the system is very easy to use.
IDRISI is available in versions for MS-DOS, 16-bit Windows and 32-bit Windows.2 The 32-bit Windows version of
IDRISI offers extended analytical, display and database capabilities and is the vehicle for all new analytical developments.
This manual, the IDRISI Guide to GIS and Image Processing Volume 1, and its companion, the IDRISI Guide to
GIS and Image Processing Volume 2, describe Idrisi32, the 32-bit Windows version of IDRISI.
IDRISI consists of a main interface program (containing the menu and toolbar system) and a collection of over 150 program modules that provide facilities for the input, display and analysis of geographic data. See the IDRISI Modules
chapter for an overview of the menu structure and a listing of all modules and their capabilities. Detailed information
about each module, as well as a variety of other technical information, may be found in the on-line Help System.
Idrisi32.2 Release Notes
Many new capabilities have been added to this release of Idrisi. Most significantly, a graphic macro modeling environment
has been added and the image processing, decision support and change/time series analysis areas have been expanded.
The Idrisi Macro Modeler provides a “drag and drop” interface for building multi-step analytical models. Idrisi macro
files may be constructed in this environment and the graphic model itself may be saved. A right click on any operation
allows the user to review and edit the input parameters. In addition to providing a graphic interface to building an Idrisi
macro file, the Macro Modeler also adds two extensions to Idrisi capabilities. First, it provides for easy looping for feeding
model output back into the model as input. Second, it allows the user to enter a group of layers to be processed one-byone through a model.
1. IDRISI and CartaLinx are registered trademarks of Clark University.
2. MS-DOS and Windows are registered trademarks of Microsoft Corporation.
Chapter 1 Introduction
1
In the Image Processing area, a variety of linear spectral unmixing classifiers have been added to the already-extensive
suite of tools for subpixel classification. The ability to on-screen digitize training site polygons by “growing” the polygon
out from a selected pixel (i.e., flood polygon) has been added. Several new facilities for working with hyperspectral data
have also been added as we expect these data to become more widely used in the coming years. Multi-image mosaicking
with color balancing is also new to this version. Finally, a host of new import routines for satellite formats from LANDSAT, SPOT and other agencies is included.
In the decision support area, an automated wizard has been added to help users move through multi-criteria/multi-objective decision problems. Through the use of the wizard, users may build a model that stores all the parameters used in the
WEIGHT, MCE, RANK and MOLA steps of the process. This model may then be easily modified and re-run to produce a range of output possibilities. A click on any wizard page will reveal more information about the operation and
options presented.
Perhaps the most extensive addition of new capabilities is to the Change/Time Series Analysis area. New capabilities for
landcover change and predictive modeling based on Markov chain theory and Cellular Automata have been added. Automated procedures for Change Vector Analysis, Image Ratioing, Image Differencing and Image-to-Image gain and bias
calibration based on linear regression are also new to this version. A set of map comparison tools have been added along
with a new module for calculating Pearson’s Product Moment Correlation Coefficient between a values file and each pixel
in an image group file.
The release is rounded out with several enhancements to existing capabilities. The optional use of mask files to limit the
area of analysis has been added to several modules including HISTO and TSA. RECLASS and CONCAT can now operate on vector files. VIEWSHED now provides options for boolean or proportion view output images. The Help System
now provides a split-window interface with immediate access to the table of contents, index, and keyword search.
Finally, this release welcomes back the SHORTCUT feature of earlier versions of Idrisi. Users may choose to select modules from this alphabetical listing rather than through the menu system.
Exploring IDRISI
The best introduction to IDRISI is through the Tutorial. Parallel to working on the exercises, you should read the
remainder of the IDRISI Guide to GIS and Image Processing, which is divided into two manuals. This one, Volume
1, presents introductory and basic information about system operation. Volume 2 presents chapters on a variety of topics
of special interest in GIS and Remote Sensing.
The first three chapters of this manual present a general overview of IDRISI (this chapter), GIS, and Remote Sensing and
Image Processing.
The next several chapters explore the use of the IDRISI system. The chapter System Overview describes the nature of
the user interface. The chapter Maps, Collections, and Data Structures outlines the logic with which IDRISI organizes
data and gives an overview of the file structures of the most commonly-used data files. The Display System chapter discusses issues related to the display of geographic data and the interactive display features available for their exploration.
The Database Workshop chapter describes the database management system, giving detailed information on all its functions, including the ability to link the database to a map, and its ability to use structured query language (SQL). The
IDRISI Modules chapter gives an overview of the capabilities of the IDRISI modules and their typical usage. It also outlines the logic of the menu structure. The chapter IDRISI Modeling Tools describes the use of IDRISI's Macro Modeler, Image Calculator, macro scripting language and API (COM Server) modeling tools. The Database Development
chapter covers some of the important issues for the development and creation of GIS databases, especially techniques for
importing data to IDRISI.
The final chapter of this volume, Georeferencing, presents issues of geodesy, geodetic datums, projections and reference
systems in understandable terminology. While many project-level applications of GIS and image processing do not
IDRISI Guide to GIS and Image Processing Volume 1
2
require georeferencing to a geodetic system, integration of data with local or national government mapping will unquestionably require that the issues treated in this chapter be addressed.
This volume also contains a series of Appendices containing Georeferencing parameters, most importantly, detailed
tables of constants used for transformation between map datums (yes, in Geodesy, the plural of datum is datums, and not
data!).
The companion volume, IDRISI Guide to GIS and Image Processing Volume 2, contains information on a series of
specialized topics. The Decision Support chapter will be of particular interest to those involved with resource allocation
and planning. It covers the special procedures required to undertake multi-criteria / multi-objective analyses, as well as
decision making in the presence of uncertainty.
Several of the chapters of that volume relate to the use of remotely-sensed data and image processing techniques. The
Image Restoration chapter suggests methods for removing or diminishing the degree of random and systematic distortions that occur in imagery. A separate chapter on Fourier Analysis continues this discussion of methods for noise
removal. The Classification of Remotely Sensed Imagery chapter outlines in detail the IDRISI approach to image
classification including the use of "soft" and "fuzzy" classifiers for this process. Use of hyperspectral data is also discussed
in this chapter. The RADAR Imaging and Analysis chapter provides some suggestions for the use of radar imagery.
The chapter on Vegetation Indices describes the vegetation index models included in IDRISI for the transformation of
satellite imagery into images that indicate the relative amount of biomass present. The Time Series/Change Analysis
chapter deals with an increasingly important set of tools in environmental monitoring. Topics covered include pairwise
comparisons, procedures for distinguishing true change from natural variability, temporal profiling, and time series analysis by means of Principal Components Analysis.
Another group of chapters in this volume addresses issues of modeling continuous raster surfaces. In the Anisotropic
Cost Analysis chapter, the brief discussion of cost distance procedures in the Introduction to GIS chapter is extended
to consider the case of anisotropic forces and frictions (i.e., forces and frictions that act differently in different directions).
These tools are somewhat experimental, but offer special opportunities for the modeling of dynamic phenomena such as
groundwater flows, forest fire movements, oil spills, and so on. Three chapters focus on issues of spatial interpolation
from sample data. The Surface Interpolation chapter gives an overview of the techniques commonly encountered in
GIS and points out some of their relative advantages and disadvantages. It also indicates how these techniques are carried
out in IDRISI. The Triangulated Irregular Networks and Surface Generation chapter details the IDRISI implementation of the TIN. The chapter Geostatistics presents background information for the use of advanced geostatistical procedures such as kriging and simulation.
The Appendices section of Volume 2 contains error propagation formulae referred to in the Decision Support chapter.
The Tutorial manual is intended as a means of learning (and teaching) the IDRISI system and the basic tools used in GIS
and image processing. The exercises are in a format suitable for classroom use as well as individual instruction. Literally
thousands of users have learned the basics of GIS by means of these exercises.
In addition to the manuals described above, IDRISI also contains a very robust on-line Help System. This does not duplicate the information in the IDRISI Guide volumes, but acts as a very important supplement to it. Specifically, the Help
System contains detailed information on the use of every module in the IDRISI set. This includes information on operation, special notes, explanations of error messages, command line syntax, and so on. Every module has a help button that
can be clicked on to get help for that module. The Help System can also be accessed by clicking on the Help menu item.
You will find there a table of contents, index, and a keyword search facility. The Help System also contains a basic glossary and detailed information about IDRISI file formats.
Keeping in Touch
We hope you enjoy your use of the IDRISI system, and you will provide us with feedback on your experiences. Addresses
Chapter 1 Introduction
3
and contact information can be found in the About Idrisi item under the Idrisi Help Menu. Also visit our website at
www.clarklabs.org for free service upgrades, news, and information.
IDRISI Guide to GIS and Image Processing Volume 1
4
Introduction to GIS
A Geographic Information System (GIS) is a computer-assisted system for the acquisition, storage, analysis and display of
geographic data. Today, a variety of software tools are available to assist this activity. However, they can differ from one
another quite significantly, in part because of the way they represent and work with geographic data, but also because of
the relative emphasis they place on these various operations. In this chapter, we will explore these differences as a means
of understanding the special characteristics of the IDRISI system.
Components of a GIS
Although we think of a GIS as a single piece of software, it is typically made up of a variety of different components. Figure 2-1 gives a broad overview of the software components typically found in a GIS. Not all systems have all of these elements, but to be a true GIS, an essential group must be found.
Spatial and Attribute Database
Central to the system is the database—a collection of maps and associated information in digital form. Since the database
is concerned with earth surface features, it can be seen to be comprised of two elements—a spatial database describing
the geography (shape and position) of earth surface features, and an attribute database describing the characteristics or
qualities of these features. Thus, for example, we might have a property parcel defined in the spatial database and qualities
such as its land use, owner, property valuation, and so on, in the attribute database.
In some systems, the spatial and attribute databases are rigidly distinguished from one another, while in others they are
Images
Maps
Map
Digitizing
System
Geographic
Analysis
System
Statistical
Analysis
System
Spatial Attribute
Data Data
Base Base
Cartographic
Display
System
Figure 2-1
Chapter 2 Introduction to GIS
Statistical
Reports
Image
Processing
System
Database
Management
System
Statistics
Tabular Data
Maps
5
closely integrated into a single entity—hence the line extending only half-way through the middle circle of Figure 2-1.
IDRISI is of the type that integrates the two components into one. However, it also offers the option of keeping some
elements of the attribute database quite separate. This will be explored further below when we examine techniques for the
digital representation of map data.
Cartographic Display System
Surrounding the central database, we have a series of software components. The most basic of these is the Cartographic
Display System. The Cartographic Display System allows one to take selected elements of the database and produce map
output on the screen or some hardcopy device such as a printer or plotter. The range of cartographic production capabilities among GIS software systems is great. Most provide only very basic cartographic output, and rely upon the use of
high quality publication software systems for more sophisticated production needs such as color separation.
IDRISI allows for highly interactive and flexible on-screen cartographic composition, including the specification of multiple data layers, customization and positioning of map elements such as annotation, scale bars, insets and so forth, and
customized color and symbol sets. IDRISI map compositions may be saved for later display, printed to Windows-compatible devices, and exported in a variety of common desktop publishing formats.
Software systems that are only capable of accessing and displaying elements of the database are often referred to as Viewers or Electronic Atlases.
Map Digitizing System
After cartographic display, the next most essential element is a Map Digitizing System. With a Map Digitizing System, one
can take existing paper maps and convert them into digital form, thus further developing the database. In the most common method of digitizing, one attaches the paper map to a digitizing tablet or board, then traces the features of interest
with a stylus or puck according to the procedures required by the digitizing software. Many Map Digitizing Systems also
allow for editing of the digitized data.
The CartaLinx software package, also developed and distributed by Clark Labs, provides complete digitizing and vector
editing capability and is fully compatible with IDRISI. There are also a number of independent digitizing software packages that support the IDRISI data format.
Scanners may also be used to digitize data such as aerial photographs. The result is a graphic image, rather than the outlines of features that are created with a digitizing tablet. Scanning software typically provides users with a variety of standard graphics file formats for export. These files are then imported into the GIS. IDRISI supports import of TIF and
BMP graphics file formats.
Digitizing packages, Computer Assisted Design (CAD), and Coordinate Geometry (COGO) are examples of software
systems that provide the ability to add digitized map information to the database, in addition to providing cartographic
display capabilities.
Database Management System
The next logical component in a GIS is a Database Management System (DBMS). Traditionally, this term refers to a type
of software that is used to input, manage and analyze attribute data. It is also used in that sense here, although we need to
recognize that spatial database management is also required. Thus, a GIS typically incorporates not only a traditional
DBMS, but also a variety of utilities to manage the spatial and attribute components of the geographic data stored.
With a DBMS, it is possible to enter attribute data, such as tabular information and statistics, and subsequently extract
specialized tabulations and statistical summaries to provide new tabular reports. However, most importantly, a Database
Management System provides us with the ability to analyze attribute data. Many map analyses have no true spatial component, and for these, a DBMS will often function quite well. For example, we might use the system to find all property parcels where the head of the household is single but with one or more child dependents, and to produce a map of the result.
IDRISI Guide to GIS and Image Processing Volume 1
6
The final product (a map) is certainly spatial, but the analysis itself has no spatial qualities whatsoever. Thus, the double
arrows between the DBMS and the attribute database in Figure 2-1 signify this distinctly non-spatial form of data analysis.
In IDRISI, a DBMS is provided by Database Workshop. One can perform analyses in Database Workshop, then immediately apply the results to the proper spatial data, viewing the results as a map. In addition to Database Workshop, an
extensive set of program modules is also available for spatial and attribute data management.
Software that provides cartographic display, map digitizing, and database query capabilities are sometimes referred to as
Automated Mapping and Facilities Management (AM/FM) systems.
Geographic Analysis System
Up to this point, we have described a very powerful set of capabilities—the ability to digitize spatial data and attach
attributes to the features stored, to analyze these data based on those attributes, and to map out the result. Indeed, there
are a variety of systems on the market that have just this set of abilities, many of which will call themselves a GIS. But useful as this is, such a set of capabilities does not necessarily constitute a full GIS. The missing component is the ability to
analyze data based on truly spatial characteristics. For this we need a Geographic Analysis System.
With a Geographic Analysis System, we extend the capabilities of traditional database query to include the ability to analyze data based on their location. Perhaps the simplest example of this is to consider what happens when we are concerned with the joint occurrence of features with different geographies. For example, suppose we want to find all areas of
residential land on bedrock types associated with high levels of radon gas. This is a problem that a traditional DBMS simply cannot solve because bedrock types and landuse divisions do not share the same geography. Traditional database
query is fine as long as we are talking about attributes belonging to the same features. But when the features are different,
it cannot cope. For this we need a GIS. In fact, it is this ability to compare different features based on their common geographic occurrence that is the hallmark of GIS. This analysis is accomplished through a process called overlay, thus named
because it is identical in character to overlaying transparent maps of the two entity groups on top of one another.
Like the DBMS, the Geographic Analysis System is seen in Figure 2-1 to have a two way interaction with the database—
the process is distinctly analytical in character. Thus, while it may access data from the database, it may equally contribute
the results of that analysis as a new addition to the database. For example, we might look for the joint occurrence of lands
on steep slopes with erodable soils under agriculture and call the result a map of soil erosion risk. This risk map was not in
the original database, but was derived based on existing data and a set of specified relationships. Thus the analytical capabilities of the Geographic Analysis System and the DBMS play a vital role in extending the database through the addition
of knowledge of relationships between features.
While overlay is still the hallmark of GIS, computer-assisted geographic analysis has matured enormously over the past
several years. However, for now it is sufficient to note that it is this distinctly geographic component that gives a true GIS
its identity. In IDRISI, these abilities are extensive and form the foundation of the software system.
Image Processing System
In addition to these essential elements of a GIS—a Cartographic Display System, a Map Digitizing System, a Database
Management System and a Geographic Analysis System—some software systems also include the ability to analyze
remotely sensed images and provide specialized statistical analyses. IDRISI is of this type. Image processing software
allows one to take raw remotely sensed imagery (such as LANDSAT or SPOT satellite imagery) and convert it into interpreted map data according to various classification procedures. In recognition of its major importance as a technique for
data acquisition, IDRISI offers a broad set of tools for the computer-assisted interpretation of remotely sensed data.
Statistical Analysis System
For statistical analysis, IDRISI offers both traditional statistical procedures as well as some specialized routines for the
statistical analysis of spatial data. Geographers have developed a series of specialized routines for the statistical description
of spatial data, partly because of the special character of spatial data, but also because spatial data pose special problems
Chapter 2 Introduction to GIS
7
for inferences drawn from statistical procedures.
Decision Support System
While decision support is one of the most important functions of a GIS, tools designed especially for this are relatively
few in most GIS software. However, IDRISI includes several modules specifically developed to aid in the resource allocation decision making process. These include modules that incorporate error into the process, help in the construction of
multi-criteria suitability maps under varying levels of tradeoff, and address allocation decisions when there are multiple
objectives involved. Used in conjunction with the other components of the system, these modules provide a powerful
tool for resource allocation decision makers.
Map Data Representation
The way in which the software components mentioned above are combined is one aspect of how Geographic Information Systems vary. However, an even more fundamental distinction is how they represent map data in digital form.
A Geographic Information System stores two types of data that are found on a map—the geographic definitions of earth
surface features and the attributes or qualities that those features possess. Not all systems use the same logic for achieving
this. Nearly all, however, use one or a combination of both of the fundamental map representation techniques: vector and
raster.
Vector
With vector representation, the boundaries or the course of the features are defined by a series of points that, when joined
with straight lines, form the graphic representation of that feature. The points themselves are encoded with a pair of numbers giving the X and Y coordinates in systems such as latitude/longitude or Universal Transverse Mercator grid coordinates. The attributes of features are then stored with a traditional database management (DBMS) software program. For
example, a vector map of property parcels might be tied to an attribute database of information containing the address,
owner's name, property valuation and land use. The link between these two data files can be a simple identifier number
that is given to each feature in the map (Figure 2-2).
Raster
The second major form of representation is known as raster. With raster systems, the graphic representation of features
and the attributes they possess are merged into unified data files. In fact, we typically do not define features at all. Rather,
the study area is subdivided into a fine mesh of grid cells in which we record the condition or attribute of the earth's surface at that point (Figure 2-2). Each cell is given a numeric value which may then represent either a feature identifier, a
qualitative attribute code or a quantitative attribute value. For example, a cell could have the value "6" to indicate that it
belongs to District 6 (a feature identifier), or that it is covered by soil type 6 (a qualitative attribute), or that it is 6 meters
above sea level (a quantitative attribute value). Although the data we store in these grid cells do not necessarily refer to
phenomena that can be seen in the environment, the data grids themselves can be thought of as images or layers, each
depicting one type of information over the mapped region. This information can be made visible through the use of a raster display. In a raster display, such as the screen on your computer, there is also a grid of small cells called pixels. The
word pixel is a contraction of the term picture element. Pixels can be made to vary in their color, shape or grey tone. To
make an image, the cell values in the data grid are used to regulate directly the graphic appearance of their corresponding
pixels. Thus in a raster system, the data directly controls the visible form we see.
Raster versus Vector
Raster systems are typically data intensive (although good data compaction techniques exist) since they must record data
IDRISI Guide to GIS and Image Processing Volume 1
8
at every cell location regardless of whether that cell holds information that is of interest or not. However, the advantage
is that geographical space is uniformly defined in a simple and predictable fashion. As a result, raster systems have substantially more analytical power than their vector counterparts in the analysis of continuous space3 and are thus ideally
suited to the study of data that are continuously changing over space such as terrain, vegetation biomass, rainfall and the
like. The second advantage of raster is that its structure closely matches the architecture of digital computers. As a result,
raster systems tend to be very rapid in the evaluation of problems that involve various mathematical combinations of the
data in multiple layers. Hence they are excellent for evaluating environmental models such as soil erosion potential and
forest management suitability. In addition, since satellite imagery employs a raster structure, most raster systems can easily
incorporate these data, and some provide full image processing capabilities.
Vector
Raster
802
812
801
2
810
806
803
809
804
3
1
808
ID
801
802
803
Vector
804
805
806
807
807
LAND-USE
201
202
112
201
312
201
111
805
AREA
6305
6412
7221
12532
14638
Figure
6120
8914
2-2
0
0
0
0
1
1
1
1
1
0
0
0
0
0
1
1
1
1
0 0 0 2
0 0 2 2
0 2 2 2
0 3 3 3
0Raster
1 3 3
1 1 3 3
1 3 3 3
1 0 3 3
1 0 0 3
2
2
2
2
2
0
3
0
0
2
2
2
2
0
0
3
0
0
2
2
2
0
0
0
0
0
0
While raster systems are predominantly analysis oriented, vector
systems tend to be more database management oriented. Vector
systems are quite efficient in their storage of map data because
they only store the boundaries of features and not that which is
inside those boundaries. Because the graphic representation of features is directly linked to the attribute database, vector systems
usually allow one to roam around the graphic display with a mouse
and query the attributes associated with a displayed feature, such as
the distance between points or along lines, the areas of regions
defined on the screen, and so on. In addition, they can produce
simple thematic maps of database queries, such as one showing all
sewer line sections over one meter in diameter installed before
1940.
Compared to their raster counterparts, vector systems do not have
as extensive a range of capabilities for analyses over continuous
Vector
Raster
space. They do, however, excel at problems concerning moveFigure 2-2
ments over a network and can undertake the most fundamental of
GIS operations that will be sketched out below. For many, it is the
simple database management functions and excellent mapping capabilities that make vector systems attractive. Because of
the close affinity between the logic of vector representation and traditional map production, vector systems are used to
produce maps that are indistinguishable from those produced by traditional means. As a result, vector systems are very
popular in municipal applications where issues of engineering map production and database management predominate.
Raster and vector systems each have their special strengths. As a result, IDRISI incorporates elements from both representational techniques. Though it is primarily a raster analytical system, IDRISI does employ vector data structures as a
major form of map data display and exchange. In addition, fundamental aspects of vector database management are also
provided.
Geographic Database Concepts
Organization
Whether we use a raster or vector logic for spatial representation, we begin to see that a geographic database—a complete
database for a given region—is organized in a fashion similar to a collection of maps (Figure 2-3). Vector systems may
come closest to this logic with what are known as coverages—map-like collections that contain the geographic definitions
3. The basic data structure of vector systems can best be described as a network. As a result, it is not surprising to find that vector systems have excellent
capabilities for the analysis of network space. Thus the difference between raster and vector is less one of inherent ability than one of the difference in
the types of space they describe.
Chapter 2 Introduction to GIS
9
of a set of features and their associated attribute tables. However, they differ from maps in two ways. First, each will typically contain information on only a single feature type, such as property parcels, soils polygons, and the like. Second, they
may contain a whole series of attributes that pertain to those features, such as a set of census information for city blocks.
water
roads
soils
elevation
Figure 2-3
Raster systems also use this map-like logic, but usually divide data sets into unitary layers. A layer contains all the data for a
single attribute. Thus one might have a soils layer, a roads layer and a land use layer. A few raster systems, including
IDRISI, can link a feature identifier layer (a layer that contains the identifiers of the features located at each grid cell) with
attribute tables. More commonly, separate layers exist for each attribute and on-screen displays and paper maps are produced from these, either singly or in combination.
Although there are subtle differences, for all intents and purposes, raster layers and vector coverages can be thought of as
simply different manifestations of the same concept—the organization of the database into elementary map-like themes.
Layers and coverages differ from traditional paper maps, however, in an important way. When map data are encoded in
digital form (digitized), scale differences are removed. The digital data may be displayed or printed at any scale. More
importantly, digital data layers that were derived from paper maps of different scales, but covering the same geographic
area, may be combined.
In addition, many GIS packages, including IDRISI, provide utilities for changing the projection and reference system of
digital layers. This allows multiple layers, digitized from maps having various projections and reference systems, to be
converted to a common system.
With the ability to manage differences of scale, projection and reference system, layers can be merged with ease, eliminating a problem that has traditionally hampered planning activities with paper maps. It is important to note, however, that
the issue of resolution of the information in the data layers remains. Although features digitized from a poster-sized world
map could be combined in a GIS with features digitized from a very large scale local map, such as a city street map, this
would normally not be done. The level of accuracy and detail of the digital data can only be as good as that of the original
maps.
Georeferencing
All spatial data files in a GIS are georeferenced. Georeferencing refers to the location of a layer or coverage in space as
defined by a known coordinate referencing system. With raster images, a common form of georeferencing is to indicate
IDRISI Guide to GIS and Image Processing Volume 1
10
the reference system (e.g., latitude/longitude), the reference units (e.g., degrees) and the coordinate positions of the left,
right, top and bottom edges of the image. The same is true of vector data files, although the left, right, top and bottom
edges now refer to what is commonly called the bounding rectangle of the coverage—a rectangle which defines the limits of
the mapped area.4 This information is particularly important in an integrated GIS such as IDRISI since it allows raster
and vector files to be related to one another in a reliable and meaningful way. It is also vital for the referencing of data values to actual positions on the ground.
Georeferencing is an extremely important consideration when using GIS. Therefore a separate chapter later in this volume treats this topic in detail.
Analysis in GIS
The organization of the database into layers is not simply for reasons of organizational clarity. Rather, it is to provide
rapid access to the data elements required for geographic analysis. Indeed, the raison d'être for GIS is to provide a medium
for geographic analysis.
The analytical characteristics of GIS can be looked at in two ways. First, one can look at the tools that GIS provides. Then
one can look at the kinds of operations that GIS allows. Regardless of whether we are using a raster or a vector system,
we will tend to find that the tools fall into four basic groups and that the operations undertaken fall into three.
Analytical Tools
Database Query
The most fundamental of all tools provided by a GIS are those involved with Database Query. Database query simply
asks questions about the currently-stored information. In some cases, we query by location—what land use is at this location?
In other cases, we query by attribute—what areas have high levels of radon gas? Sometimes we undertake simple queries such as
those just illustrated, and at other times we ask about complex combinations of conditions—show me all wetlands that are
larger than 1 hectare and that are adjacent to industrial lands.
In most systems, including IDRISI, these query operations are undertaken in two steps. The first step, called a reclassification, creates a new layer of each individual condition of interest (Figure 2-4). For example, consider a query to find residential areas on bedrock associated with high levels of radon gas. The first step would be to create a layer of residential
areas alone by reclassifying all landuse codes into only two—a 1 whenever an area is residential and a 0 for all other cases.
The resulting layer is known as a Boolean layer since it shows only those areas that meet the condition (1 = true, residential)
and those that don't (0 = false, not residential). Boolean layers are also called logical layers since they show only true/false
relationships. They are also sometimes called binary layers since they contain only zeros and ones. We will avoid using that
term, however, since it also describes a particular kind of data storage format. Here we will call them Boolean layers.
Once the residential layer has been created, a geology layer is then also reclassified to create a Boolean layer showing areas
with bedrock associated with high levels of radon gas. At this point we can combine the two conditions using an overlay
operation (Figure 2-4). As mentioned previously, it is only a GIS that can combine conditions such as this that involve
features with different geographies. Typically, an overlay operation in GIS will allow the production of new layers based
on some logical or mathematical combination of two or more input layers. In the case of database query, the key logical
operations of interest are the AND and OR relational operators, also known as the INTERSECTION and UNION
operations respectively. Here we are looking for cases of residential land AND high radon gas—the logical intersection of
4. The bounding rectangle is defined by the study region of interest and does not necessarily refer to the actual minimum and maximum coordinates
in the data file.
Chapter 2 Introduction to GIS
11
our two Boolean layers.
Soils
Boundaries
Reclassification
Overlay
Figure 2-4
Map Algebra
The second set of tools that a GIS will typically provide is that for combining map layers mathematically. Modeling in particular requires the ability to combine layers according to various mathematical equations. For example, we might have an
equation that predicts mean annual temperature as a result of altitude. Or, as another example, consider the possibility of
creating a soil erosion potential map based on factors of soil erodability, slope gradient and rainfall intensity. Clearly we
need the ability to modify data values in our map layers by various mathematical operations and transformations and to
combine factors mathematically to produce the final result.
The Map Algebra tools will typically provide three different kinds of operations:
1. the ability to mathematically modify the attribute data values by a constant (i.e., scalar arithmetic);
2. the ability to mathematically transform attribute data values by a standard operation (such as the trigonometric
functions, log transformations and so on);
3. the ability to mathematically combine (such as add, subtract, multiply, divide) different data layers to produce
a composite result.
This third operation is simply another form of overlay—mathematical overlay, as opposed to the logical overlay of database query.
To illustrate this, consider a model for snow melt in densely forested areas:5
M = (0.19T + 0.17D)
where M is the melt rate in cm/day, T is the air temperature and D is the dewpoint temperature. Given layers of the air
temperatures and dewpoints for a region of this type, we could clearly produce a snow melt rate map. To do so would
require multiplying the temperature layer by 0.19 (a scalar operation), the dewpoint layer by 0.17 (another scalar opera5. Equation taken from Dunne, T., and Leopold, L.B., (1978) Water in Environmental Planning, (W.H. Freeman and Co.: San Francisco), 480.
IDRISI Guide to GIS and Image Processing Volume 1
12
tion) and then using overlay to add the two results. While simple in concept, this ability to treat map layers as variables in
algebraic formulas is an enormously powerful capability.
Distance Operators
The third tool group provided by GIS consists of the Distance Operators. As the name suggests, these are a set of techniques where distance plays a key role in the analysis undertaken. Virtually all systems provide the tools to construct
buffer zones—areas within a specified distance of designated target features. Some can also evaluate the distance of all
locations to the nearest of a set of designated features, while others can even incorporate frictional effects and barriers in
distance calculations (Figure 2-5).
When frictional effects are incorporated, the distance calculated is often referred to as a cost distance. This name is used
because movement through space can be considered to incur costs, either in money, time or effort. Frictions increase
those costs. When the costs of movement from one or more locations are evaluated for an entire region, we often refer to
the result as a cost surface (Figure 2-5). In this case, areas of low cost (presumably near to the starting point) can be seen as
valleys and areas of high cost as hills. A cost surface thus has its lowest point(s) at the starting location(s) and its highest
point(s) at the locations that are farthest away (in the sense of the greatest accumulated cost).6
Target Features
Buffers
Distance
Cost-Distance
Figure 2-5
There may be cases in which frictions do not affect the cost of movement the same way in all directions. In other words,
they act anisotropically. For example, going up a steep slope might incur a cost that would be higher than going down the
same steep slope. Thus the direction of movement through the friction is important, and must be taken into account
when developing the cost surface. IDRISI provides modules to model this type of cost surface which are explained in
detail in the Anisotropic Cost Analysis chapter in the IDRISI Guide to GIS and Image Processing Volume 2.
Given the concept of a cost surface, Geographic Information Systems also commonly offer least cost path analysis—another
important distance operation. As the name suggests, our concern is to evaluate the least-cost path between two locations.
The cost surface provides the needed information for this to be evaluated (Figure 2-8).
Regardless of how distance is evaluated, by straight line distance or by cost distance, another commonly provided tool is
allocation. With allocation, we assign locations to the nearest of a set of designated features. For example, we might establish a set of health facilities and then wish to allocate residents to their nearest facility, where "near" might mean linear distance, or a cost-distance such as travel time.
6. It should be noted here that a cost surface as just described can only be evaluated with a raster system. For vector systems, the closest equivalent
would be cost distances evaluated over a network. Here we see a simple, but very powerful, illustration of the differences between raster and vector systems in how they conceive of space.
Chapter 2 Introduction to GIS
13
Context Operators
Finally, most Geographic Information Systems provide a variety of Context Operators (also known as neighborhood or local
operators). With context operators, we create new layers based on the information on an existing map and the context in
which it is found. One of the simplest examples of this is surface analysis where we use a digital elevation model to produce
a slope layer by examining the heights of locations in comparison to the heights of neighboring locations. In a similar
fashion, the aspect (the direction of maximum downward slope) can also be evaluated. We might also position an artificial
light source and calculate a shaded relief model. These context operator products and the elevation model from which
they were derived are illustrated in Figure 2-6.
Elevation Model
Slope
Aspect
Shaded Relief
Figure 2-6
A second good example of a context operator is a digital filter. Digital filters operate by changing values according to the
character of neighboring values. For example, a surface of terrain heights can be smoothed by replacing values with the
average of the original height and all neighboring heights. Digital filters have a broad range of applications in GIS and
remote sensing, ranging from noise removal to the visual enhancement of images.
Because of their simple and uniform data structure, raster systems tend to offer a broad range of context operators. In
IDRISI, for example, these include surface analysis and digital filtering, identification of contiguous areas, watershed analysis, viewshed analysis (an evaluation of all areas in view of one or more designated features) and a special supply/demand
modeling procedure where demands are satisfied by taking supplies in a radial fashion from neighboring locations.
Analytical Operations
Given these basic tools, a broad range of analytical operations can be undertaken. However, it would appear that most of
these fall into one of three basic groups: Database Query, Derivative Mapping and Process Modeling.
Database Query
With database query, we are simply selecting out various combinations of variables for examination. The tools we use are
largely the Database Query tools previously discussed (hence the name), but also include various measurement and statistical analysis procedures. The key thing that distinguishes this kind of analysis is that we have taken out no more than we
have put into the system. While we may extract combinations we have never examined before, the system provides us
with no new information—we are simply making a withdrawal from a data bank we have built up.
One of the key activities in database query is pattern seeking. Typically we are looking for spatial patterns in the data that
may lead us to hypothesize about relationships between variables.
IDRISI Guide to GIS and Image Processing Volume 1
14
Derivative Mapping
With derivative mapping, we combine selected components of our database to yield new derivative layers. For example,
we might take our digital elevation data to derive slope gradients, and then take our slope data and combine it with information on soil type and rainfall regime to produce a new map of soil erosion potential. This new map then becomes an
addition to our growing database.
How is it that we can create new data from old? Unlike database query where we simply extracted information that was
already in the database, with derivative mapping we take existing information and add to it something new—knowledge of
relationships between database elements. We can create a soil erosion potential map using a digital elevation layer, a soils
layer and a rainfall regime layer, only if we know the relationship between those factors and the new map we are creating.
In some cases, these relationships will be specified in logical terms (such as creating a suitability map for industrial location based on the condition that it be on existing forest land, outside protective buffers around wetlands and on low
slopes) and we will use our database query tools. In other cases, however, these relationships will be specified in mathematical terms and we will rely heavily on the Map Algebra tools. Regardless, the relationships that form the model will
need to be known.
In some cases, the relationship models can be derived on logical or theoretical grounds. However, in many instances it is
necessary that the relationships be determined by empirical study. Regression analysis, for example, is one very common
way in which empirical testing is used to develop a mathematical relationship between variables. If one takes the soil erosion example, one might set up a series of test sites at which the soil erosion is measured along with the slope, soil type
and rainfall data. These sample points would then be used to develop the equation relating soil erosion to these variables.
The equation would then be used to evaluate soil erosion potential over a much broader region.
Process Modeling
Database query and derivative mapping make up the bulk of GIS analysis undertaken today. However, there is a third area
that offers incredible potential—Process or Simulation Modeling.
With process modeling, we also bring something new to the database—knowledge of process. Process refers to the causal
chain by which some event takes place. For example, a simple model of fuel-wood demand satisfaction might run as follows:
1.
Take all the wood you need (if you can) from your present location.
2.
If your demand is satisfied or if you have traveled more than 10 kilometers from home, go to step 4.
3.
If your demand is not met, move to an immediately adjacent location not already visited and repeat step 1.
4.
Stop.
Process modeling is a particularly exciting prospect for GIS. It is based on the notion that in GIS, our database doesn't
simply represent an environment, it is an environment! It is a surrogate environment, capable of being measured, manipulated and acted upon by geographic and temporal processes. Our database thus acts as a laboratory for the exploration of
processes in a complex environment. Traditionally, in science, we have had to remove that complexity in order to understand processes in isolation. This has been an effective strategy and we have learned much from it. However, technologies
such as GIS now offer the tools to reassemble those simple understandings in order to gain an understanding and appreciation of how they act in the full complexity of a real environmental setting. Often even very simple understandings yield
complex patterns when allowed to interact in the environment.
A different sort of process, the decision making process, may also be supported and in some ways modeled with the use
of GIS. GIS technology is becoming more important as a tool for decision support. Indeed, even the simplest database
query results may prove to be invaluable input to the decision maker. However, the more complex process of decision making, in which decision makers often think in terms of multiple criteria, soft boundaries (non-Boolean) and levels of
Chapter 2 Introduction to GIS
15
acceptable risk, may also be modeled using GIS. IDRISI provides a suite of decision support modules to help decision
makers develop more explicitly rational and well-informed decisions. The Decision Making chapter in the IDRISI
Guide to GIS and Image Processing Volume 2 discusses this important use of GIS in detail and provides illustrative
case examples.
Despite its evident attraction, process modeling, both in environmental processes and decision making, is still a fairly
uncommon activity in GIS. The reason is quite simple. While more and more modeling tools are becoming available in
the GIS, it is not uncommon that the process of interest requires a capability not built into the system. These cases require
the creation of a new program module. Many systems are not well set up for the incorporation of user-developed routines. IDRISI, however, has been designed so programs in any computer language may be merged into the system and
called from the IDRISI interface.
The Philosophy of GIS
GIS has had an enormous impact on virtually every field that manages and analyzes spatially distributed data. For those
who are unfamiliar with the technology, it is easy to see it as a magic box. The speed, consistency and precision with
which it operates is truly impressive, and its strong graphic character is hard to resist. However, to experienced analysts,
the philosophy of GIS is quite different. With experience, GIS becomes simply an extension of one's own analytical
thinking. The system has no inherent answers, only those of the analyst. It is a tool, just like statistics is a tool. It is a tool
for thought.
Investing in GIS requires more than an investment in hardware and software. Indeed, in many instances this is the least
issue of concern. Most would also recognize that a substantial investment needs to be placed in the development of the
database. However, one of the least recognized yet most important investment is in the analysts who will use the system.
The system and the analyst cannot be separated—one is simply an extension of the other. In addition, the process of
incorporating GIS capabilities into an institution requires an investment in long-term and organization-wide education
and training.
In many ways, learning GIS involves learning to think—learning to think about patterns, about space and about processes
that act in space. As you learn about specific procedures, they will often be encountered in the context of specific examples. In addition, they will often have names that suggest their typical application. But resist the temptation to categorize
these routines. Most procedures have many more general applications and can be used in many novel and innovative
ways. Explore! Challenge what you see! What you will learn goes far beyond what this or any software package can provide.
IDRISI Guide to GIS and Image Processing Volume 1
16
Introduction to Remote Sensing and Image
Processing
Of all the various data sources used in GIS, one of the most important is undoubtedly that provided by remote sensing.
Through the use of satellites, we now have a continuing program of data acquisition for the entire world with time frames
ranging from a couple of weeks to a matter of hours. Very importantly, we also now have access to remotely sensed
images in digital form, allowing rapid integration of the results of remote sensing analysis into a GIS.
The development of digital techniques for the restoration, enhancement and computer-assisted interpretation of remotely
sensed images initially proceeded independently and somewhat ahead of GIS. However, the raster data structure and
many of the procedures involved in these Image Processing Systems (IPS) were identical to those involved in raster GIS. As a
result, it has become common to see IPS software packages add general capabilities for GIS, and GIS software systems
add at least a fundamental suite of IPS tools. IDRISI is a combined GIS and image processing system that offers
advanced capabilities in both areas.
Because of the extreme importance of remote sensing as a data input to GIS, it has become necessary for GIS analysts
(particularly those involved in natural resource applications) to gain a strong familiarity with IPS. Consequently, this chapter gives an overview of this important technology and its integration with GIS. The Image Processing exercises in the
Tutorial illustrate many of the concepts presented here.
Definition
Remote sensing can be defined as any process whereby information is gathered about an object, area or phenomenon
without being in contact with it. Our eyes are an excellent example of a remote sensing device. We are able to gather
information about our surroundings by gauging the amount and nature of the reflectance of visible light energy from
some external source (such as the sun or a light bulb) as it reflects off objects in our field of view. Contrast this with a
thermometer, which must be in contact with the phenomenon it measures, and thus is not a remote sensing device.
Given this rather general definition, the term remote sensing has come to be associated more specifically with the gauging of
interactions between earth surface materials and electromagnetic energy. However, any such attempt at a more specific
definition becomes difficult, since it is not always the natural environment that is sensed (e.g., art conservation applications), the energy type is not always electromagnetic (e.g., sonar) and some procedures gauge natural energy emissions
(e.g., thermal infrared) rather than interactions with energy from an independent source.
Fundamental Considerations
Energy Source
Sensors can be divided into two broad groups—passive and active. Passive sensors measure ambient levels of existing
sources of energy, while active ones provide their own source of energy. The majority of remote sensing is done with passive sensors, for which the sun is the major energy source. The earliest example of this is photography. With airborne
cameras we have long been able to measure and record the reflection of light off earth features. While aerial photography
is still a major form of remote sensing, newer solid state technologies have extended capabilities for viewing in the visible
and near-infrared wavelengths to include longer wavelength solar radiation as well. However, not all passive sensors use
energy from the sun. Thermal infrared and passive microwave sensors both measure natural earth energy emissions. Thus
Chapter 3 Introduction to Remote Sensing and Image Processing
17
the passive sensors are simply those that do not themselves supply the energy being detected.
By contrast, active sensors provide their own source of energy. The most familiar form of this is flash photography. However, in environmental and mapping applications, the best example is RADAR. RADAR systems emit energy in the
microwave region of the electromagnetic spectrum (Figure 3-1). The reflection of that energy by earth surface materials is
then measured to produce an image of the area sensed.
Red
Blue
Green
0.4 0.5 0.6 0.7
UV
(µm)
Near - infrared
Visible
10 −6 10 −5 10 −4 10 −3 10 −2 10 −1 1
Wavelength (µm)
ic
M
10 8
10 9
w
e
av
Te
an l e v
d isio
n
io
ad
R
ro
R
ra
ys
(1m)
10 2 10 3 10 4 10 5 10 6 10 7
M T
i d he
- I rm
R a
lI
IR
V)
r(U
ea
N ible let
s io
Vi r a v
lt
U
ic
ys
ra
m
X
os
s
ay
γr
C
(1mm)
10
Wavelength
(µm)
a
From Lillesand and Kiefer 1987
Figure 3-1: The Electromagnetic Spectrum
Wavelength
As indicated, most remote sensing devices make use of electromagnetic energy. However, the electromagnetic spectrum is
very broad and not all wavelengths are equally effective for remote sensing purposes. Furthermore, not all have significant
interactions with earth surface materials of interest to us. Figure 3-1 illustrates the electromagnetic spectrum. The atmosphere itself causes significant absorption and/or scattering of the very shortest wavelengths. In addition, the glass lenses
of many sensors also cause significant absorption of shorter wavelengths such as the ultraviolet (UV). As a result, the first
significant window (i.e., a region in which energy can significantly pass through the atmosphere) opens up in the visible
wavelengths. Even here, the blue wavelengths undergo substantial attenuation by atmospheric scattering, and are thus
often left out in remotely sensed images. However, the green, red and near-infrared (IR) wavelengths all provide good
opportunities for gauging earth surface interactions without significant interference by the atmosphere. In addition, these
regions provide important clues to the nature of many earth surface materials. Chlorophyll, for example, is a very strong
absorber of red visible wavelengths, while the near-infrared wavelengths provide important clues to the structures of
plant leaves. As a result, the bulk of remotely sensed images used in GIS-related applications are taken in these regions.
Extending into the middle and thermal infrared regions, a variety of good windows can be found. The longer of the middle infrared wavelengths have proven to be useful in a number of geological applications. The thermal regions have
proven to be very useful for monitoring not only the obvious cases of the spatial distribution of heat from industrial activity, but a broad set of applications ranging from fire monitoring to animal distribution studies to soil moisture conditions.
After the thermal IR, the next area of major significance in environmental remote sensing is in the microwave region. A
number of important windows exist in this region and are of particular importance for the use of active radar imaging.
The texture of earth surface materials causes significant interactions with several of the microwave wavelength regions.
This can thus be used as a supplement to information gained in other wavelengths, and also offers the significant advantage of being usable at night (because as an active system it is independent of solar radiation) and in regions of persistent
cloud cover (since radar wavelengths are not significantly affected by clouds).
IDRISI Guide to GIS and Image Processing Volume 1
18
Interaction Mechanisms
When electromagnetic energy strikes a material, three types of interaction can follow: reflection, absorption and/or transmission (Figure 3-2). Our main concern is with the reflected portion since it is usually this which is returned to the sensor
system. Exactly how much is reflected will vary and will depend upon the nature of the material and where in the electromagnetic spectrum our measurement is being taken. As a result, if we look at the nature of this reflected component over
a range of wavelengths, we can characterize the result as a spectral response pattern.
Light Source
reflection
absorption
transmission
Figure 3-2
Spectral Response Patterns
A spectral response pattern is sometimes called a signature. It is a description (often in the form of a graph) of the degree
to which energy is reflected in different regions of the spectrum. Most humans are very familiar with spectral response
patterns since they are equivalent to the human concept of color. For example, Figure 3-3 shows idealized spectral
response patterns for several familiar colors in the visible portion of the electromagnetic spectrum, as well as for white
and dark grey. The bright red reflectance pattern, for example, might be that produced by a piece of paper printed with a
red ink. Here, the ink is designed to alter the white light that shines upon it and absorb the blue and green wavelengths.
What is left, then, are the red wavelengths which reflect off the surface of the paper back to the sensing system (the eye).
The high return of red wavelengths indicates a bright red, whereas the low return of green wavelengths in the second
example suggests that it will appear quite dark.
Chapter 3 Introduction to Remote Sensing and Image Processing
19
B
G
R
B
bright red
B
G
purple
G
R
B
dark green
R
B
G
white
G
R
yellow
R
B
G
R
dark gray
Figure 3-3
The eye is able to sense spectral response patterns because it is truly a multi-spectral sensor (i.e., it senses in more than
one place in the spectrum). Although the actual functioning of the eye is quite complex, it does in fact have three separate
types of detectors that can usefully be thought of as responding to the red, green and blue wavelength regions. These are
the additive primary colors, and the eye responds to mixtures of these three to yield a sensation of other hues. For example,
the color perceived by the third spectral response pattern in Figure 3-3 would be a yellow—the result of mixing a red and
green. However, it is important to recognize that this is simply our phenomenological perception of a spectral response
pattern. Consider, for example, the fourth curve. Here we have reflectance in both the blue and red regions of the visible
spectrum. This is a bimodal distribution, and thus technically not a specific hue in the spectrum. However, we would perceive this to be a purple! Purple (a color between violet and red) does not exist in nature (i.e., as a hue—a distinctive dominant wavelength). It is very real in our perception, however. Purple is simply our perception of a bimodal pattern
involving a non-adjacent pair of primary hues.
In the early days of remote sensing, it was believed (more correctly hoped) that each earth surface material would have a
distinctive spectral response pattern that would allow it to be reliably detected by visual or digital means. However, as our
common experience with color would suggest, in reality this is often not the case. For example, two species of trees may
have quite a different coloration at one time of the year and quite a similar one at another.
Finding distinctive spectral response patterns is the key to most procedures for computer-assisted interpretation of
remotely sensed imagery. This task is rarely trivial. Rather, the analyst must find the combination of spectral bands and
the time of year at which distinctive patterns can be found for each of the information classes of interest.
For example, Figure 3-4 shows an idealized spectral response pattern for vegetation along with those of water and dry
bare soil. The strong absorption by leaf pigments (particularly chlorophyll for purposes of photosynthesis) in the blue and
red regions of the visible portion of the spectrum leads to the characteristic green appearance of healthy vegetation. However, while this signature is distinctively different from most non-vegetated surfaces, it is not very capable of distinguishing
between species of vegetation—most will have a similar color of green at full maturation. In the near-infrared, however,
we find a much higher return from vegetated surfaces because of scattering within the fleshy mesophyllic layer of the
leaves. Plant pigments do not absorb energy in this region, and thus the scattering, combined with the multiplying effect
of a full canopy of leaves, leads to high reflectance in this region of the spectrum. However, the extent of this reflectance
will depend highly on the internal structure of leaves (e.g., broadleaf versus needle). As a result, significant differences
between species can often be detected in this region. Similarly, moving into the middle infrared region we see a significant
dip in the spectral response pattern that is associated with leaf moisture. This is, again, an area where significant differences can arise between mature species. Applications looking for optimal differentiation between species, therefore, will
typically involve both the near and middle infrared regions and will use imagery taken well into the development cycle.
IDRISI Guide to GIS and Image Processing Volume 1
20
Relative Reflectance
Dry bare soil
Vegetation
Water (clear)
0.4
0.8
Figure 3-4
1.2
1.6
Wavelength (mm)
2.0
2.4
Adapted from Lillesand and Kiefer 1987
Multispectral Remote Sensing
In the visual interpretation of remotely sensed images, a variety of image characteristics are brought into consideration:
color (or tone in the case of panchromatic images), texture, size, shape, pattern, context, and the like. However, with computer-assisted interpretation, it is most often simply color (i.e., the spectral response pattern) that is used. It is for this reason that a strong emphasis is placed on the use of multispectral sensors (sensors that, like the eye, look at more than one
place in the spectrum and thus are able to gauge spectral response patterns), and the number and specific placement of
these spectral bands.
Figure 3-5 illustrates the spectral bands of the LANDSAT Thematic Mapper (TM) system. The LANDSAT satellite is a
commercial system providing multi-spectral imagery in seven spectral bands at a 30 meter resolution.
It can be shown through analytical techniques such as Principal Components Analysis, that in many environments, the
bands that carry the greatest amount of information about the natural environment are the near-infrared and red wavelength bands. Water is strongly absorbed by infrared wavelengths and is thus highly distinctive in that region. In addition,
plant species typically show their greatest differentiation here. The red area is also very important because it is the primary
region in which chlorophyll absorbs energy for photosynthesis. Thus it is this band which can most readily distinguish
between vegetated and non-vegetated surfaces.
Given this importance of the red and near-infrared bands, it is not surprising that sensor systems designed for earth
resource monitoring will invariably include these in any particular multispectral system. Other bands will depend upon the
range of applications envisioned. Many include the green visible band since it can be used, along with the other two, to
produce a traditional false color composite—a full color image derived from the green, red, and infrared bands (as
opposed to the blue, green, and red bands of natural color images). This format became common with the advent of color
infrared photography, and is familiar to many specialists in the remote sensing field. In addition, the combination of these
three bands works well in the interpretation of the cultural landscape as well as natural and vegetated surfaces. However,
it is increasingly common to include other bands that are more specifically targeted to the differentiation of surface materials. For example, LANDSAT TM Band 5 is placed between two water absorption bands and has thus proven very useful
in determining soil and leaf moisture differences. Similarly, LANDSAT TM Band 7 targets the detection of hydrothermal
alteration zones in bare rock surfaces. By contrast, the AVHRR system on the NOAA series satellites includes several
thermal channels for the sensing of cloud temperature characteristics.
Chapter 3 Introduction to Remote Sensing and Image Processing
21
Band 1, visible blue
0.45-0.52 mm
Band 2, visible green
0.52-0.60 mm
Band 3, visible red
0.63-0.69 mm
Band 5, middle-infrared
1.55-1.75 mm
Band 6, thermal infrared
10.4-12.5 mm
Band 7, middle-infrared
2.08-2.35 mm
Band 4, near-infrared
0.76-0.90 mm
Figure 3-5
Hyperspectral Remote Sensing
In addition to traditional multispectral imagery, some new and experimental systems such as AVIRIS and MODIS are
capable of capturing hyperspectral data. These systems cover a similar wavelength range to multispectral systems, but in
much narrower bands. This dramatically increases the number of bands (and thus precision) available for image classification (typically tens and even hundreds of very narrow bands). Moreover, hyperspectral signature libraries have been created in lab conditions and contain hundreds of signatures for different types of landcovers, including many minerals and
other earth materials. Thus, it should be possible to match signatures to surface materials with great precision. However,
environmental conditions and natural variations in materials (which make them different from standard library materials)
make this difficult. In addition, classification procedures have not been developed for hyperspectral data to the degree
they have been for multispectral imagery. As a consequence, multispectral imagery still represents the major tool of
remote sensing today.
Sensor/Platform Systems
Given recent developments in sensors, a variety of platforms are now available for the capture of remotely sensed data.
Here we review some of the major sensor/platform combinations that are typically available to the GIS user community.
IDRISI Guide to GIS and Image Processing Volume 1
22
Aerial Photography
Aerial photography is the oldest and most widely used method of remote sensing. Cameras mounted in light aircraft flying
between 200 and 15,000 m capture a large quantity of detailed information. Aerial photos provide an instant visual inventory of a portion of the earth's surface and can be used to create detailed maps. Aerial photographs commonly are taken
by commercial aerial photography firms which own and operate specially modified aircraft equipped with large format (23
cm x 23 cm) mapping quality cameras. Aerial photos can also be taken using small format cameras (35 mm and 70 mm),
hand-held or mounted in unmodified light aircraft.
Camera and platform configurations can be grouped in terms of oblique and vertical. Oblique aerial photography is taken
at an angle to the ground. The resulting images give a view as if the observer is looking out an airplane window. These
images are easier to interpret than vertical photographs, but it is difficult to locate and measure features on them for mapping purposes.
Vertical aerial photography is taken with the camera pointed straight down. The resulting images depict ground features
in plan form and are easily compared with maps. Vertical aerial photos are always highly desirable, but are particularly useful for resource surveys in areas where no maps are available. Aerial photos depict features such as field patterns and vegetation which are often omitted on maps. Comparison of old and new aerial photos can also capture changes within an
area over time.
Vertical aerial photos contain subtle displacements due to relief, tip and tilt of the aircraft and lens distortion. Vertical
images may be taken with overlap, typically about 60 percent along the flight line and at least 20 percent between lines.
Overlapping images can be viewed with a stereoscope to create a three-dimensional view, called a stereo model.
Large Format Photography
Commercial aerial survey firms use light single or twin engine aircraft equipped with large-format mapping cameras.
Large-format cameras, such as the Wild RC-10, use 23 cm x 23 cm film which is available in rolls. Eastman Kodak, Inc.,
among others, manufactures several varieties of sheet film specifically intended for use in aerial photography. Negative
film is used where prints are the desired product, while positive film is used where transparencies are desired. Print film
allows for detailed enlargements to be made, such as large wall-sized prints. In addition, print film is useful when multiple
prints are to be distributed and used in the field.
Small Format Photography
Small-format cameras carried in chartered aircraft are an inexpensive alternative to large-format aerial photography. A
35mm or 70mm camera, light aircraft and pilot are required, along with some means to process the film. Because there are
inexpensive commercial processing labs in most parts of the world, 35mm systems are especially convenient.
Oblique photographs can be taken with a hand-held camera in any light aircraft; vertical photographs require some form
of special mount, pointed through a belly port or extended out a door or window.
Small-format aerial photography has several drawbacks. Light unpressurized aircraft are typically limited to altitudes
below 4000 m. As film size is small, sacrifices must be made in resolution or area covered per frame. Because of distortions in the camera system, small-format photography cannot be used if precise mapping is required. In addition, presentation-quality wall-size prints cannot be made from small negatives. Nonetheless, small-format photography can be very
useful for reconnaissance surveys and can also be used as point samples.
Color Photography
Normal color photographs are produced from a composite of three film layers with intervening filters that act to isolate,
in effect, red, green, and blue wavelengths separately to the different film layers. With color infrared film, these wavelengths are shifted to the longer wavelengths to produce a composite that has isolated reflectances from the green, red
and near-infrared wavelength regions. However, because the human eye cannot see infrared, a false color composite is
produced by making the green wavelengths appear blue, the red wavelengths appear green, and the infrared wavelengths
Chapter 3 Introduction to Remote Sensing and Image Processing
23
appear red.
As an alternative to the use of color film, it is also possible to group several cameras on a single aircraft mount, each with
black and white film and a filter designed to isolate a specific wavelength range. The advantage of this arrangement is that
the bands are independently accessible and can be photographically enhanced. If a color composite is desired, it is possible to create it from the individual bands at a later time.
Clearly, photographs are not in a format that can immediately be used in digital analysis. It is possible to scan photographs
with a scanner and thereby create multispectral datasets either by scanning individual band images, or by scanning a color
image and separating the bands. However, the geometry of aerial photographs (which have a central perspective projection and differential parallax) is such that they are difficult to use directly. More typically they require processing by special
photogrammetric software to rectify the images and remove differential parallax effects.
Aerial Videography
Light, portable, inexpensive video cameras and recorders can be carried in chartered aircraft. In addition, a number of
smaller aerial mapping companies offer videography as an output option. By using several cameras simultaneously, each
with a filter designed to isolate a specific wavelength range, it is possible to isolate multispectral image bands that can be
used individually, or in combination in the form of a color composite. For use in digital analysis, special graphics hardware
boards known as frame grabbers can be used to freeze any frame within a continuous video sequence and convert it to digital format, usually in one of the more popular exchange formats such as TIF or TARGA. Like small-format photography,
aerial videography cannot be used for detailed mapping, but provides a useful overview for reconnaissance surveys, and
can be used in conjunction with ground point sampling.
Satellite-Based Scanning Systems
Photography has proven to be an important input to visual interpretation and the production of analog maps. However,
the development of satellite platforms, the associated need to telemeter imagery in digital form, and the desire for highly
consistent digital imagery have given rise to the development of solid state scanners as a major format for the capture of
remotely sensed data. The specific features of particular systems vary (including, in some cases, the removal of a true
scanning mechanism). However, in the discussion which follows, an idealized scanning system is presented that is highly
representative of current systems in use.
The basic logic of a scanning sensor is the use of a mechanism to sweep a small field of view (known as an instantaneous
field of view—IFOV) in a west to east direction at the same time the satellite is moving in a north to south direction.
Together this movement provides the means of composing a complete raster image of the environment.
A simple scanning technique is to use a rotating mirror that can sweep the field of view in a consistent west to east fashion. The field of view is then intercepted with a prism that can spread the energy contained within the IFOV into its spectral components. Photoelectric detectors (of the same nature as those found in the exposure meters of commonly
available photographic cameras) are then arranged in the path of this spectrum to provide electrical measurements of the
amount of energy detected in various parts of the electromagnetic spectrum. As the scan moves from west to east, these
detectors are polled to get a set of readings along the east-west scan. These form the columns along one row of a set of
raster images—one for each detector. Movement of the satellite from north to south then positions the system to detect
the next row, ultimately leading to the production of a set of raster images as a record of reflectance over a range of spectral bands.
There are several satellite systems in operation today that collect imagery that is subsequently distributed to users. Several
of the most common systems are described below. Each type of satellite data offers specific characteristics that make it
more or less appropriate for a particular application.
In general, there are two characteristics that may help guide the choice of satellite data: spatial resolution and spectral resolution.
The spatial resolution refers to the size of the area on the ground that is summarized by one data value in the imagery.
This is the Instantaneous Field of View (IFOV) described earlier. Spectral resolution refers to the number and width of
IDRISI Guide to GIS and Image Processing Volume 1
24
the spectral bands that the satellite sensor detects. In addition, issues of cost and imagery availability must also be considered.
LANDSAT
The LANDSAT system of remote sensing satellites is currently operated by the EROS Data Center of the United States
Geological Survey. This is a new arrangement following a period of commercial distribution under the Earth Observation
Satellite Company (EOSAT) which was recently acquired by Space Imaging Corporation. As a result, the cost of imagery
has dramatically dropped, to the benefit of all. Full or quarter scenes are available on a variety of distribution media, as
well as photographic products of MSS and TM scenes in false color and black and white.
There have been seven LANDSAT satellites, the first of which was launched in 1972. The LANDSAT 6 satellite was lost
on launch. However, as of this writing, LANDSAT 5 is still operational. LANDSAT 7 was launched in April, 1999.
LANDSAT carries two multispectral sensors. The first is the Multi-Spectral Scanner (MSS) which acquires imagery in four
spectral bands: blue, green, red and near infrared. The second is the Thematic Mapper (TM) which collects seven bands:
blue, green, red, near-infrared, two mid-infrared and one thermal infrared. The MSS has a spatial resolution of 80 meters,
while that of the TM is 30 meters. Both sensors image a 185 km wide swath, passing over each day at 09:45 local time, and
returning every 16 days. With LANDSAT 7, support for TM imagery is to be continued with the addition of a co-registered 15 m panchromatic band.
SPOT
The Système Pour L'Observation de la Terre (SPOT) was launched and has been operated by a French consortium since 1985.
SPOT satellites carry two High Resolution Visible (HRV) pushbroom sensors7 which operate in multispectral or panchromatic mode. The multispectral images have 20 meter spatial resolution while the panchromatic images have 10 meter
resolution. SPOT satellites 1-3 provide three multi-spectral bands: Green, Red and Infrared. SPOT 4, launched in 1998,
provides the same three bands plus a short wave infrared band. The panchromatic band for SPOT 1-3 is 0.51-0.73m while
that of SPOT 4 is 0.61-0.68m.
All SPOT images cover a swath 60 kilometers wide. The SPOT sensor may be pointed to image along adjacent paths.
This allows the instrument to acquire repeat imagery of any area 12 times during its 26 day orbital period. The pointing
capability makes SPOT the only satellite system which can acquire useful stereo satellite imagery.
SPOT Image Inc. sells a number of products, including digital images on a choice of magnetic media, as well as photographic products. Existing images may be purchased, or new acquisitions ordered. Customers can request the satellite to
be pointed in a particular direction for new acquisitions.
IRS
The Indian Space Research Organization currently has 5 satellites in the IRS system, with at least 7 planned by 2004.
These data are distributed by ANTRIX Corp. Ltd (the commercial arm of the Indian Space Research Organization), and
also by Space Imaging Corporation in the United States. The most sophisticated capabilities are offered by the IRS-1C
and IRS-1D satellites that together provide continuing global coverage with the following sensors:
IRS-Pan: 5.8 m panchromatic
IRS-LISS38: 23.5 m multispectral in the following bands:
Green (0.52-0.59)
7. The pushbroom sensor produces output like a scanner. However, there is no actual scanning motion. Rather, the sensor consists of a dense array of
detectors—one for each raster cell in the scan line—that is moved across the scene like a pushbroom.
8. LISS = Linear Imaging and Self Scanning Sensor. Image format is approximately 140 km x 140 km.
Chapter 3 Introduction to Remote Sensing and Image Processing
25
Red (0.62-0.68)
Near-Infrared (0.77-0.86)
Shortwave Infrared (1.55-1.7)
IRS-WiFS9: 180 m multispectral in the following bands:
Red (0.62-0.68)
Near-Infrared (0.77-0.86)
NOAA-AVHRR
The Advanced Very High Resolution Radiometer (AVHRR) is carried on board a series of satellites operated by the U.S.
National Oceanic and Atmospheric Administration (NOAA). It acquires data along a 2400-km-wide swath each day.
AVHRR collects five bands: red, near-infrared, and three thermal infrared. Spatial resolution of the sensor is 1.1 km and
this data is termed Local Area Coverage (LAC). For studying very large areas, a resampled version with resolution of
about 4 km is also available, and is termed Global Area Coverage (GAC).
AVHRR may be "high" spatial resolution for meteorological applications, but the images portray only broad patterns and
little detail for terrestrial studies. However, they do have a high temporal resolution, showing wide areas on a daily basis
and are therefore a popular choice for monitoring large areas. AVHRR imagery is used by several organizations engaged
in famine prediction and is an integral part of many early warning activities.
RADARSAT
RADARSAT is an earth observation satellite launched in November 1995 by the Canadian Space Agency. The data is distributed by RADARSAT International (RSI) of Richmond, British Columbia, Canada (or through Space Imaging in the
US). Spatial resolution of the C-band SAR imagery ranges from 8 to 100 meters per pixel and the ground coverage repeat
interval is 24 days. Sensors can be pointed at the location of interest which enables collection of stereo RADAR imagery.
RADAR signals also penetrate cloud cover, thus accessing areas not available to other remote sensing systems. In contrast
to other remotely sensed imagery, the returned RADAR signal is more affected by electrical and physical (primarily textural) characteristics in the target than by its reflection and spectral pattern, therefore requiring special interpretation and
spatial georegistration techniques. Compared to other types of remotely sensed imagery, the use of RADAR data is still in
its infancy, but has strong potential.
ERS
ERS-1 and ERS-2 (European Remote Sensing Satellite) were developed by the European Space Agency. These identical
systems provide an interesting complement to the other commercial imagery products in that they offer a variety of Cband RADAR imagery output formats. For GIS applications, the main output of interest is the side-looking airborne
RADAR (SAR) output that provides 100 km wide swaths with a 30 meter resolution. This should prove to be of considerable interest in a variety of applications, including vegetation studies and mapping projects where cloud cover is a persistent problem.
JERS
The Japanese Earth Resource Satellite offers 18 m resolution L-band side-looking RADAR imagery. This is a substantially
longer wavelength band than the typical C-band used in earth resources applications. L-band RADAR is capable of penetrating vegetation as well as unconsolidated sand and is primarily used in geologic, topographic and coastal mapping applications. JERS data is available in the United States from Space Imaging Corporation.
9. WiFS = Wide-Field Sensor. Image format is 810 km x 810 km.
IDRISI Guide to GIS and Image Processing Volume 1
26
AVIRIS
AVIRIS is an experimental system developed by the Jet Propulsion Lab (JPL) that produces hyperspectral data. It captures data in 224 bands over the same wavelength range as LANDSAT.
MODIS
The MODIS sensor onboard the EOS AM-1 platform (scheduled for launch in July, 1999) will provide a logical extension
of the AVHRR by providing no fewer than 36 bands of medium-to-coarse resolution imagery with a high temporal repeat
cycle (1-2 days). Bands 1 and 2 will provide 250 m resolution images in the red and near-infrared regions. Bands 3-7 will
provide 500 m resolution multispectral images in the visible and infrared regions. Finally, bands 8-36 will provide hyperspectral coverage in the visible, reflected infrared, and thermal infrared regions, with a 1 km resolution.
Digital Image Processing
Overview
As a result of solid state multispectral scanners and other raster input devices, we now have available digital raster images
of spectral reflectance data. The chief advantage of having these data in digital form is that they allow us to apply computer analysis techniques to the image data—a field of study called Digital Image Processing.
Digital Image Processing is largely concerned with four basic operations: image restoration, image enhancement, image
classification, image transformation. Image restoration is concerned with the correction and calibration of images in order
to achieve as faithful a representation of the earth surface as possible—a fundamental consideration for all applications.
Image enhancement is predominantly concerned with the modification of images to optimize their appearance to the visual
system. Visual analysis is a key element, even in digital image processing, and the effects of these techniques can be dramatic. Image classification refers to the computer-assisted interpretation of images—an operation that is vital to GIS. Finally,
image transformation refers to the derivation of new imagery as a result of some mathematical treatment of the raw image
bands.
In order to undertake the operations listed in this section, it is necessary to have access to Image Processing software.
IDRISI is one such system. While it is known primarily as a GIS software system, it also offers a full suite of image processing capabilities.
Image Restoration
Remotely sensed images of the environment are typically taken at a great distance from the earth's surface. As a result,
there is a substantial atmospheric path that electromagnetic energy must pass through before it reaches the sensor.
Depending upon the wavelengths involved and atmospheric conditions (such as particulate matter, moisture content and
turbulence), the incoming energy may be substantially modified. The sensor itself may then modify the character of that
data since it may combine a variety of mechanical, optical and electrical components that serve to modify or mask the
measured radiant energy. In addition, during the time the image is being scanned, the satellite is following a path that is
subject to minor variations at the same time that the earth is moving underneath. The geometry of the image is thus in
constant flux. Finally, the signal needs to be telemetered back to earth, and subsequently received and processed to yield
the final data we receive. Consequently, a variety of systematic and apparently random disturbances can combine to
degrade the quality of the image we finally receive. Image restoration seeks to remove these degradation effects.
Broadly, image restoration can be broken down into the two sub-areas of radiometric restoration and geometric restoration.
Radiometric Restoration
Radiometric restoration refers to the removal or diminishment of distortions in the degree of electromagnetic energy reg-
Chapter 3 Introduction to Remote Sensing and Image Processing
27
istered by each detector. A variety of agents can cause distortion in the values recorded for image cells. Some of the most
common distortions for which correction procedures exist include:
uniformly elevated values, due to atmospheric haze, which preferentially scatters short wavelength bands (particularly the
blue wavelengths);
striping, due to detectors going out of calibration;
random noise, due to unpredictable and unsystematic performance of the sensor or transmission of the data; and
scan line drop out, due to signal loss from specific detectors.
It is also appropriate to include here procedures that are used to convert the raw, unitless relative reflectance values
(known as digital numbers, or DN) of the original bands into true measures of reflective power (radiance).
See the chapter on Image Restoration in the IDRISI Guide to GIS and Image Processing Volume 2 for a more
detailed discussion of radiometric restoration and how it can be implemented in IDRISI.
Geometric Restoration
For mapping purposes, it is essential that any form of remotely sensed imagery be accurately registered to the proposed
map base. With satellite imagery, the very high altitude of the sensing platform results in minimal image displacements
due to relief. As a result, registration can usually be achieved through the use of a systematic rubber sheet transformation
process10 that gently warps an image (through the use of polynomial equations) based on the known positions of a set of
widely dispersed control points. This capability is provided in IDRISI through the module RESAMPLE.
With aerial photographs, however, the process is more complex. Not only are there systematic distortions related to tilt
and varying altitude, but variable topographic relief leads to very irregular distortions (differential parallax) that cannot be
removed through a rubber sheet transformation procedure. In these instances, it is necessary to use photogrammetric rectification to remove these distortions and provide accurate map measurements11. Failing this, the central portions of high
altitude photographs can be resampled with some success.
RESAMPLE is a module of major importance, and it is essential that one learn to use it effectively. Doing so also requires
a thorough understanding of reference systems and their associated parameters such as datums and projections. The
chapter on Georeferencing later in this volume provides an in-depth discussion of these issues.
Image Enhancement
Image enhancement is concerned with the modification of images to make them more suited to the capabilities of human
vision. Regardless of the extent of digital intervention, visual analysis invariably plays a very strong role in all aspects of
remote sensing. While the range of image enhancement techniques is broad, the following fundamental issues form the
backbone of this area:
10. Satellite-based scanner imagery contains a variety of inherent geometric problems such as skew (caused by rotation of the earth underneath the satellite as it is in the process of scanning a complete image) and scanner distortion (caused by the fact that the instantaneous field of view (IFOV) covers
more territory at the ends of scan lines, where the angle of view is very oblique, than in the middle). With commercially-marketed satellite imagery, such
as LANDSAT, IRS and SPOT, most elements of systematic geometric restoration associated with image capture are corrected by the distributors of the
imagery. Thus, for the end user, the only geometric operation that typically needs to be undertaken is a rubber-sheet resampling in order to rectify the
image to a map base. Many commercial distributors will perform this rectification for an additional fee.
11. Photogrammetry is the science of taking spatial measurements from aerial photographs. In order to provide a full rectification, it is necessary to have
stereoscopic images—photographs which overlap enough (e.g., 60% in the along-track direction and 10% between flight lines) to provide two independent
images of each part of the landscape. Using these stereoscopic pairs and ground control points of known position and height, it is possible to fully recreate the geometry of the viewing conditions, and thereby not only rectify measurements from such images, but also derive measurements of terrain
height. The rectified photographs are called orthophotos. The height measurements may be used to produce digital elevation models.
IDRISI Guide to GIS and Image Processing Volume 1
28
Contrast Stretch
Digital sensors have a wide range of output values to accommodate the strongly varying reflectance values that can be
found in different environments. However, in any single environment, it is often the case that only a narrow range of values will occur over most areas. Grey level distributions thus tend to be very skewed. Contrast manipulation procedures
are thus essential to most visual analyses. Figure 3-6 shows TM Band 3 (visible red) and its histogram. Note that the values of the image are quite skewed. The right image of the figure shows the same image band after a linear stretch between
values 12 and 60 has been applied. In IDRISI, this type of contrast enhancement may be performed interactively through
Composer’s Layer Properties while the image is displayed. This is normally used for visual analysis only—original data
values are used in numeric analyses. New images with stretched values are produced with the module STRETCH.
Linear Stretch
Figure 3-6
Composite Generation
For visual analysis, color composites make fullest use of the capabilities of the human eye. Depending upon the graphics
system in use, composite generation ranges from simply selecting the bands to use, to more involved procedures of band
combination and associated contrast stretch. Figure 3-7 shows several composites made with different band combinations from the same set of TM images. (See Figure 3-5 for TM band definitions.) The IDRISI module COMPOSITE is
used to construct three-band composite images.
RGB=bands 3,2,1
RGB=bands 4,3,2
RGB=bands 4,5,3
RGB=bands 7,4,2
Figure 3-7
Digital Filtering
One of the most intriguing capabilities of digital analysis is the ability to apply digital filters. Filters can be used to provide
edge enhancement (sometimes called crispening), to remove image blur, and to isolate lineaments and directional trends, to
mention just a few. The IDRISI module FILTER is used to apply standard filters and to construct and apply user-defined
Chapter 3 Introduction to Remote Sensing and Image Processing
29
filters.
Image Classification
Image classification refers to the computer-assisted interpretation of remotely sensed images. The procedures involved
are treated in detail in the IDRISI Guide to GIS and Image Processing Volume 2 chapter Classification of
Remotely Sensed Imagery. This section provides a brief overview.
Although some procedures are able to incorporate information about such image characteristics as texture and context,
the majority of image classification is based solely on the detection of the spectral signatures (i.e., spectral response patterns) of land cover classes. The success with which this can be done will depend on two things: 1) the presence of distinctive signatures for the land cover classes of interest in the band set being used; and 2) the ability to reliably distinguish
these signatures from other spectral response patterns that may be present.
There are two general approaches to image classification: supervised and unsupervised. They differ in how the classification is
performed. In the case of supervised classification, the software system delineates specific landcover types based on statistical characterization data drawn from known examples in the image (known as training sites). With unsupervised classification, however, clustering software is used to uncover the commonly occurring landcover types, with the analyst
providing interpretations of those cover types at a later stage.
Supervised Classification
The first step in supervised classification is to identify examples of the information classes (i.e., land cover types) of interest in the image. These are called training sites. The software system is then used to develop a statistical characterization of
the reflectances for each information class. This stage is often called signature analysis and may involve developing a characterization as simple as the mean or the range of reflectances on each band, or as complex as detailed analyses of the mean,
variances and covariances over all bands.
Once a statistical characterization has been achieved for each information class, the image is then classified by examining
the reflectances for each pixel and making a decision about which of the signatures it resembles most. There are several
techniques for making these decisions, called classifiers. Most Image Processing software will offer several, based on varying decision rules. IDRISI offers a wide range of options falling into three groups depending upon the nature of the output desired and the nature of the input bands.
hard classifiers
The distinguishing characteristic of hard classifiers is that they all make a definitive decision about the landcover class to
which any pixel belongs. IDRISI offers three supervised classifiers in this group: Parallelepiped (PIPED), Minimum Distance to Means (MINDIST), and Maximum Likelihood (MAXLIKE). They differ only in the manner in which they
develop and use a statistical characterization of the training site data. Of the three, the Maximum Likelihood procedure is
the most sophisticated, and is unquestionably the most widely used classifier in the classification of remotely sensed imagery.
soft classifiers
Contrary to hard classifiers, soft classifiers do not make a definitive decision about the land cover class to which each
pixel belongs. Rather, they develop statements of the degree to which each pixel belongs to each of the land cover classes
being considered. Thus, for example, a soft classifier might indicate that a pixel has a 0.72 probability of being forest, a
0.24 probability of being pasture, and a 0.04 probability of being bare ground. A hard classifier would resolve this uncertainty by concluding that the pixel was forest. However, a soft classifier makes this uncertainty explicitly available, for any
of a variety of reasons. For example, the analyst might conclude that the uncertainty arises because the pixel contains
more than one cover type and could use the probabilities as indications of the relative proportion of each. This is known
as sub-pixel classification. Alternatively, the analyst may conclude that the uncertainty arises because of unrepresentative
training site data and therefore may wish to combine these probabilities with other evidence before hardening the decision
IDRISI Guide to GIS and Image Processing Volume 1
30
to a final conclusion. IDRISI offers three soft classifiers (BAYCLASS, BELCLASS and FUZCLASS) and three corresponding hardeners (MAXBAY, MAXBEL, and MAXFUZ). The difference between them relates to the logic by which
uncertainty is specified—Bayesian, Dempster-Shafer, and Fuzzy Sets respectively. In addition, the system supplies a variety of additional tools specifically designed for the analysis of sub-pixel mixtures (e.g., UNMIX, FUZSIG, MIXCALC
and MAXSET).
hyperspectral classifiers
All of the classifiers mentioned above operate on multispectral imagery—images where several spectral bands have been
captured simultaneously as independently accessible image components. Extending this logic to many bands produces
what has come to be known as hyperspectral imagery.
Although there is essentially no difference between hyperspectral and multispectral imagery (i.e., they differ only in
degree), the volume of data and high spectral resolution of hyperspectral images does lead to differences in the way that
they are handled. IDRISI provides special facilities for creating hyperspectral signatures either from training sites or from
libraries of spectral response patterns developed under lab conditions (HYPERSIG) and an automated hyperspectral signature extraction routine (HYPERAUTOSIG). These signatures can then be applied to any of several hyperspectral classifiers: Spectral Angle Mapper (HYPERSAM), Minimum Distance to Means (HYPERMIN), Linear Spectral Unmixing
(HYPERUNMIX), Orthogonal Subspace Projection (HYPEROSP), Absorption area analysis (HYPERABSORB) hyperspectral classifiers. An unsupervised classifier (see next section) for hyperspectral imagery (HYPERUSP) is also available.
Unsupervised Classification
In contrast to supervised classification, where we tell the system about the character (i.e., signature) of the information
classes we are looking for, unsupervised classification requires no advance information about the classes of interest.
Rather, it examines the data and breaks it into the most prevalent natural spectral groupings, or clusters, present in the
data. The analyst then identifies these clusters as landcover classes through a combination of familiarity with the region
and ground truth visits.
The logic by which unsupervised classification works is known as cluster analysis, and is provided in IDRISI primarily by
the CLUSTER module. CLUSTER performs classification of composite images (created with COMPOSITE) that combine the most useful information bands. It is important to recognize, however, that the clusters unsupervised classification produces are not information classes, but spectral classes (i.e., they group together features (pixels) with similar
reflectance patterns). It is thus usually the case that the analyst needs to reclassify spectral classes into information classes.
For example, the system might identify classes for asphalt and cement which the analyst might later group together, creating an information class called pavement.
While attractive conceptually, unsupervised classification has traditionally been hampered by very slow algorithms. However, the clustering procedure provided in IDRISI is extraordinarily fast (unquestionably the fastest on the market) and
can thus be used iteratively in conjunction with ground truth data to arrive at a very strong classification. With suitable
ground truth and accuracy assessment procedures, this tool can provide a remarkably rapid means of producing quality
land cover data on a continuing basis.
In addition to the above mentioned techniques, two modules bridge both supervised and unsupervised classifications.
ISOCLUST uses a procedure known as Self-Organizing Cluster Analysis to classify up to 7 raw bands with the user specifying the number of clusters to process. The procedure uses the CLUSTER module to initiate a set of clusters that seed an
iterative application of the MAXLIKE procedure, each stage using the results of the previous stage as the training sites
for this supervised procedure. The result is an unsupervised classification that converges on a final set of stable members
using a supervised approach (hence the notion of "self-organizing"). MAXSET is also, at its core, a supervised procedure.
However, while the procedure starts with training sites that characterize individual classes, it results in a classification that
includes not only these specific classes, but also significant (but unknown) mixtures that might exist. Thus the end result
has much the character of that of an unsupervised approach.
Chapter 3 Introduction to Remote Sensing and Image Processing
31
Accuracy Assessment
A vital step in the classification process, whether supervised or unsupervised, is the assessment of the accuracy of the final
images produced. This involves identifying a set of sample locations (such as with the SAMPLE module) that are visited
in the field. The land cover found in the field is then compared to that which was mapped in the image for the same location. Statistical assessments of accuracy may then be derived for the entire study area, as well as for individual classes
(using ERRMAT).
In an iterative approach, the error matrix produced (sometimes referred to as a confusion matrix), may be used to identify
particular cover types for which errors are in excess of that desired. The information in the matrix about which covers are
being mistakenly included in a particular class (errors of commission) and those that are being mistakenly excluded (errors of
omission) from that class can be used to refine the classification approach.
Image Transformation
Digital Image Processing offers a limitless range of possible transformations on remotely sensed data. Two are mentioned
here specifically, because of their special significance in environmental monitoring applications.
Vegetation Indices
There are a variety of vegetation indices that have been developed to help in the monitoring of vegetation. Most are based
on the very different interactions between vegetation and electromagnetic energy in the red and near-infrared wavelengths. Refer back to Figure 3-4, which includes a generalized spectral response pattern for green broad leaf vegetation.
As can be seen, reflectance in the red region (about 0.6 - 0.7m) is low because of absorption by leaf pigments (principally
chlorophyll). The infrared region (about 0.8 - 0.9 m), however, characteristically shows high reflectance because of scattering by the cell structure of the leaves. A very simple vegetation index can thus be achieved by comparing the measure of
infrared reflectance to that of the red reflectance.
Although a number of variants of this basic logic have been developed, the one which has received the most attention is
the normalized difference vegetation index (NDVI). It is calculated in the following manner:
NDVI = (NIR - R) / (NIR + R)
where
NIR = Near Infrared
and
R
= Red
Figure 3-8 shows NDVI calculated with TM bands 3 and 4 for the same area shown in Figures 3-5, 3-6 and 3-7.
Normalized Difference Vegetation Index
-1.00
-0.88
-0.77
-0.65
-0.53
-0.41
-0.30
-0.18
-0.06
0.05
0.17
0.29
0.41
0.52
0.64
0.76
0.88
Figure 3-8
This kind of calculation is quite simple for a raster GIS or Image Processing software system, and the result has been
shown to correlate well with ground measurements of biomass. Although NDVI needs specific calibration to be used as
IDRISI Guide to GIS and Image Processing Volume 1
32
an actual measure of biomass, many agencies have found the index to be useful as a relative measure for monitoring purposes. For example, the United Nations Food and Agricultural Organization (FAO) Africa Real Time Information System (ARTEMIS) and the USAID Famine Early Warning System (FEWS) programs both use continental scale NDVI
images derived from the NOAA-AVHRR system to produce vegetation index images for the entire continent of Africa
every ten days.12
While the NDVI measure has proven to be useful in a variety of contexts, a large number of alternative indices have been
proposed to deal with special environments, such as arid lands. IDRISI offers a wide variety of these indices (over 20 in
the VEGINDEX and TASSCAP modules combined). The chapter on Vegetation Indices in the IDRISI Guide to GIS
and Image Processing Volume 2 offers a detailed discussion of their characteristics and potential application.
Principal Components Analysis
Principal Components Analysis (PCA) is a linear transformation technique related to Factor Analysis. Given a set of
image bands, PCA produces a new set of images, known as components, that are uncorrelated with one another and are
ordered in terms of the amount of variance they explain from the original band set.
PCA has traditionally been used in remote sensing as a means of data compaction. For a typical multispectral image band
set, it is common to find that the first two or three components are able to explain virtually all of the original variability in
reflectance values. Later components thus tend to be dominated by noise effects. By rejecting these later components, the
volume of data is reduced with no appreciable loss of information.
Given that the later components are dominated by noise, it is also possible to use PCA as a noise removal technique. The
output from the PCA module in IDRISI includes the coefficients of both the forward and backward transformations. By
zeroing out the coefficients of the noise components in the reverse transformation, a new version of the original bands
can be produced with these noise elements removed.
Recently, PCA has also been shown to have special application in environmental monitoring. In cases where multispectral
images are available for two dates, the bands from both images are submitted to a PCA as if they all came from the same
image. In these cases, changes between the two dates tend to emerge in the later components. More dramatically, if a time
series of NDVI images (or a similar single-band index) is submitted to the analysis, a very detailed analysis of environmental changes and trends can be achieved. In this case, the first component will show the typical NDVI over the entire series
while each successive component illustrates change events in an ordered sequence of importance. By examining these
images, along with graphs of their correlation with the individual bands in the original series, important insights can be
gained into the nature of changes and trends over the time series. The TSA (Time Series Analysis) module in IDRISI is a
specially tailored version of PCA to facilitate this process.
Other Transformations
As mentioned in earlier, IDRISI offers a variety of other transformations. These include color space transformations
(COLSPACE), texture calculations (TEXTURE), blackbody thermal transformations (THERMAL), and a wide variety of
ad hoc transformations (such as image ratioing) that can be most effectively accomplished with the image calculator utility.
12. An archive dataset of monthly NDVI images for Africa is available on CD from Clark Labs. The Africa NDVI data CD contains monthly NDVI
maximum value composite images (1982-1999), average and standard deviation of monthly NDVI images for each month over the same time period,
monthly NDVI anomaly images, and ancillary data (DEM,land use and land cover, country boundaries and coast line) for Africa in IDRISI format.
Contact Clark Labs for more information.
Chapter 3 Introduction to Remote Sensing and Image Processing
33
Conclusions
Remotely sensed data is important to a broad range of disciplines. This will continue to be the case and will likely grow
with the greater availability of data promised by an increasing number of operational systems. The availability of this data,
coupled with the computer software necessary to analyze it, provides opportunities for environmental scholars and planners, particularly in the areas of landuse mapping and change detection, that would have been unheard of only a few
decades ago.
The inherent raster structure of remotely sensed data makes it readily compatible with raster GIS. Thus, while IDRISI
provides a wide suite of image processing tools, they are completely integrated with the broader set of raster GIS tools the
system provides.
IDRISI Guide to GIS and Image Processing Volume 1
34
Idrisi System Overview
IDRISI consists of a main interface program (with a menu and toolbar system) and a collection of approximately 200 program modules that provide facilities for the input, display, and analysis of geographic and remotely sensed data. These geographic data are described in the form of map layers—elementary map components that describe a single theme. Examples
of map layers might include a roads layer, an elevation layer, a soil type layer, a remotely sensed reflectance layer and so
on. All analyses act upon map layers. For display, a series of map layers may be brought together into a map composition.
Because geographic data may be of different types, IDRISI incorporates the two basic forms of map layers: raster image layers and vector layers.13 Although IDRISI is adept at the input and display of both image and vector layers, analysis is primarily oriented towards the use of image layers. In addition, IDRISI offers a complete image processing system for remotely
sensed image data. As a result, it is commonly described as a raster system. However, IDRISI does offer strong capabilities for the analysis of vector attribute data, as well as rapid vector-to-raster conversion routines. Thus the system offers a
powerful set of tools for geographic analyses that require both types of map layers.
System Operation
The IDRISI Application Window
When IDRISI is open, the application window will completely occupy the screen (in Windows terminology, it is automatically maximized). Though not required, it is recommended that it be kept maximized because many of the dialog boxes
and images you will display will require a substantial amount of display space.14 The IDRISI application window includes
the menu, the toolbar, and the status bar.
The Menu System
The menu system is at the top of the application window. You can activate it either with the mouse or by holding the
ALT key and pressing the underlined key of the main menu entry. You can then use the mouse or arrow keys to move
around.
If you select a menu option that includes a right-pointing arrow, a submenu will appear. Clicking on a menu option without a right-pointing arrow will cause a dialog box for that module to appear.
The Tool Bar
Just below the menu is a set of buttons that are collectively known as the tool bar. Each button represents either a program module or an interactive operation that can be selected by clicking on that button with the mouse. Some of these
buttons toggle between an active and an inactive state. When active, the button appears to remain depressed after being
clicked. In these cases, the button can be released by clicking it again. You will also notice that some buttons may not
always be available for use. This is indicated when the button icon appears grey. Hold the cursor over an icon to cause the
name of the function or module represented by that icon to momentarily appear. The set of icons represent interactive
display functions as well as some of the most commonly-used modules.
13. For an in-depth discussion of raster and vector data structures, see the chapter Introduction to GIS in this volume.
14. For this reason, we recommend that the operating system taskbar be set to "autohide" so that it does not constantly occupy much-needed screen
space. To do so, go to Start/Settings/Taskbar and toggle on the autohide option.
Chapter 4 Idrisi System Overview
35
Menu
Tool Bar
Status Bar
The Status Bar
At the bottom of the screen is the status bar. The status bar provides a variety of information about program operation.
When maps and map layers are displayed on the screen and the mouse is moved over one of these windows, the status bar
will indicate the position of the cursor within that map in both column and row image coordinates and X and Y map reference system coordinates. In addition, the status bar indicates the scale of the screen representation as a Representative
Fraction (RF).15
The status bar will also indicate the progress of the most recently launched analytical operation with a graphic progress
bar as well as a percent done measure. Since IDRISI has been designed to permit multitasking of operations, it is possible
that more than one operation may be working simultaneously. To see a listing of all active processes and their status, simply double click on the progress bar panel at the bottom right of the screen. Modules may also be terminated from here.
Program Modules
Program modules may be accessed in three ways:
1.
by selecting the module in the menu structure and activating it with a click of the mouse,
2.
by selecting its program icon on the toolbar just below the menu, and
3.
by typing in or selecting the module name from the alphebetical list provided by the ShortCut utility.
15. A Representative Fraction expresses the scale of a map as a fractional proportion. Thus, an RF of 1/10,000 indicates that the image on the screen is
one ten-thousandth of the actual size of the corresponding features on the ground. IDRISI automatically determines your screen resolution and dimensions, which, in combination with the minimum and maximum x and y coordinates of the image or vector layer, are used to determine the RF for any
display window.
IDRISI Guide to GIS and Image Processing Volume 1
36
Each of these methods activates a dialog box for that module. After entering the required information for the operation
to be performed and clicking on the OK button, the program module will run.
Data Paths and Projects
Program modules act upon data—map layers and tabular data stored in data files. These files are stored in folders (also
known as directories) and sub-folders (sub-directories) on drives in the computer. The location of any specific data file is
thus designated by a name consisting of the file name plus the drive, folder and sub-folder locations. Collectively, these
are known as a data path (since they specify the route to a particular collection of data). For example, "c:\massachusetts\middlesex\census_tracts.vct" might designate a vector layer of census tracts contained in the Middlesex County
sub-folder of the Massachusetts folder on hard disk drive C.
IDRISI can work with files from any data path (including network paths). However, it can often be a nuisance to continually specify folder and sub-folder names, particularly when a specific project typically accesses data in a limited set of
folders. To simplify matters, IDRISI allows you to specify these data paths as belonging to a specific project. When the
paths to these folders are designated in a Project File, IDRISI finds data very easily and with minimal user intervention.
The Project Working Folder
Within any Project File, the most important folder is the Working Folder. Users may choose to store all of the data for a
project in the Working Folder (particularly for smaller projects). However, for larger projects, or projects wishing to
develop libraries of carefully organized and protected data sets, additional Resource Folders can also be added to the Project
File.
While the user is always able to specify any location for input or output data, designating the most commonly-used folders
in a Project File facilitates the use of IDRISI. For instance, if input file names are given without a path designation,
IDRISI automatically looks in the Working Folder first, then in each Resource Folder in turn to locate the file. Similarly,
if no alternative path is specified, output data are automatically written to the Working Folder.
Project Resource Folders
Resource Folders contain data that can be accessed quickly through the IDRISI pick list (see below). A Project File can
contain any number of Resource Folders. To the user, the Working Folder and all Resource Folders function as a single
project folder. For example, one might ask IDRISI to display a raster map layer named "landuse". If no folder information was supplied, IDRISI first looks for the file in the Working Folder, and then each Resource Folder in turn, ultimately
displaying the first instance it finds. This is very convenient as it allows the user to enter file names without paths. However, if duplicate file names exist in separate folders that are part of the same project, the user must exercise caution and
remember the order in which IDRISI accesses the project folders.
Setting the Data Paths
The data paths for the Working and Resource Folders can be set in the Data Paths option of the File menu. You can also
do so by clicking the leftmost icon on the toolbar to access the Project Environment dialog.
Working with IDRISI Dialog Boxes
When any module is activated, a dialog box appears with information on data or option choices that are required for the
module to run. Virtually all input boxes must be filled with information before the module can be run. Only the title and
measurement units input boxes under the Output Documentation button can be left blank. (It is recommended, however,
Chapter 4 Idrisi System Overview
37
that these boxes also be filled.) In some cases, input boxes already contain values, or will fill in with values as other information is entered. These are default values that may be freely edited. In those cases where a set of choices should be
made, the most common settings are normally pre-selected. These should be examined and changed if necessary. The system will display an error message in cases where a needed element of data has been left out.
Pick Lists
Whenever a dialog box requires the name of an input data file (such as a raster image), you have two choices for input of
that value. The first is simply to type the name into the input box. Alternatively, you can activate a pick list of choices by
either double clicking into the input box, or clicking the pick list button to the right of the input box (the small button with
the ellipses characters "..."). The pick list is organized as a tree directory of all files of the required type that exist in the
Working and Resource Folders. The Working folder is always at the top of the list, followed by each Resource Folder in
the order in which they are listed in the Project Environment. The files displayed in the pick list depend upon the particular input box from which the pick list was launched. All IDRISI file types are designated by unique file name extensions.
If an input box requires a raster image, for example, the pick list launched from that input box will display only those files
that have an .rst extension.
You may select a file from the pick list by first highlighting it with a click of the left mouse button, and then selecting it by
clicking the OK button or pressing the Enter key. Alternatively, double clicking the highlighted entry will also cause it to
be selected. The pick list will then disappear, and the name of that file will appear in the input box. Note that you can also
choose a file that is not in the current project environment by clicking on the Browse button at the bottom of the pick list.
Output File Names
You will normally want to give output file names that are meaningful to you. It is common to generate many output files
in a single analysis and descriptive file names are helpful for keeping track of these files.16 File names may be long and can
contain spaces and most keyboard characters (all except / \ : * ? " < > and |). It is not necessary for the user to specify
the three-letter file name extension as IDRISI takes care of this. A full path may be entered in the output file name box. If
16. The documentation file for output raster and vector data files includes in the lineage field the command line from which the layer was created. This
may be viewed with the Metadata function.
IDRISI Guide to GIS and Image Processing Volume 1
38
no path is entered, the output file will automatically be written to the working folder that is specified in the project environment.
Clicking on the pick list button to the right of the output file name box will bring up a standard Windows Save As file
name dialog box. Here the user may select any folder for the output file and may also enter the file name. When the pick
list button is used, an automatically-generated output file name will be entered by default. To change this name, simply
click into the file name box and type the desired name.
As stated above, users typically give descriptive file names to help avoid confusion as the database grows. However,
IDRISI can also generate automatic names for output data files. Automatic output names are intended for the quick naming of temporary files.
Automatic output file names begin with a user-defined three-letter prefix (set to TMP by default),17 followed by a three
digit number. By default, filenames begin with TMP001 and progress up to TMP999. As these files are normally temporary, overwrite protection (see below) is disabled for files with names that begin with the designated prefix followed by
three digits. Note that after exiting IDRISI, the cycle will begin again with TMP001 the next time IDRISI is used. Since
the numbering sequence starts from 001 with each new session, the data in such files is likely to be lost in a subsequent
session unless it is intentionally saved under a new name.
To use the autoname feature, move to the output file name box concerned, and either double click in the box or click the
pick list button. If the double click is used, an automatic file name will be generated and placed into the output file name
box. Since no path is specified with this method, output will automatically go to the Working Folder.
In the case where the pick button is clicked beside an output file name box, a standard Windows Save As dialog box will
appear with the temporary name (e.g., TMP001) already selected.
Overwrite Protection
By default, IDRISI checks if the given output file already exists. If it does, it will ask whether you wish to overwrite it. If
not, you will be returned to the dialog box where you can enter a different name.
17. The automatically-generated output file prefix may be changed in the User Preferences option of the File menu.
Chapter 4 Idrisi System Overview
39
There are three exceptions to this logic. The first is when an automatic output name is generated (see the section immediately above). The second is when the user removes overwrite protection in the User Preferences dialog under the File
menu. And the third is when modules are run in macro mode.
Getting Help
While using IDRISI, help is always close at hand. IDRISI contains an extensive on-line Help System that can be accessed
in a variety of ways. The most general means of accessing help is to select it from the main IDRISI menu. However, each
program module also contains a help button. In this case, help is context sensitive, accessing the section in Help associated with that particular module. In either case, it is worth becoming familiar with the many options the Help System provides. To do so, select the Using Help option from the main Help menu.
IDRISI Guide to GIS and Image Processing Volume 1
40
Map Layers, Collections and Data Structures
Map Layers
Map layers are the fundamental units of display and analysis in GIS. A map layer is an elementary theme, a single phenomenon that can be mapped across space. Examples would include a land use layer, a roads layer, an elevation layer, a soils
layer, and so on. Map layers, then, are somewhat different from traditional maps. Traditional maps usually consist of several map layers. For example, a topographic map sheet would typically consist of a contour layer, a forest cover layer, an
administrative boundaries layer, a roads layer and a settlements layer. In IDRISI, this same concept is applied. A map is a
graphic representation of space that is composed of one or more map layers using a tool called Composer.
This breakdown of the map into its elementary constituents offers important advantages. Clearly it allows for the simple
production of highly customized maps; we simply pull together the layers we wish to see together. However, the more
important reason for storing data this way is that layers are the basic variables used by the GIS in analytical modeling.
Thus, for example, we might create a map of soil erosion as a mathematical function of soil type, slope, land cover, and
precipitation layers. Indeed, the breakdown of geographic data into layers allows for the production of an extraordinary
range of mathematical and logical models.
Map Layer Types
As detailed in the Introduction to GIS chapter, there are two basic types of layers in GIS: raster image layers, and vector
layers.18 Some systems deal exclusively with one or the other type. IDRISI incorporates both since they each have special
advantages. However, while they have equal stature in terms of display and database query, IDRISI offers a far broader
range of analytical operations for raster layers. This is partly because of IDRISI's historic development from an almost
purely raster system and partly because the range of analytical operations possible in GIS is far greater with raster layers
(as a consequence of their simplicity and predictable regularity).
All IDRISI raster layers have the same basic structure. Sub-types exist only in terms of the numeric data type that is used
to record cell values. As one can easily imagine, raster images are high in data volume. Thus small changes in the underlying data type can make huge differences in storage requirements. IDRISI supports raster images stored with byte, integer,
and real numbers as well as two special formats for the storage of full-color images.
Vector layers describe the location and character of distinct geographic features. Sub-types include Point, Line, Polygon
and Text layers. Point features are phenomena that can be described by a single point, such as meteorological data or (at
small scales) town locations. As the name suggests, line vector layers describe linear features such as rivers and roads. The
term polygon may be less familiar. Polygons refer to areas. Because the boundaries of the areas are made up of straight line
segments between points, the figure formed is technically a polygon. The term polygon has thus become part of the terminology of GIS. Finally, text layers are used to describe the location and orientation of text labels.
Map Layer Names
All analytical modules in IDRISI act on map layers. Thus the input and output file names that are given in IDRISI operations are typically those of map layers. In your normal use of IDRISI, map layers are specified with simple names that do
not need to express their data path (the IDRISI Project Environment defines the paths to your working and resource
folders). It is also not necessary to specify their file name extensions (since IDRISI looks for and creates specific extensions automatically).
18. In IDRISI, the terms raster layer, raster image layer and image are all used interchangeably.
Chapter 5 Map Layers, Collections and Data Structures
41
Consistent with the 32-bit variants of the Windows operating system, layer names may contain any letter of the alphabet
as well as numerals, spaces and most symbols (excluding \ / : * ? " < > and |). For example, layers might have names
such as "soils" or "landuse 1996." In principle, file names can be up to 255 characters in length. However, in practical use,
it is recommended that they be only as long as necessary so the entire file name will be visible in the pick lists and dialog
input boxes.
In normal use, no distinction is made between image and vector layer names. However, their type is always known by the
system—by the context in which they were created and by the file extension that the system uses internally to store each
layer. Thus a raster image layer named "soils" would be stored as "soils.rst" internally, while a vector layer named "soils"
would be stored as "soils.vct". As a result, it is possible to have raster and vector layers of the same name. However, most
users of IDRISI allow this to occur only in cases where the two files store vector and image manifestations of the same
underlying data.
When file names are input into IDRISI dialog boxes, the system always knows the appropriate layer type in use. Indeed,
only those files of the appropriate type for that operation will display in the pick list.
Map Layer Data Files
While we refer to a map layer by a simple name, it is actually stored by IDRISI as a pair of data files. One contains the spatial data, while the second contains information about those data (i.e., documentation or metadata). Thus, while raster
images are stored in data files with an ".rst" extension, they each have an accompanying documentation file with an ".rdc"
extension (i.e., raster documentation). Similarly, while vector data files have a ".vct" extension, each has a companion documentation file with a ".vdc" extension (i.e., vector documentation). In use, however, you would refer to these layers by
their simple layer names (i.e., "soils" and "districts" respectively). The contents of data documentation files are described
in detail below in the section on file structures.
Map Layer Collections
Sometimes it makes sense to group map layers together into collections. A good example is when you have a set of vector
features associated with a data table. For instance, a vector file might be created of census tracts within a city. Each of
these tracts possesses multiple attributes such as median income, population density, median educational attainment, and
so on. Each of these attributes can be linked to a vector file defining the geography of the census tracts and displayed as a
map layer. Thus the combination of the vector file and the entire database table constitutes a collection of map layers.
IDRISI recognizes two kinds of collections. One is a collection produced by linking a geographic or feature definition
vector file (such as the digital representation of census tracks mentioned above) with a database table.19 This is called a
linked-table collection and is established with a vector link file. The second is simply a collection of independent raster
layers into an associated group which is defined by a raster group file. In both cases, all the elements of the collection
must reside in the same folder.
Link Files
Link files associate a feature definition vector file with a database table. As a consequence, each of the fields (columns) of
data in the data table becomes a layer, with the entire set of fields producing a linked-table collection. As you will see,
IDRISI has special features for the display and analysis of relationships between layers in such a collection. All that is
required is a link file that can establish four important properties of the collection:
1.
the feature definition (i.e., spatial frame) vector file;
19. Feature definition vector files and image files describe the geography of features but not their attributes. Rather, the features have identifier values
that are used to associate attributes stored in database tables or attribute values files.
IDRISI Guide to GIS and Image Processing Volume 1
42
2.
an associated attribute database;
3.
a specific table within that database; and
4.
field).
the database field that contains the identifiers which can associate database records with vector features (the link
The Collection Editor is available on the toolbar and under the File menu to help construct these link files. Only vector
point, line or polygon layers can be specified as a feature definition vector file. The database files and associated tables
must be Microsoft Access Database Engine files with an ".mdb" extension.
Group Files
The second type of collection is a raster group file. As the name suggests, a group file is simply a listing of a set of independent layers that one wishes to associate together. For example, a group file might consist of the seven bands of imagery that make up a Landsat satellite image. Many of the image processing procedures in IDRISI allow you to specify a
group file as input rather than each individual band to be used. This saves a lot of time when the same group of images is
used in several procedures. However, there are more compelling reasons for creating group files than as a simple means
for specifying collections of files.
1.
When more than one member of a group is displayed on the screen, you have the option to treat them identically during zoom and pan operations. Thus when one member is zoomed or panned, all other members of that group
that are on the screen will zoom or pan simultaneously.
2.
When the feature properties option is activated, queries about the data values for a specific layer will yield a table (or
optionally, a graph) of the values for all attributes in the collection at that location.
Again, the Collection Editor (from the toolbar or the File menu) can be used to facilitate the creation of group files. Note
that there is also a special variant of a group file known as a time series file that is used for time series analysis. Signature
files used in image classification may also be collected together in signature group files and hyperspectral signature group files.
Collection Member Naming Conventions
To specify a layer that belongs to a collection, one uses the "dot" convention—i.e., the reference is a combination of the
collection name and the layer name, separated by a dot. For example, suppose one had a link file named
CensusTractData99 associating a feature definition vector file with a database table of census statistics. Each of the layers
in the collection would be referenced by a combination of the collection name and the appropriate field name in the table.
Thus the Median_Income layer would be referred to as:
CensusTractData99.Median_Income
Similarly, one might create a group file of the seven multispectral bands in a Landsat image. Perhaps the bands have
names such as OahuBand1, OahuBand2, etc. If the group file that was created to associate them was named LandsatTM,
then a reference to the second band would be as follows:
LandsatTM.OahuBand2
Collections and Pick Lists
The pick lists that IDRISI constructs to facilitate the selection of input files are collection-aware. For example, when a raster
image is required, the pick list will display raster group files as well as raster images. Collections are immediately evident in
a pick list by the "+" sign at their point of connection to the directory tree. If you click on the collection file, it will expand
to show each of the members. These members are shown with their simple names. However, if you choose one, IDRISI
will enter its full reference (using the dot convention) into the input box. In some cases, a group file name may itself be
selected as input for a dialog box.
Chapter 5 Map Layers, Collections and Data Structures
43
Attribute Files
An attribute file contains information about the characteristics of features, but not the geographic definition of those features. IDRISI supports two forms of attribute files, database tables and ASCII values files.
Data Tables
IDRISI incorporates the Microsoft Access Database Engine. Thus IDRISI attribute tables have ".mdb" extensions, and
can also be used and modified with Microsoft Access or any other Access-compatible system. The linked layer collection
can only be created with this type of attribute file. Microsoft Access Database files may contain multiple tables, each one
of which can be linked to form a layer collection. See the chapter Database Workshop for more information about using
database tables with IDRISI.
Values Files
The second form of attribute file that IDRISI uses is an ASCII values file. This has a very simple file structure that contains the values for a single attribute. It is stored in ASCII text format and consists of two columns of data separated by
one or more spaces. The first column contains an identifier that can be used to associate the value with a feature (either
raster or vector), while the second column contains the attribute value. Attribute values files have an ".avl" extension.
Some modules create these as output files while others require (or can use) them as input for analyses. Attribute values
files may be imported or exported to or from a database table using Database Workshop. In addition, IDRISI provides a
procedure (ASSIGN) to assign those values to a set of features included in a raster image or vector file, and a companion
procedure (EXTRACT) to create a new values file as a statistical summary of the values in another image for the features
of a feature definition image.
Import/Export Formats
IDRISI recognizes xBASE table formats strictly for purposes of moving data to and from IDRISI.
Map Layer File Structures
Raster Layers (.rst)
Raster images are the most fundamental and important data type in IDRISI. They are automatically assigned an ".rst" file
extension by the system. Raster layers are both simple in structure and regular in their organization, allowing an extraordinary range of analytical operations. The data structures that IDRISI uses for storing raster images are optimized for simplicity and efficiency.
Raster images define a rectangular region of space by means of a fine matrix of numeric data values that describe the condition or character of the landscape in each cell of a fine grid. These numeric grid cell values are used not only for analysis,
but also for display. By assigning specific colors (in a palette) to designated numeric ranges, a very fine matrix-like color
image is formed (so fine that typically the individual cells cannot be seen without zooming in to a very large scale).
In this grid-cell structure used by IDRISI, rows and columns are numbered starting from zero. Thus an image of 1000
columns and 500 rows has columns numbered from 0-999 and rows numbered from 0-499. Unlike the normal Cartesian
coordinate system, the 0,0 cell is in the top left-hand corner. Cells are numbered left-to-right as you might expect, but
rows are numbered top-to-bottom. This is common with most raster systems because of the directionality of many output devices, particularly printers, that print from top to bottom.
While the logical structure of an image file is a grid, the actual structure, as it is stored, is a single column of numbers. For
instance, an image consisting of 3 rows by 5 columns is stored as a single column of 15 numbers. It is the image's docu-
IDRISI Guide to GIS and Image Processing Volume 1
44
mentation file that allows IDRISI modules to reconstruct the grid from this list. An image that looks like this:
10
15
9
10
1
14
10
11
14
13
11
10
has an image file that looks like this (if one could see it):
10
15
9
10
1
14
10
11
14
13
11
10
The raster documentation file, containing the number of rows and columns, allows the image to be correctly recreated for
display and analysis.
As mentioned earlier, the major sub-types of raster images are differentiated on the basis of the data types that are used to
represent cell values. IDRISI recognizes five raster data types: integer, byte, real, RGB8, and RGB24.
1.
Integers are numbers that have no fractional part and lie within the range of -32768 to +32767. Integer files are
sometimes called 16-bit integer files since they require 16 bits (i.e., 2 bytes) of memory per value (and therefore per pixel).
Integer values can be used to represent quantitative values or can be used as codes for categorical data types. For example,
a soils map may record three soil types in a particular region. Since IDRISI images are stored in numeric format, these
types can be given integer codes 1, 2 and 3. The documentation file records a legend for these, on the relationship
between the integer codes and the actual soil types.
2.
Byte values are positive integer numbers ranging from 0 to 255. The byte data type is thus simply a subrange of
the integer type. It is used in cases where the number range is more limited. Since this more limited range only requires 8
bits of memory per value for storage (i.e., 1 byte), only half as much hard disk or memory space for each image is needed
compared to integer. This data type is probably the most commonly used in GIS since it provides an adequate range to
describe most qualitative map data sets and virtually all remotely-sensed data.
3.
Real numbers have a fractional part such as 3.14. Real numbers are used whenever a continuous (rather than discrete) data variable is being stored with great precision, or whenever the data range exceeds that of the integer data type.
The real data type of IDRISI raster images is known as single precision real numbers. Thus it can store values within a
range of ± 1 x 1038 with a precision of 7 significant figures. Single precision values provide a nice balance between range,
Chapter 5 Map Layers, Collections and Data Structures
45
precision and data volume. Each number (and thus each pixel) requires 4 bytes of storage. Note that real numbers have
no discrete representation (e.g., 4.0 is the same as 3.99999999999999999999 at some level of precision). Thus, while it is
possible to store whole numbers in a real data type, they cannot then be used as feature identifiers (e.g., for ASSIGN or
EXTRACT) or in modules that require byte or integer values (e.g., GROUP).
4.
RGB8 data files are simply byte binary files, but their values have special meaning. They are given this special
designation to indicate that they contain color codes (in a 0-255 range) for a specially-encoded color image that requires a
special palette (the "Composite" palette) for display. They are constructed, using the COMPOSITE module, from three
separate images which are assigned to the Red, Green and Blue additive primaries to yield a full color image using only 8
bits per pixel. In most cases, RGB24 is a preferable format for the display of full color images. However, the RGB8 file
format is required input for the CLUSTER and ISOCLUST unsupervised image classification modules.
5.
RGB24 data files use 24 bits (3 bytes) for each pixel to encode full color images. The internal structure is bandinterleaved-by-pixel. Compared to RGB8 images which can represent only 256 colors, RGB24 images can encode over 16
million separate colors. RGB24 images are also constructed using the COMPOSITE module (in the Display menu). The
IDRISI display system allows for interactive and independent contrast manipulation of the three input bands of a displayed RGB24 image. They are the preferred format for the display of full color images.
The documentation file associated with each image records the file data type. When a new image is created, it maintains
the prevailing data type of the input image, or it produces a data type that is logical based upon standard mixed arithmetic
rules. Thus, dividing one integer data image by a second integer data image yields a real data image. IDRISI accepts integer, byte and real data types for almost all operations that would logically allow this. Some modules, though, do not make
sense with real data. For example, the GROUP operation that extracts contiguous groups can only do so for categorical
data coded with integer numbers. If the data type in an image is incorrect for a particular operation, an error message will
be displayed.
The CONVERT module can be used at any time to convert between the integer, byte, and real data types. In the case of
conversion from real numbers to either the integer or byte formats, CONVERT offers the option of converting by
rounding or by truncation.
The data type described above indicates the type of numbers stored in an image. Another parameter, the file type, indicates how these numbers are stored. As with the data type, the file type of an image is recorded in its documentation file.
Image files may be stored in ASCII, binary or packed binary formats, although only the binary form is recommended (for
reasons of efficiency). Binary files are those which are stored in the native binary encoding format of the operating system
in use. In the case of IDRISI, this will be one of the Windows operating system variants. Thus, for example, the binary
coding of a real number is that adopted by Windows and the Intel hardware platform.
Binary files are efficient in both hard disk space utilization and processing time. However, the format is not universal and
not always very accessible. As a consequence, IDRISI also provides limited support for data recorded in ASCII format. A
file in ASCII format is also referred to as a text file, and can be viewed directly with any text editor (such as the Edit module in IDRISI). The ASCII file type is primarily used to transfer files to and from other programs, since the coding system
is a recognized standard (ASCII = American Standard Code for Information Interchange). ASCII files are not an efficient
means of storing data and they cannot be displayed. You should therefore convert your files to binary format before getting too involved with your analyses. The CONVERT module converts files from ASCII to binary or binary to ASCII.
Although binary files cannot be examined directly, IDRISI provides strong data examination facilities through the IDRISI
File Explorer on the File menu. Binary files may be viewed as a matrix of numbers using the View Structure utility of the
File Explorer. There is rarely a need then to convert images to ASCII form.
The packed binary format is a special data compression format for binary integer or byte data, known as run-length
encoding. It played a special role in earlier MS-DOS versions of IDRISI before file compression was available at the operating system level. However, it is of limited use now that Windows takes an active role in data compression. Its support is
largely for purposes of backward compatibility with earlier versions of IDRISI. Like ASCII files, most IDRISI modules
still support the packed binary data type. However, also like ASCII files, packed binary files cannot be directly displayed.
IDRISI Guide to GIS and Image Processing Volume 1
46
Thus the use of this file type is also discouraged. As with other conversions, converting raster images to or from packed
binary is undertaken with the CONVERT module.20
Raster Documentation Files (.rdc)
Each of the primary file types used by IDRISI (Raster, Vector and Attribute) is associated with a companion documentation file. Raster documentation files are automatically assigned an ".rdc" file extension by IDRISI.
Documentation files are always stored in ASCII format. They can be viewed and modified using the Metadata utility from
the toolbar icon or File menu.21 After specifying the file type you wish to view, you can select a file of that type by choosing it from the file list. The contents of its documentation file will then be displayed.
The documentation file consists of a series of lines containing vital information about the corresponding image file. The
first 14 characters describe the contents of the line, while the remaining characters contain the actual data. For example,
the documentation file for a soils image (soils.rst) might look like this:
file format
: IDRISI Raster A.1
file title
: Major Soils Groups
data type
: byte
file type
: binary
columns
: 512
rows
: 480
ref. system
: US83TM18
ref. units
: m
unit dist.
: 1
min. X
: 503000
max. X
: 518360
min. Y
: 4650000
max. Y
: 4664400
pos'n error : unknown
resolution
: 30
min. value
: 0
max. value
: 3
display min : 0
display max : 3
value units
: classes
value error
: 0.15
flag value
: 0
20. There are some important considerations in determining whether the packed form will actually reduce data storage requirements. Primarily, the packed
binary format will only save space where there are cells with a frequency of identical values next to one another. See the CONVERT module description in the on-line
Help System for some important tips in this respect.
21. The Metadata utility replaces the modules DOCUMENT and DESCRIBE in earlier versions of IDRISI.
Chapter 5 Map Layers, Collections and Data Structures
47
flag def'n
: background
legend cats : 3
code 1
: Podzol Soils
code 2
: Brown Podzolic Soils
code 3
: Gray-Brown Podzolic Soils
lineage
: Soil polygons derived from 1:5000 scale color air photography
lineage
: and ground truth, with the final compilation being adjusted to the
lineage
: map base by hand.
comment
: Value error determined by statistical accuracy assessment
comment
: based on a stratified random sample of 37 points.
This file contains information on major soils groups and is in byte format stored as a binary file. The image contains 512
columns and 480 rows, for a total of 245,760 values. The image is georeferenced with a reference system named
US83TM1822 with each coordinate unit representing 1 meter. The minimum and maximum X and Y coordinate values
indicate the reference system coordinates of the left, right, top and bottom edges.
The position error indicates how close a feature's actual position is to its mapped position in the image. It is marked in the
example as unknown. If known, this field should record the RMS (Root Mean Square) error of locational references
derived from the bounding rectangle coordinates. This field is for documentation purposes only and is not currently used
analytically by any module.
The resolution refers to the inherent resolution of the image. In most cases, it should correspond with the result of dividing the range of reference coordinates in X (or Y) by the number of columns (or rows) in the image. However, there are
some rare instances where it might differ from this result. A good example is the case of Landsat Band 6 (Thermal) imagery. The resolution of these data is actually 120 meters and would be recorded as such in this field. However, the data is
distributed in an apparent 30 meter format to make them physically match the dimensions of the other bands in the
image. This is done by duplicating each 120 meter pixel 4 times in X and then each row 4 times in Y. The resolution field
is a way of correctly indicating the underlying resolution of these data.
In most images, the resolution in X and Y will be equal (i.e., pixels will be square). However, when this is not the case, the
coarser resolution is recorded in the documentation file.23 For example, with Landsat MSS data, the pixel resolution is 80
meters in X and 60 meters in Y. In this case, the documentation file would show a resolution value of 80. Rectangular pixels will display correctly, however, preserving the true resolution in both X and Y. In addition, all analytical modules using
measures of distance or area calculate resolution in X and Y independently. They therefore do not rely on the resolution
value recorded in the documentation file.
The minimum and maximum value fields record the actual range of data values that occurs in image cells, while the display min and display max values designate what are sometimes called the saturation points of the data range. Values less than
or equal to display min are assigned the lowest color in the palette sequence, while values greater than or equal to display
max are assigned the highest color in the palette sequence. All values in between are assigned palette colors according to
their position in the range (i.e., with a linear stretch). It is very common for the display min and max to match the min and
max values. However, in cases where one wishes to alter the brightness and contrast of the image, altering the saturation
22. The reference system indicated in the documentation file is the name of a reference system parameter file (.ref). The parameters of this file are
detailed in the chapter Georeferencing in this volume.
23. Earlier versions of IDRISI calculated and reported the resolution field as that of the X dimension only.
IDRISI Guide to GIS and Image Processing Volume 1
48
points can produce a significant enhancement.24 Saturation points may be manipulated interactively in the display system
(see the chapter Display System in this volume).
The value units read classes in this example to indicate that the numbers are simply qualitative codes for soil classes, not
quantitative values. The value units field is informational only, therefore any description can be used. It is suggested that
the term classes be used for all qualitative data sets. However, when standard linear units are appropriate, use the same
abbreviations that are used for reference units (m, ft, mi, km, deg, rad).
The value error field is very important and should be filled out whenever possible. It records the error in the data values
that appear in image cells. For qualitative data, this should be recorded as a proportional error. In the example, the error is
recorded as 0.15, indicating cell accuracy of 85% (i.e., what is mapped is expected to be found on the ground 85% of the
time). For quantitative data, the value here should be an RMS error figure. For example, for an elevation image, an RMS
error of 3 would indicate 68% of all values will be within ± 3 meters of the mapped value, that approximately 95% will be
within ± 6 meters, and so on. This field is analytical for some modules (e.g., PCLASS) and is intended to be incorporated
into more modules in the future.
The flag value and flag definition fields can be used to indicate any special meaning that certain cell values might carry.
The flag value indicates the numeric value that indicates a flag while the flag definition describes the nature of the flagged
areas. The most common data flags are those used to indicate background cells and missing data cells. These flag definition field entries are specifically recognized and used analytically by some modules (e.g., SURFACE). Other terms would
be considered only informational at this stage. Note that some modules also output data flags. For example, when SURFACE is used to create an aspect image, aspects derived from a slope of zero are flagged with a -1 (to indicate that the
aspect cannot be evaluated). In the output documentation file, the flag value field reads -1 while the flag definition field
reads "aspect not evaluated, slope=0".
The legend cats field records the number of legend captions that are recorded in the file. Following this, each legend caption is described, including the symbol code (a value from 0-255) and the text caption.
Finally, the image documentation file structure allows for any number of occurrences in four optional fields: comment,
lineage, consistency and completeness. At present, these fields are for information only and are not read by IDRISI modules (although some modules will write them). Note that the lineage, consistency and completeness fields are intended to
meet the recommendations of the U.S. National Committee for Digital Cartographic Data Standards (NCDCDS). Along
with the positional (pos'n) error and value error fields, they provide a means of adhering to the current standards for
reporting digital cartographic data quality.25 Multiple occurrences of any of these field types can occur (but only at the end of
the file) as long as they are correctly indicated in the 14 character descriptive field to the left. The lineage field can be used
to record the history of an image. The command line used to generate a new file is recorded in a lineage field of the documentation file. The user may add any number of additional lineage lines. The consistency field is used to report the logical
consistency of the file; it has particular application for vector files where issues of topological errors would be reported.
The completeness field refers to the degree to which the file comprehensively describes the subject matter indicated. It
might record, for example, the minimum mapping unit by which smaller features were eliminated. Finally, the comment
field can be used for any informational purpose desired. Note that the Metadata utility and CONVERT module both read
and maintain these optional fields. Metadata also allows one to enter, delete or update fields of this nature (as well as any
documentation file fields).
Vector Layers (.vct)
IDRISI supports four vector file types: point, line, polygon and text.26 All are automatically stored with a ".vct" file exten24. Note that one would normally only change the saturation points with quantitative data. In the example illustrated here, the numeric values representative qualitative classes. In these cases, the saturation points should match the actual minimum and maximum values.
25. For further information about the NCDCDS standard, refer to the January 1988 issue of The American Cartographer, Volume 15, No. 1.
26. Note that the binary vector file structures for Idrisi32 are different from those of earlier versions. Consult the on-line Help System for details.
Chapter 5 Map Layers, Collections and Data Structures
49
sion. They are stored in a feature encoded structure in which each feature is described in its entirety before the next is
described. The File Structures section of the on-line Help System provides specific details about the structures of each
vector file type. In addition, vector files may be viewed as ascii representations using the View Structure feature of the
IDRISI File Explorer. However, the following descriptions provide all the information that most users will require.
Common Features of All Vector Files
All vector files describe one or more distinct features. Unlike raster images that describe the totality of space within a rectangular region, vector files may describe only a small number of features within a similarly defined rectangular region.
Each feature is described by means of a single numeric attribute value and one or more X,Y coordinate pairs that describe
the location, course or boundary of that feature. These points will be joined by straight line segments when drawn. Thus,
curved features require a great number of closely spaced points to give the appearance of smooth curves.
The numeric attribute values can represent either identifiers (to link to values in a data table) or actual numeric data values, and can be stored either as integer or real numbers (although identifiers must be integers).27
A special feature of vector files is that their data types (integer or real) are less specific in their meaning than raster files.
The integer type has a very broad range, and is compatible with both the byte and integer type of raster layers as well as
the long integer type commonly used for identifiers in database tables. Very approximately, the vector integer type covers
the range of whole numbers from -2,000,000,000 to +2,000,000,000. Similarly, the vector real data type is compatible with
the single precision real numbers of raster layers, but is in fact stored as a double precision real number with a minimum
of 15 significant figures. IDRISI's conversion procedures from vector to raster handle these translations automatically, so
there is little reason to be concerned about this detail. Simply recognize that integer values represent whole numbers while
real numbers include fractional parts.
Point Files
Point files are used to represent features for which only the location (as a single point location designation) is of importance. Examples include meteorological station data, fire hydrants, and towns and cities (when their areal extent is not of
concern). Each point feature is described with an attribute value that can be integer or real, and an X,Y coordinate pair.
Line Files
Line files describe linear features such as rivers or roads. Each line feature in a layer is described by an attribute value that
can be integer or real, a count of the number of points that make up the line, and a series of X,Y coordinate pairs (one for
each point).
Polygon Files
Polygon files describe areal features such as forest stands or census tracts. Each polygon feature in a polygon layer is
described by an attribute value that can be integer or real, a count of the number of parts in the polygon, and for each part,
a list of the points (by means of an internal index) that make up that part. This is then followed by an indexed list of the
X,Y coordinate pairs of all points in the polygon. The parts of a polygon are concerned with the issue of holes. A polygon
with one part has no holes, whereas one with two parts has one hole. The first part listed is always the polygon which
encloses the holes.28
27. All features in a single vector layer should be encoded with numeric values of the same type. Thus, if integer identifiers are used, all features must
have integer identifiers. Similarly, if real number attributes are encoded, all features in that layer must have real number attributes.
28. Note that islands are treated as separate polygons by IDRISI. Thus, if one were to take the case of the State of Hawaii as an example, one might consider it to be a single polygon with five parts. In IDRISI, though, these would be five separate polygons. However, they can each have the same identifier allowing them to be linked to a common data entry in a database table.
IDRISI Guide to GIS and Image Processing Volume 1
50
Text Files
Text vector files represent text captions that can be displayed as a layer on a map. They store the caption text, their position and orientation, and a symbol code that can be used to link them to a text symbol file. Text vector files can be created
with the on-screen digitizing feature in IDRISI or by exporting text from a CartaLinx coverage. Text symbol files are created with Symbol Workshop.
Vector Documentation Files (.vdc)
As with image files, all vector files are paired with documentation files. Vector documentation files have a ".vdc" extension. Any IDRISI module that creates or imports a vector file will automatically create a documentation file. The Metadata utility can be used to update or create documentation files as required.
A sample vector documentation file appears as follows:
file format
:
file title
: Land Use / Land Cover
id type
: integer
file type
: binary
IDRISI Vector A.1
object type : polygon
ref. system
: utm16spe
ref. units
: m
unit dist.
: 1
min. X
: 296000
max. X
: 316000
min. Y
: 764000
max. Y
: 775000
pos'n error : unknown
resolution
: unknown
min. value
: 1
max. value
: 9
display min : 1
display max : 7
value units
: classes
value error
: 0.15
flag value
: 9
flag def'n
: Unknown: Obscured by Clouds
legend cats : 7
code 1
: Residential
code 2
: Industrial
code 3
: Commercial
code 4
: Other Urban or Built Up Land
code 5
: Open Water
Chapter 5 Map Layers, Collections and Data Structures
51
code 6
: Barren
code 7
: Transitional
As can be seen, the structure of a vector documentation file is virtually identical to that for a raster layer. The only differences are:
1.
the row and column information is absent;
2.
data type has been replaced by id type, although the intention is identical. Choices here are integer or real. The reason for the slightly different wording is that coordinates are always stored as double precision real numbers. The id's
though can vary in their data type according to whether they encode identifiers or numeric attribute values.
3.
the file type is always binary. (ASCII format for vector files is supported through the Vector Export Format.)
4.
the minimum and maximum X and Y values recorded in the documentation file do not necessarily refer to the
minimum and maximum coordinates associated with any feature but to the boundary or limits of the study area. Thus
they correspond to the BND (boundary) coordinates in vector systems such as Arc/Info.
Attribute Files (.mdb and .avl)
In addition to the attributes stored directly in raster or vector layers, IDRISI permits the use of freestanding attribute files.
Two types are recognized, data tables and values files. Only the former can be linked to a vector file for display or to produce a vector collection. The latter may be used to assign new values to a raster image and will be produced when summary values for features are extracted from a raster image. Fields from data tables may be exported as values files and
values files may be imported into data tables using Database Workshop.
Data Tables (.mdb)
IDRISI data tables are Microsoft Access-compatible relational database files. Thus IDRISI attribute tables have an
".mdb" extension. Internally, each can carry multiple tables and can also be used and modified with Microsoft Access or
any other Access-compatible system.
Values Files (.avl)
Values files contain the values for a single attribute. They are stored in ASCII text format with two columns of data separated by one or more spaces. The first column contains an identifier that can be used to associate the value with a feature
(either raster or vector), while the second column contains the attribute value. Attribute values files have an ".avl" extension. The following is an illustration of a simple values file listing the populations (*1000) of the 10 provinces of Canada:
where 1 = Newfoundland, 2 = Nova Scotia, 3 = Prince Edward Island, and so forth.
1
2
3
4
5
6
7
8
9
10
560
850
129
690
6360
8889
1002
956
2288
2780
The feature definition image to be used with this values file would have the value 1 for all pixels in Newfoundland, the
IDRISI Guide to GIS and Image Processing Volume 1
52
value 2 for Nova Scotia, and so on.
Attribute Documentation Files (.adc)
As with images and vector files, attribute files (of either type) also carry documentation files, but this time with an ".adc"
extension. Similarly, Metadata is the utility which can be used to create or modify the documentation file associated with
an attribute file.
As with the other documentation files in IDRISI, those for attribute files are stored in ASCII format with the first 14
characters used purely for descriptive purposes.
A sample attribute documentation file appears as follows:
file format
: IDRISI Values A.1
file title
: Roads
file type
: ascii
records
: 12
fields
: 2
field 0
: IDR_ID
data type
: integer
format
: 0
min. value
: 105
max. value
: 982
display min
: 105
display max
: 982
value units
: ids
value error
: unknown
flag value
: none
flag def'n
: none
legend cats
: 0
field 1
: ROAD_TYPE
data type
: integer
format
: 0
min. value
: 1
max. value
: 3
display min
: 1
display max
: 3
value units
: classes
value error
: unknown
flag value
: none
flag def'n
: none
legend cats
: 3
Chapter 5 Map Layers, Collections and Data Structures
53
code 1
: Major Road
code 2
: Secondary Road
code 3
: Minor Road
This example shows the documentation file for an ASCII attribute values file. This type of file always has two fields.
Database tables will normally have many more fields. For each field, the parameters shown are repeated.
In this version of IDRISI, the following file types are supported:
ascii
Simple 2-column ASCII format files where the left field contains integer feature identifiers and the right field
contains data about those features. These values files have an ".avl" file extension.
access
A database file in Microsoft Access format having an ".mdb" extension. This format is the current resident database format supported by IDRISI, and is supported by Database Workshop. It is also the format used in the creation and display of vector collections.
In the case of the simple ASCII form (".avl" file), format information is unimportant, and thus reads 0 in this example.
Spaces or tabs can be used to separate fields. With fixed length ASCII and database files, format information is essential.
The format line simply indicates the number of character positions occupied by the field.
byte and integer data
A single number to indicate the number of character positions to be used.
character string data
A single number to indicate the maximum number of characters.
real number data
Two numbers separated by a colon to indicate the total number of columns and the number of those columns to
be used for recording decimal values. For example, 5:2 indicates that five columns are to be used, two of which
are for the decimal places. Note that the decimal itself occupies one of these columns. Thus the number "25.34"
occupies this field completely.
For most other entries, the interpretation is the same as for raster image layers. Valid data types include byte, integer, real
and string (for character data). However, for data tables (.mdb), many of the data types recognized by the Microsoft Access
Jet Engine are supported. These include:
single
(single precision real, 1.5x10-45 to 3.4x1038, 6-7 significant figures)
double (double precision real, 5.0 x 10-324 to 1.7 x 10308, 15-16 significant figures)
byte
(0-255, whole numbers)
integer (-32768 to 32767, whole numbers)
long
(long integer, -2,147,483,648 to 2,147,483,647, whole numbers)
text
(character strings)
IDRISI will document these types automatically, and knows how to convert them when they are used in the context of
display (in a vector collection) or when creating an ASCII values file.
IDRISI Guide to GIS and Image Processing Volume 1
54
Other File Types
While the majority of data files you will work with are those describing layers and their attributes, many others exist within
the IDRISI system and some of them are described below. Other more specialized file types are described in the context
of specific modules and in the File Structures section of the on-line Help System.
Map Composition Files (.map)
Map composition files store the graphic instructions necessary to create a map composition using data from a set of map
layers and associated symbol and palette files. They are described further in the Display System chapter in this volume.
Symbol and Palette Files (.sm0, .sm1, .sm2, .smt, .smp)
In order to display map layers, it is necessary to set up an association between vector features or image values and particular graphic renditions. This is done through the use of Symbol Files and Palette Files. A set of symbol and palette files is
included with IDRISI, and may meet most user needs. However, for final output, it is often desirable to create custom
symbol and palette files. This is done with the Symbol Workshop utility under the Display Menu. See the Display System chapter in this volume for more information about Symbol Workshop.
Symbol files indicate the manner in which vector features should be symbolized (such as the type, thickness and color of
lines or the font, style, size and color of text). Each symbol file lists the characteristics of up to 256 symbols that are identified by index numbers from 0 to 255. In all, four kinds of symbol files are used, one each for the vector feature types:
point, line, polygon and text. They are stored with file extensions of ".sm0," ".sm1," ".sm2" and ".smt" respectively. As
usual, IDRISI always knows the appropriate symbol file type to use. As a result, you never need to specify a symbol file
with its extension.
For raster images, graphic renditions are specified by the use of palette files. Like symbol files, palette files also define up
to 256 renditions identified by index numbers from 0 to 255. However, in this case, only the color mixture (defined by the
relative amounts of red, green and blue primaries) is specified. Palette files are stored with an ".smp" extension.29
Reference System Parameter Files (.ref)
Reference system parameter files record information about specific geographic referencing systems. They include data on
the projection, ellipsoid, datum, and numbering conventions in use with a reference system. IDRISI includes over 400
such files. The user can modify these and also create new reference system parameter files with the Metadata utility. See
the Georeferencing chapter in this volume for more information about these files.
29. The MS-DOS version of IDRISI used a different palette file structure with a ".pal" extension. The PALIDRIS module in the import group can read
and convert these files.
Chapter 5 Map Layers, Collections and Data Structures
55
IDRISI Guide to GIS and Image Processing Volume 1
56
Display System
Introduction
The previous chapter described map layers as each representing a single elementary theme. By combining map layers and
giving them graphic renditions, we create a map. Map compositions may contain as few as one layer and as many as 32 layers. Map compositions are created as a natural consequence of working with the IDRISI display system. As you work,
IDRISI keeps track of all changes and additions you make to the composition. You can then save the composition at any
time. The composition is stored in a file with a ".map" extension and is simply called a map file.
The display system in IDRISI consists of several separate but interdependent program components, each of which is
described in this chapter.
Display Launcher
DISPLAY Launcher is used to open a new display window. It begins the map composition process, and is always the first
operation required to create a new map display. DISPLAY Launcher can be accessed from its toolbar icon (shown above)
or by choosing it from the Display menu. Doing so opens a dialog box with options to display an image layer, a vector
layer, or an existing map composition.
When you select a raster or a vector layer, Idrisi uses a set of decision rules based on the values in the layer to suggest an
appropriate palette or symbol file. You may change this selection. You must also specfy if the layer should be autoscaled
(see below). In the case of a map composition, you will only be required to specify its name since all display parameters
are stored in the map file. See the chapter System Overview in this volume for options on choosing file names from the
pick list or typing them in directly. Click OK to display the map layer or composition.
Palette and Symbol Files
Palette files define the way the information stored in image pixel values will be displayed on the screen. Similarly, symbol
files define the way vector features with particular ids or linked attribute values will appear. Each stores the character of
up to 256 graphic renditions referenced by index numbers 0-255. Palette files store the Red, Green and Blue (RGB) color
mixture for each index while symbol files store a variety of parameters related to specific vector object types (e.g., point
size, line width, color, etc.). The section below on Symbol Workshop lists the parameters that may be defined for each
type of symbol file.
IDRISI comes with a set of standard palettes and symbol files. These are installed in the symbols folder of the Idrisi32
program directory (e.g., C:\Idrisi32\symbols). You may wish to extend that set, either by modifying existing palettes and
symbol files, or by creating entirely new ones. Symbol Workshop can be used for this purpose. User-created palette and
symbol files may be stored anywhere, including the Idrisi32 symbols directory. Typically users store commonly-used symbol files in the symbols directory, while they store files specific to particular data sets with those data.
Autoscaling
Autoscaling concerns the relationship between raster cell values and palette indices (and, similarly, vector ids or linked
attribute values and symbol indices). By default, a direct relationship between them is assumed (i.e., a cell with a numeric
value of 12 should be displayed with the color defined by palette index 12). However, not all images contain integer values
that fall nicely within the allowable range of values for palette indices (0-255). As a result, it is often necessary to scale the
Chapter 6 Display System
57
actual range of values into this more limited range. For example, we might have an image layer of temperature anomalies
ranging from (-7.2) degrees to (+4.6) degrees. These values cannot be directly displayed because there is no palette index
of (-7.2). Autoscaling offers a solution to this problem.
When autoscaling is used, any layer can be displayed, no matter what its range of values. When autoscaling is invoked, the
display system automatically calculates a relationship between input values and palette indices such that the lowest value is
given the lowest palette index, the highest is given the highest palette index, and all other values are given a palette index
in direct proportion to their position within this range. Thus, in the temperature anomaly example given above, the value
(-7.2) would be displayed with palette index 0, the value (+4.6) with index 255 and the value (+1.3) with index 184.30
While the autoscaling discussion has thus far focused on raster images and palette files, the same concepts apply to vector
features with their associated id's or linked attribute values and symbol files. When a vector file is displayed with autoscaling, its id's are autoscaled. When a vector layer that is part of a vector collection (i.e., it is linked to a data table) is displayed, it is the linked field attribute values that are autoscaled for display.
There are several further issues that should be noted about autoscaling. First, autoscaling is automatically invoked whenever real numbers or integer values outside the 0-255 range must be displayed. Second, while the maximum range of palette and symbol indices is 0-255, some palette or symbol files may use a more limited range of values for the purpose of
autoscaling (e.g., 1-100). The autoscaling range of a palette or symbol file can be inspected and changed with Symbol
Workshop. Third, while autoscaling works well for many images, those with extremely skewed distributions of data values
(i.e., with a very small number of extremely high or extremely low values) may be rendered with poor contrast. In these
cases, you can alter the display min and display max values (known as the saturation points) using the Layer Properties
option of Composer (see below). Finally, note that autoscaling does not change the values stored in a data file; it only
alters the display. To create a new raster image with contrast-stretched values, use the module STRETCH.
Automatic Display
IDRISI includes an automatic display feature that can be enabled or disabled from the User Preferences option of the File
menu. With automatic display, the results of analytical operations are displayed immediately after the operation has been
completed. The system will determine whether to use the default qualitative or quantitative palette (both of which are
user-defined in the Preferences dialog) for display and if autoscaling should be invoked. The artificial intelligence that is
used to make these determinations is not fool-proof, however. Palette and autoscaling choices may be quickly corrected in
the Layer Properties dialog if necessary. The automatic display feature is intended as a quick-look facility, not as a substitute for DISPLAY Launcher.
Launching Maps and Layers from IDRISI File Explorer
You will notice that the IDRISI File Explorer has a series of display options. The default is to view the file as a graphic
rendition in a map window. This uses the automatic display feature to display any listed data layer or map composition.
However, it should be noted that automatic display has limited information about the layer, and thus it may not produce
the best looking display. Again, this is intended only as a quick look facility and not as a substitute for DISPLAY
Launcher. Map compositions displayed from IDRISI File Explorer are displayed exactly as they have been saved, since all
palette and symbol file information is included in the map file.
Map Windows and Layer Frames
A map layer is displayed in a Layer Frame. This and all other components of the map composition (e.g., north arrow, legend) are placed in the Map Window. A single Map Window includes a single Layer Frame. The other components that may
30. The equation used by autoscaling to calculate the index to use for any particular value is as follows:
Display_Index=[((palette autoscale max - palette autoscale min)/(display max - display min))*(value - display min)] The minimum and maximum index
values for autoscaling are recorded in the palette or symbol file and may be altered in Symbol Workshop. The display min and max values are recorded
in the documentation file and may be altered with the Metadata utility.
IDRISI Guide to GIS and Image Processing Volume 1
58
be placed in a Map Window are described below in the section on Map Properties. The Map Window may be thought of
as the graphic "page" upon which a map composition is arranged. Its color is set in the Map Properties dialog and it can
be maximized or resized interactively by placing the cursor on the window border until it changes to a double arrow, then
dragging it to the desired position.
A Layer Frame can be moved by first double clicking upon it (you will notice a set of sizing buttons then appear) and then
dragging it to its new position.31 It can also be resized by grabbing one of the visible sizing buttons and moving the layer
frame border. In either case (moving or resizing), the action will be completed by clicking on any other map component
or the map window banner. This will cause the sizing buttons to disappear and the layers to resize (preserving the original
aspect ratio) to fit into the layer frame.
Note that there are several tool bar buttons that are useful in layer frame manipulations. These are shown in the table
below. The first causes the layer frame to close up around the currently displayed layers. This is particularly useful after
resizing, when the new layer frame is not exactly the right shape to hold the layers. The second causes the map window
and layer frame to expand as large as they can while still being fully visible (the button can also be activated by pressing
the END key). Note that in doing so, the system reserves space for Composer. If you wish to ignore Composer, hold the
shift key while pressing the END key. The third button returns the map window and layer frame to their initially-displayed state. This action can also be triggered by pressing the HOME key.
Toolbar Icon
Keyboard
Action
Fit Map Window to Layer Frame
END
Maximize display of Layer Frame
(leave space for Composer)
SHIFT+END
Maximize display of Layer Frame
(ignore Composer)
HOME
Restore original window
Composer
As soon as the first IDRISI map window has been opened, the Composer dialog box appears on the screen. Composer
can be considered a cartographic assistant that allows one to:
a)
add or remove layers from the composition;
b)
change the order in which layers are drawn (called the priority of a layer);
c)
temporarily toggle a layer to be invisible (i.e., hide it) without removing it;
d)
examine and alter the properties of a layer, including the symbol or palette file in use, display saturation points,
and autoscaling;
e)
add, delete and modify a variety of map components in the map window;
31. To drag a component, place the cursor over the component (not on the sizing buttons) and hold the left mouse button down while moving the cursor to the desired position. "Drop" the component in place by releasing the mouse button.
Chapter 6 Display System
59
f)
activate cursor query mode to examine feature properties of any layer;
g)
save the current composition (as displayed) as a MAP file; and
h)
print the composition.
If you have several map windows open, you will notice that Composer displays the information for the map window that
has focus (or the last one to have received focus if a non-map window, such as a dialog box or other window, currently has
the focus). Focus refers to the ability of a window to receive input messages from the mouse and keyboard. Windows designates the window that has focus by displaying its banner with a specific color.32 To bring a window into focus, click on
any part of it. To change the characteristics of a particular map composition, first give it focus.
Add Layer
To add a layer, click the Add Layer button on Composer and choose the desired file.33 You
will then be presented with a similar dialog to indicate the palette or symbol file to be used
and whether the new layer should be autoscaled. All layers in a map composition must share a
common reference system. If this is not the case, a warning message will appear indicating
this and warning that the layer may not display properly. The bounds of files do not need to
match to be displayed together in a map composition.
When a new layer is added, it is automatically given the highest priority (i.e., it is drawn on top
of the other layers). The first layer displayed will thus, by default, have the lowest priority (priority 0) and will appear to be on the bottom of the composition. The layer with priority 0 is
often referred to as the base layer in this documentation.
When multiple layers are present, their priority can be changed by dragging the name of the
layer concerned, and dropping it in the desired position. Whenever a map window is redrawn,
the layers are drawn in their priority order, starting with 0 and progressing to the highest
value.
Note that if a raster layer is added, whether on top of another raster layer or on top of a vector layer, it will obscure all the layers beneath it. Commonly, a single raster layer is used in a
map composition, and it has lowest priority so the vector layers will display on top of the raster layer.
Remove Layer
To remove a layer, select its name in the list of layers shown on Composer, then click the Remove Layer button. If you
wish to only temporarily hide the layer, do not remove it, but rather, change its visibility.
Layer Names, Types, and Visibility
Composer displays one row for each layer in the composition. You can toggle a layer visible or invisible by clicking on the
button to the left of its name. Note that toggling a layer off does not remove it from the composition. Rather, it simply
makes it temporarily hidden. This can greatly facilitate visual analysis in some instances.
To the right of the layer name is a graphic symbol which indicates the layer's type: raster, vector point, vector line, vector
polygon or vector text. These are illustrated in the table below. This information is used, for example, to indicate which
32. This is only one of many settings in the Windows system display that is defined when you choose a color scheme or define the display characteristics
of individual Windows components.
33. Only one layer frame is present in any map window, so all layers are added to the same layer frame. Note that layers may be added from several paths
for display. However, if the composition is to be saved as a map composition file, all the layers must exist in the same folder.
IDRISI Guide to GIS and Image Processing Volume 1
60
file is which when a raster layer and a vector layer are in the same composition.
Layer Type
Icon on Composer
Raster
Vector Point
Vector Line
Vector Polygon
Vector Text
Layer Properties
The properties of any layer, including key documentation information and
display parameters, are shown in the Layer Properties dialog. To view this,
highlight the layer concerned with a single mouse click on the layer name in
Composer, and then click the Layer Properties button. This will display the
Layer Properties dialog associated with that layer.
The Layer Properties dialog summarizes several important elements of the
layer's documentation file and gives you the option of accessing the Metadata utility from which you may view and alter all the documentation file
values. If the layer is raster, a histogram of the layer may be launched using
the Histogram button. If the layer is a vector feature definition file linked to
an attribute table field, the full data table may be displayed using the Access
Table button.
This dialog also lists the symbol or palette file in use and the status of
autoscaling, which can be changed. At the bottom of the Layer Properties
dialog, a set of sliders for modifying the saturation points is shown if the
layer is utilizing autoscaling. If the layer is a 24-bit raster image, three sets of
sliders will appear, one for each band. This facility is described in more
detail below.
The OK button closes the dialog, leaving the display as it currently is but not saving any changes to the file. The Cancel
button reverts and closes the dialog. Changes made in the Layer Properties dialog to the display min and max values may
be saved to the layer documentation file with the Save Changes button. The original display may be restored with the
Revert button.
Only one Layer Properties box may be open at a time. To see the layer properties for another layer, first close the Layer
Properties, select the new layer on Composer, then launch Layer Properties again.
Changing the Display Min / Display Max Saturation Points
When autoscaling is in effect, all values less than or equal to the display min recorded in the documentation file will be
assigned the lowest symbol or palette color in the sequence. Similarly, all values greater than or equal to the display max
will be assigned the highest symbol or color in the sequence.34 It is for this reason that these values are known as saturation points. Adjusting them will alter the brightness and contrast of the image.
34. Note that the lowest and highest symbols or colors in the sequence are those set as the autoscale min and max in the symbol file or palette. These may
be adjusted in Symbol Workshop.
Chapter 6 Display System
61
Two options exist for changing these values within the Layer Properties dialog. They can be edited within the appropriate
input boxes, or they can be adjusted by means of the sliders. The sliders can be moved by dragging them with the mouse,
or by clicking on the particular slider and then pressing either the left or right arrow keys. The arrow keys move the slider
by very small increments by default, but will yield larger movements if you hold down the shift key at the same time.35
The best way to become familiar with the effects of changing the saturation points is to experiment. In general, moving
the saturation points toward the center from their corresponding minimum and maximum data values will increase the
contrast. In effect, you sacrifice detail at the tails of the distribution in order to increase the visibility of detail in the rest of
the distribution. If the image seems too dark overall, try moving the display max down. Similarly, if the image seems too
bright, move the display min up. Depending on the distribution of the original data values (which can be assessed with
HISTO), interactively adjusting the saturation points and observing the effects can be very helpful in visual image analysis. This is particularly the case with 24-bit color composite images.
Map Properties
While the Layer Properties dialog displays information about a single map layer, the Map Properties dialog describes the
entire map composition. It is used to set the visibility and characteristics of map components such as layer legends, north
arrow and so forth.
The Map Properties dialog can be brought forward by clicking the Map Properties button on Composer or by clicking the
right mouse button in a map window. The right click method is context sensitive. For example, a right click over a legend
launches the Map Properties dialog with the legend options visible.
Map Properties is a tabbed dialog. Simply click on the appropriate tab to access the options associated with a particular
map component. Ten tabs are provided: Legends, Georeferencing, Map Grid, North Arrow, Scale Bar, Text Inset,
Graphic Insets, Titles, Background and Placemarks. The options for each are described below.
You can enlarge the Map Window at any time to increase the size of your graphic page and make space for new map components. In all cases, the components must be set to Visible to be displayed. All components are movable and many may
35. If you always wish to display a layer with particular display min/max values that are not the actual min/max values of
the layer, use METADATA to set the display min and display max fields of the layer documentation file to the desired
values. Whenever the file is displayed, these values will be used to set the range of values covered by the palette or symbol
file. The actual data values remain unchanged.
IDRISI Guide to GIS and Image Processing Volume 1
62
be interactively resized. To do so, double click on the component, make the necessary changes, then click any other element to release move/resize mode.
Legends
Up to five layers may have displayed legends in a map composition. Any raster layer or vector point, line or polygon layer
can have a legend. The legend text entries are stored in the layer documentation file and may be entered or altered with
the Metadata utility. Up to 256 entries are allowed. By default, the legend will display up to 20 categories (this can be
changed in User Preferences under the File menu). However, whenever there are more than 20 legend entries in the documentation file, a vertical scrollbar will be attached to the legend, and you may scroll through the remaining entries.
For a quantitative data layer with byte or integer data type and a range of values of 20 or less, a legend will be formed and
labeled automatically (i.e., no legend entries are needed in the documentation file). If the data are real or the data range is
greater than 20, a special continuous legend will be displayed with automatically generated representative category labels.
Georeferencing
The Georeferencing tab shows the reference system and units of the current composition, as well as its bounding rectangle. It also shows the actual bounding rectangle of all features within the composition. The former can be changed using
the input boxes provided. In addition, a button is provided that will allow you to set the composition bounding rectangle
to the current feature bounds.
The Georeferencing tab is also used to set a very important property regarding vector text layers. Text sizes are set in
points—a traditional printing measurement equal to 1/72 inch. However, in the dynamic environment of a GIS, where one
is constantly zooming in and out (and thus changing scale), it may be more useful to relate point size to ground units. The
Convert Text Point Sized to Map Reference Units option on the Georeferencing Tab does this. When it is enabled, text
captions associated with text layers (but not map components, such as titles) will change size as you zoom in and out,
appearing as if the text is attached to the landscape. The scaling relationship between point size and map reference units is
also user-defined. When this option is disabled, the text captions retain their original size as you zoom in and out of the
map layers.
Map Grid
A map grid can be placed on the layer frame. The intervals may be specified as well as the starting values in X and Y (all in
reference units). You can also choose the color and width of the grid lines and the label font. The grid will be automatically labeled.
North Arrow
A north arrow can be added to a composition. You set the declination of the arrow and may specify any text to be displayed to the left and right of the arrow. This would allow you, for example, to change the text to read "Magnetic North"
rather than "Grid North," or to change to a different language.
Scale Bar
Adding a scale bar can also be achieved with the Map Properties dialog. You set the length (in map reference system
units), the number of divisions, color properties and text label of the scale bar. The scale bar automatically adjusts in
width as you zoom.36
36. This can sometimes prove to be a nuisance since the scale bar may need to expand to be greater than the width of the map window. In these cases,
simply set its visibility to "off" using Map Properties. Otherwise, use Map Properties to reset the size of the scale bar to a smaller length.
Chapter 6 Display System
63
Text Inset
The text inset can be used to display blocks of text. This is done by specifying the name of an ASCII text file. The text
inset frame is sizeable and incorporates automatic wordwrap. Font style and background color may be set.
Graphic Insets
Graphic insets can be used to hold inset maps, pictures, logos, or any other graphic image. Any Windows Bitmap
(".bmp") or Metafile (".wmf" or ".emf") may be used as a graphic inset. All graphic insets are resizeable. However, if the
Stretchable option is disabled, resizing is controlled to preserve the original aspect ratio of the inset.
Titles
A title, subtitle, and caption may be added to a map composition. These are not associated with any particular map layer,
but rather belong to the map composition as a whole. However, the title of the base layer, read from its documentation
file, will appear as the title text by default. This may be edited and font properties may be set.
Background
The Background tab allows you to change the color of the backgrounds for the Layer Frame and the Map Window. The
backgrounds of other components, such as legends and the scale bar, are set in their respective tabs. However, the Background tab also allows you to set the backgrounds of all components to match that of the Map Window. Note that setting
the layer frame background may not be noticeable if a raster layer is present in the composition (since it may cover up the
entire frame).
Placemarks
Finally, the Placemarks tab keeps track of placemarks associated with the map composition. A placemark is a particular
window of the map layers. The Placemarks tab allows you to define new placemarks, delete existing ones, and go to any
specific placemark. Placemarks are described further below in the section on Interactive Display Features.
Feature Properties
The Feature Properties button on Composer invokes the Feature Properties cursor mode, which brings up a table of
properties associated with features identified with the mouse. This may also be invoked with the Feature Properties toolbar icon. See the section below on Interactive Display Features for a complete description.
Save Composition
At any point in your use of Composer, it is possible to save your map composition in several formats. The Save Composition button brings up a dialog presenting the format choices. The first is to save the map composition as a MAP file. A
MAP file contains a complete record of the composition in a file with a ".map" extension.37 This can then be redisplayed
at any time by starting DISPLAY Launcher and indicating that you wish to display a map composition file. With this form
of composition storage, it is possible to restore the composition to its exact previous state, and then continue with the
composition process.
The next three options allow you to save the composition to either a BMP, WMF or EMF graphics file format. These file
types are used in desktop publishing, spreadsheet and word processing software programs. The BMP is a raster bitmap,
produced by simply copying the screen. This is often called a screen dump, since it copies the exact appearance of the map
window, without copying the details of how the composition was formed (i.e., the names of the layers, their symboliza37. More information about the MAP file structure may be found in the chapter Map Layers, Collections and Data Structures. Note especially that
the MAP file contains the instructions to construct the map composition, but does not include the map data layers themselves. These separate files are
used to recreate the composition.
IDRISI Guide to GIS and Image Processing Volume 1
64
tion, and so on). The WMF format is the "Windows Metafile" structure used through Windows 3.1. It stores the Windows instructions that can be used to reconstruct both the vector and raster elements of the map window. These can then
be imported into a desktop publishing or graphics program. The EMF format is the newer Windows "Enhanced Metafile" structure. This is a follow-on to the WMF structure, and is specifically designed for 32-bit variants of Windows (95,
98, NT, 2000). Whenever you have a choice between WMF and EMF, choose the EMF format.
The next option copies the map window to the Windows clipboard rather than to a file. This facilitates copying compositions immediately into other software programs.
Finally, the last option allows you to save the currently windowed region of the active layer as a new IDRISI layer. If the
layer is raster, only the portion of the raster image that is currently displayed will be written to the new raster file. This is
thus an interactive version of the WINDOW module. However, if the active layer is vector, the entire file will be copied,
but the bounding coordinates will be altered to match the current window. The new file will display the window region,
but all the original data outside that window still exists in the file.
Print Composition
To print a map composition, first display it. Ensure that that map window has focus, then click the Print Composition
button on Composer. In the Print Composition dialog you will be able to select the desired printer, access its setup
options, choose to fit the map window as large as possible on the page or print to a user-specified scale. Page margins may
be set and the line widths may be scaled. A preview of the page as it will print is shown. All Windows-compatible printing
devices are supported.
Navigation
At the bottom of Composer are several small navigation buttons. These allow you to pan (move) to the left, right, top and
bottom and to zoom in and out. These are described later in this chapter in the section on Interactive Display Features.
Symbol Workshop
The map layers in a composition are rendered by means of symbol and palette files. While a set of symbol files is included
with IDRISI, commonly you will want to develop specific symbol files to optimize the impact of the information presented on your final maps.
Symbol and palette files are created and modified using Symbol Workshop,38 available under the Display menu and
through its toolbar icons. In all, five types of files can be created: point symbol files, line symbol files, polygon symbol
files, text symbol files and palette files. Symbol files record the graphic renditions for up to 256 symbols, indexed 0-255.
For example, a text symbol file might indicate that features assigned symbol index 5 are to have their names rendered
using bold, italic, 10 point Times New Roman text in a red color.
From the Symbol Workshop File menu, you can choose to open an existing file or create a new one. If you choose the latter, you will need to indicate the symbol type: point, line, polygon, text, or palette. The 256 symbols are then arrayed in a
16 x 16 grid. To change a symbol, simply click within its symbol grid cell. A symbol-specific dialog will then appear, allow-
38. IDRISI for Windows had a separate utility named Palette Workshop for the manipulation of palettes. This has been incorporated into Symbol
Workshop with Idrisi32. Two icons on the toolbar open Symbol Workshop. The first opens it ready to work with vector symbol files while the second
opens it ready to work with palette files.
Chapter 6 Display System
65
ing you to alter any of the following settings:
Symbol File Type
Point Symbols
Attribute
Symbol Type
Fill Style
Size
Fill Color
Outline Color
Line Symbols
Line Style
Line Width
Color
Polygon Symbols
Fill Style
Color
Text Symbols
Font
Size
Color
Style (normal, bold, italic, underline)
Palette
Color
A very important feature of Symbol Workshop is the ability to copy or blend attributes. For example, imagine creating a
point symbol file of graduated circles. Click on symbol 0 and set its properties to yield a very small yellow circle. Then
move to symbol 255 and set it to be a large red circle. Now set the blend end points to be 0 and 255 and click the blend
button. You will now have a sequence of circles smoothly changing in both size and color.
As a companion to the blend option, Symbol Workshop also provides a copy function. With both the blend and copy
functions, it is possible to set options that will cause them only to copy or blend specific attributes (such as color or size).
To get a sense of how the symbols will look on a particular background you may also set a background color for the Symbol Workshop display.
Finally, the autoscaling range of a symbol file may be specified or altered in Symbol Workshop. All symbol files contain
definitions for 256 symbols. However, by altering the autoscale range, it is possible to create sequences of more limited
range. For example, to set up a palette with 8 colors, set the autoscale min and max to be 0 and 7 respectively. Then define
these 8 colors. In use, it will then appear that the palette has only 8 colors if the layer with which it is used is autoscaled.
Media Viewer
Media Viewer is a facility for creating and displaying video images composed of a series of IDRISI images. Media Viewer
is located under the Display menu. When activated it will present the basic control dialog with its own separate menu.
Click on the File menu to create a new video or to open an existing video. If you choose to create a new video, you will be
presented with a new dialog that will require the name of either a Raster Group file (".rgf") or a Time Series file (".ts") that
defines the images to be used and their order in the video. You will also be asked to specify a palette to be used and the
time delay to be used between each successive image. The result of the operation will be an ".avi" multi-media video file
that can be displayed at any time using the Media Viewer controls. Note that the viewer can be sized by dragging its borders. The image can be sized to fit within the viewer by selecting the Fit to Window option under the Properties menu. Also
IDRISI Guide to GIS and Image Processing Volume 1
66
note that you can place a video into a continuous loop.
Interactive Display Features
One of the remarkable features of GIS is that map displays are not static. Rather, they provide a highly interactive
medium for the exploration of map data. IDRISI includes a number of features that form the basis for this interaction.
Move and Resize
As discussed earlier, all map components (layer frame, title, legend, etc.) can be moved, and most can also be resized.
Moving is achieved by placing the cursor over the component in question, double clicking to enter move/resize mode,
and then holding down the left mouse button to "drag" that element to its new location. Clicking on any other component, or the map window banner, will exit move/resize mode. While in move/resize mode, resizing is achieved by dragging one of the resizing buttons on the margins of that component.
Cursor Inquiry Mode
As you move the cursor over a map window, the status bar at the bottom of the screen will indicate the X and Y coordinates of the cursor in the geographic reference system (for raster layers, the column/row position is also shown). You can
find out what is at that location by using Cursor Inquiry Mode. To enable cursor inquiry mode, click on its button on the
tool bar. The button will appear to be depressed. When this mode is active, you can query the value at any position on the
active layer of any map window. In the case of a raster layer, it will show the numeric value and legend interpretation (if
one exists) of the grid cell immediately below the cursor. For a vector layer, it will show the numeric value and legend
interpretation of the nearest feature. Note especially that when a map window contains several layers, the displayed value
is for the active layer (the one highlighted in Composer). The active layer can easily be changed by clicking onto the
desired layer name in the Composer list. Cursor inquiry can remain active, but you may want to turn it off if you wish
move and resize map components. To do so, simply click the cursor inquiry button again.
Feature Properties
Cursor Inquiry Mode allows one to view the values of features for a single layer at a time.
For layers that are members of a collection, the Feature Properties option allows one to
look at a simple tabular view of the values for all members of the collection at the query
location. To activate this feature, either click the Feature Properties button on Composer
or click the Feature Properties button on the tool bar. Click it again if you wish to turn it
off.
For multi-layer query mode to work, the layer must have been displayed or added to the
map window as a part of a collection. This is accomplished by either typing in the layer
name using the "dot logic" described in the chapter Map Layers, Collections and Data
Structures in this volume, or by choosing the layer name from within the list of members
under the collection file name in the pick list. When Feature Properties is used with layers
that are not members of a group some simple information regarding the layer is presented
in the Feature Properties table along with the value and position of the location queried.
Clearly the major advantage of the Feature Properties table is that one can see the values of
all group members at one time. In addition, it can display these as a graph. Note that controls are also provided to change the position of the separator line in table mode. To turn off Feature Properties mode,
click on the Feature Properties button again, or click its corresponsing icon on the tool bar. Cursor Inquiry mode is auto-
Chapter 6 Display System
67
matically invoked whenever Feature Properties is active.
Pan and Zoom
Pan and zoom allow one to navigate around the display at varying scales (magnifications). The simplest method of pan
and zoom is to use the special buttons at the bottom of Composer, though the keyboard may also be used. Functions are
summarized in the table below. The easiest way to understand the actions of these keys is to imagine that you are in an airplane. Zooming in lowers your altitude, thus increasing your scale (i.e., you see less area, but with greater detail). Similarly,
zooming out increases your altitude, thus reducing you scale (i.e., you will see more, but with less detail). The representative fraction (RF) in the status bar changes as you zoom in and out. A similar logic is associated with the arrow keys.
Pressing the right arrow key (Pan Right) moves your imaginary airplane to the right, causing the scene to appear to move
to the left.
Composer Button
Keyboard
Action
Left Arrow
Pan Left
Right Arrow
Pan Right
Up Arrow
Pan Up
Down Arrow
Pan Down
PgDn
Zoom In
PgUp
Zoom Out
In addition to these continuous zoom and pan options, the tool bar also offers a zoom window option. This feature
allows you to draw a rectangle around the area you wish to zoom into. To do so, click the Zoom Window button (shown
below) and move your cursor to any corner of the area to be zoomed into. Then hold down the left button and drag out
the rectangle to the corner diagonally oposite the one you started with. When you release the left mouse button, the zoom
will take place. The Restore Original Window button on the tool bar or the HOME key will revert the map window and
layer frame back to their original size.
Toolbar Icon
Keyboard
Action
Zoom Window
HOME
Restore Original Window
All zoom and pan operations take place within the context of the layer frame. See the earlier section of this chapter on
Map Windows and Layer Frames for information about enlarging both of these.
Collection Linked Zoom
When a layer is part of a collection, it is possible to have all displayed members of the collection zoom and pan simultaneously and identically. To activate this, click the Collection Linked Zoom button on the tool bar. Notice that the button
depresses and remains so until clicked again. This feature is especially useful when ground truthing remotely sensed imagery, particularly when used with the GPS link (see below).
The Collection Linked Zoom works only when layers are launched with their complete collection reference. This is not an
issue for linked-table collection members as they can only be launched with their full collection reference. However, for
IDRISI Guide to GIS and Image Processing Volume 1
68
layers belonging to raster group file collections, these layers exist in isolation as well as being referenced as part of a group.
Thus, if you wish to use the collection linked zoom feature, be sure to enter the full collection reference (e.g.,
"spotxs.band3" rather than simply "band3").
Placemarks
Placemarks are the spatial equivalent of bookmarks. Any particular view of a composition can be saved as a placemark.
To do so, either launch the Map Properties dialog and click the Placemarks tab or click the Placemarks icon on the tool
bar. Then choose the appropriate option to name, rename, delete, or go to any placemark. Note that placemarks are saved
with map compositions. They will not be saved if the composition is not saved.
GPS Support
IDRISI also provides real-time GPS support, intended for use with a laptop computer. A Global Positioning System
receiver receives continuous position updates in latitude/longitude39 from the system of 21 active GPS navigation satellites. When the IDRISI GPS link is active, this position is shown graphically with a blinking cursor on all map windows
that cover your current location and have a common reference system. IDRISI automatically projects the incoming positions to the reference system specified (so long as the specified projection is supported by PROJECT).40 In addition,
IDRISI will automatically save your route and allow you to save waypoints—positionally tagged notes.
Most receiver units available today support communication with a computer over an RS232C communications channel
(e.g., the COM1 or COM2 port on your computer), and provide support for the NMEA (National Marine Electronics
Association) communications protocol. In most cases, using such a GPS with IDRISI is remarkably simple, and only
involves the following steps:
1.
Set the GPS to output NMEA data (this is usually an option in its setup menu and can remain as a permanent
setting, avoiding this step in future use).
2.
Connect the special communication cable that is designed for your GPS to one of the serial communication
ports on your computer. For a laptop, this is typically a 9-pin port known as COM1. By default, IDRISI expects communication over COM1, but this can be changed (see below).
3.
Display an image or map layer that includes your current location. Be sure that it has focus and then click the
GPS button on the tool bar. You will then see the blinking position cursor appear simultaneously in all map windows that
use the same reference system (as long as your actual position is within the currently displayed area).
When you have finished a GPS session, click the GPS button again to turn it off. At that point, it will give you the option
of saving your route as a vector line layer and your waypoints (if any) as a vector point layer.
Saving Routes and Waypoints
While the GPS is connected and communicating with IDRISI, it automatically saves the route that was started when the
GPS button was first clicked. It also keeps a record of an waypoints that are entered along the way. When you click the
GPS button again to terminate GPS communication, you have the option of saving these as layers. The route will be
saved as a vector line layer and the waypoint file will be stored as a text layer.
To save waypoints, simply press the "w" key at those positions for which you wish to record information. A dialog will
then appear, allowing you to enter a text description. IDRISI will keep track of both the location and the text descriptor.
39. Actually, the native reference system is a three-dimensional datum known as WGS84. However, the NMEA interface for communication of fixes
converts these to latitude/longitude and elevation readings.
40. See the chapter Georeferencing in this volume for a discussion of projections and reference systems.
Chapter 6 Display System
69
Note that as a text layer, your waypoint information is easily displayed after the file has been saved. However, you will
probably want to keep these waypoint descriptors very brief. Position readings continue to be recorded while waypoint
information is being entered.
Finally, note that IDRISI is not intended as a spatial database development tool. Thus the route and waypoint saving features are more suited to simple ground truthing operations rather than database development. For much more extensive
capabilities, we recommend CartaLinx, which provides complete database development facilities with GPS support.
How It Works
When you click the GPS button, IDRISI checks to see if there is a map layer with focus and with a valid reference system.
This can be any system with a Reference System Parameter file that has a projection supported by PROJECT (this does
not include the system labeled "plane"). This then becomes the output reference system, and all positions will be automatically converted into that form.
Next, IDRISI launches (if it is not already active) a special GPS server program named IDRNMEA.EXE from the folder
called GPS located in the Idrisi32 program folder.41 Then it establishes communication with the GPS using communication parameters stored in a special file named "IDRNMEA.CFG," also found in the GPS folder of the Idrisi32 program
folder. The default settings stored in this file (1,4800,n,8,1) are probably correct for your unit (since they assume the use
of serial port 1 and comply with the NMEA standard) and can, in most cases, be used without modification. However, if
you have trouble communicating with the GPS, this file can be edited using an ASCII text editor such as the IDRISI edit
module or Windows Notepad. Details can be found in the on-line Help System.
Once communication has been established, IDRISI converts the native latitude/longitude format of the NMEA GPS fix
to the designated IDRISI reference system, polls all map windows to find which ones are using this system, and then
moves a blinking cursor to that location on each.
Interactive Screen Digitizing
Another very important interactive capability that IDRISI provides is the ability to digitize on screen. It is important to
recognize, however, that this facility is largely intended as a means of undertaking simple digitizing tasks such as the delineation of training sites for the classification of remotely sensed imagery, or creating text vector layers in a very rapid fashion. For larger and more complex tasks, we recommend CartaLinx, a full spatial database builder software, also available
from the Clark Labs.
The Digitize button on the toolbar is shaped like a cross within a circle, and can be found among a group of three related
digitizing buttons.
Toolbar Icon
Action
Digitize
Delete Feature
Save Digitized Data
To digitize on screen, make sure that the appropriate map window has focus, and then click on the Digitize button. If the
active layer (the one highlighted in Composer) is a raster layer, you will be presented with a dialog box in which you can
define the new vector layer to be created. If the active layer is a vector layer, you will first be asked whether you wish to
41. This is the same system used by CartaLinx. The GPS server program can simultaneously serve multiple applications. Thus, if it is already loaded for
use by CartaLinx, IDRISI does not load another copy, but simply registers itself as another user.
IDRISI Guide to GIS and Image Processing Volume 1
70
create a new vector layer for the digitized features or append the new features to those existing in the active layer. If you
choose to append to an existing layer, you will notice that the file information is already filled out and you only need enter
the ID or value. If you are digitizing a new file, you will be asked to specify the following:
Data Type
The data type refers to the attribute stored for each feature. This can be either integer or real. If you are digitizing identifiers to be associated with data in a table, this must be integer.42 In all other cases, use integer for coding qualitative
attributes and real for recording quantitative data (where a fractional number may be required), or whenever the value is
expected to exceed the integer range.43
Layer Type
Specify the nature of the features you wish to create. Note that if you choose to append to an existing layer, both the layer
type and data type are predetermined to match those of the existing layer and you will not be able to change those settings. Layer type options include points, lines, polygons or text.44
Automatic Index
This option is for those cases where you are storing identifiers for features (rather than some attribute directly). When
checked, this option will automatically increment the identifier from one feature to the next. This offers speed when digitizing features whose ids increment in this simple manner.
ID or Value / Index of First Feature
This allows you to specify the ID or attribute of the feature to be digitized. When the Automatic Index feature is specified, this will be used as the starting value for the numeric sequence.
Once you click OK to this startup dialog, a new layer will be added to your composition (unless you chose to append to
an existing layer), and the digitzing cursor will appear in the Layer Frame. Note that if you were in Cursor Inquiry mode
or Feature Properties mode, these will be temporarily disabled until you exit digitizing mode.
Mouse Button Functions When Digitizing
Once the Digitize dialog has been completed and you click on the OK button, you may begin digitizing. To digitize, use
the left mouse button to identify points that define the position, course or boundary of a feature. You will notice it being
formed as you digitize. To finish the feature (or the current sequence of points in the case of point features), click the
right mouse button.
Deleting Features
The Delete Feature button to the immediate right of the Digitize button on the toolbar allows you to select and delete
vector features from the active vector layer. First click on the Delete Feature icon, then select the feature to be deleted
(the cursor will become a hand). When the feature is selected it will turn red. To delete it from the file press the Delete key
on the keyboard.
42. With vector data, no distinction is made between different storage formats of integer. Vector integer data can have values between ±2,147,483,647.
43. Although vector layers can carry integer attribute values within a ±2,147,483,647 range, integer raster layers are restricted to a range of ±32,767. Thus,
if you intend to rasterize your digitized data at a later stage, you may wish to specify real if the range is outside the ±32,767 range.
44. When polygons are chosen, an option to digitize a flood polygon is presented. This is useful in image classification for delineating training sites. See
the Image Classification chapter in the IDRISI Guide to GIS and Image Processing Volume 2 for more information.
Chapter 6 Display System
71
Digitizing Additional Features Within a Single File
Once you have finished a feature by clicking the right mouse button, you can continue to add further features to the same
vector layer data file. Make sure the map window still has focus, and that the layer to which you wish to append another
feature is highlighted, and then click the Digitize button on the toolbar. You will be asked whether you wish to append
the new feature into the designated layer.
Saving The Vector Layer
As you digitize features, they will each in turn be added to the vector layer designated. However, these data are not committed to disk until you click on the Save Digitized Data button. This can be done repeatedly throughout a session to save
your work incrementally. If a map window is closed before a digitized layer is saved, a message will ask whether or not to
save the layer or changes made to the layer.
Digitizing Multiple Layers
You may have multiple vector layers open for digitizing at the same time. Any actions you take will apply only to the
active layer, the name of which is highlighted in Composer.
A Special Note About Digitizing Point Features
With both line and polygon layers, you digitize a single feature at a time, right click, then click the Digitize button again to
digitize the next feature. However, with point vector layers, you can digitize multiple features before ending a sequence
with the right mouse button. To do this, select the Automatic Indexing option.
A Special Note About Digitizing Polygon Features
Polygons are defined by lines that join up to themselves. When digitizing polygonal features, clicking the right button not
only terminates the definition of the feature, but it also adds a final point which is identical to the first, thereby closing the
feature perfectly. As a result, it is not necessary to try to close the feature by hand—it will be done automatically.
A Final Note
Perhaps the best way to learn more about the IDRISI Display System is to work through the first few tutorial exercises in
the Tutorial. These give step-by-step instructions and guide you through the basics of map layer display and map composition in IDRISI.
IDRISI Guide to GIS and Image Processing Volume 1
72
Database Workshop
Database Workshop is IDRISI's relational database manager, and lies at the heart of IDRISI's support for layer collections that link vector feature definition files to database tables. IDRISI uses the Microsoft Access Jet Engine as the basis
for Database Workshop. With this facility, one can undertake a wide variety of database operations. However, more
importantly, one can interact directly with linked-table collections: database queries can be shown immediately on the
associated map layer, and map layer queries can be directly linked to the data table. In addition, database field values can
be assigned or extracted from raster layers. Each of these is discussed below.
Working with Linked-Table Collections
A linked-table collection consists of a vector feature definition file, a database table and a link file associating the two. The
collection is defined using the Collection Editor from the File menu.45 The link file contains information about the vector
file, database file, table of the database file (a database file may have several tables), and link field for the collection. In a
linked-table collection each field (column) in the database, linked to the geographic definition of the features in the vector
file, becomes a map layer. These can each be displayed using DISPLAY Launcher by selecting the layer of interest from
below the collection file name, or by typing in the full "dot-logic" name of the layer. Database Workshop offers several
additional ways to examine these data.
Launching Database Workshop
To launch Database Workshop, either click its icon on the tool bar or select its entry in the Data Entry or Analysis/Database Query menus. When launched, if the selected layer of the map window with focus is from a linked-table collection,
Database Workshop will automatically open that table. Otherwise, use the File/Open menu option on the Database
Workshop menu to select the desired database file and table.
Displaying Layers from Database Workshop
Simply click the mouse into any record (row) of the field (column) you wish to view and then click the Database Workshop Display icon from the Database Workshop toolbar. The selected field will then be displayed using autoscaling and
the IDRISI default symbol file. Note that each such action launches a new map window,46 so it is very easy to overload
the display system if you do not have a great deal of RAM. To avoid this, close windows periodically. The first time you
display a layer you will be prompted to indicate the link file to use.
Database Query using an SQL Filter
Database query by attribute is accomplished in Database Workshop by filtering the database. This is simply the identification of which records have attributes that meet our query (i.e. filter) criteria. To query the active database table, click on
the Filter Table icon or choose the Filter Table option from the Query menu. This opens the SQL Filter dialog which
provides a simple interface to the construction of a Structured Query Language (SQL) statement.
The Select option at the top of the filter dialog specifies which fields to display in the result. The default asterisk indicates
all fields and is fine in most instances. To specify a subset of fields, type their names into this input box separated by com-
45. For more about collections, see the chapter Map Layers, Collections and Data Structures in this volume.
46. This may not be evident since each map window will exactly overlay the previous one. We recommend moving each new window to an unused area
of the screen as it is created so that all can be seen.
Chapter 7 Database Workshop
73
mas. Remember that all field names require square brackets around them if they contain spaces in the names. (To avoid
ambiguity, it is a good habit to place square brackets around every field name.)
The Where input box is where the main part of the filter is constructed. The tabbed options to the right facilitate the placement of filter elements. SQL requires spaces on either side of each operator. If you select elements rather than typing
them in, IDRISI will ensure that this is so. Note also than any valid SQL clause can be typed in; you are not restricted to
the options shown on the tabs.
The Order By input box is optional. It simply causes the results of the query to be sorted according to the field chosen.
Clicking OK causes the filter to be executed. Database Workshop will then show only the records that meet the filter criteria. The records that do not meet the criteria are still in the database, but they are hidden. (Remove the filter to restore
the full database—see below.)
Mapping the Filtered Records
When a filter is executed, IDRISI checks all open map windows to see if any contain an active layer that is linked to the
database that was filtered. If so, it will automatically display the results of the query as a Boolean map with features that
met the filter criteria shown in red and all others shown in black.
Toggling the Filter on the Map
When a filter has been executed, the Boolean filter mask on all map windows can be toggled on and off by clicking the
toggle button. This facilitates comparison with the original displayed results.
Removing the Filter
To remove any filter, choose the Remove Filter icon from the toolbar or choose the option from the Query menu.
Query by Location
When a map window contains an active layer (the one highlighted in Composer) linked to a database, you can use Cursor
Inquiry Mode (from the IDRISI toolbar) to perform database query by location. When you click on a feature in the map
display, Database Workshop will automatically locate the corresponding record in the database. The located record is
indicated with a triangular marker at the left edge of the record in the table display.
IDRISI Guide to GIS and Image Processing Volume 1
74
Other Database Operations
Calculating Field Values
In addition to querying the database, it is sometimes necessary to create new fields, either through importing external values or calculating new values from existing fields. For example, one might calculate a new field of population density values based on existing fields of population and area. The Calculate Field Values option of the Query menu (also accessed
by clicking its icon on the Database Workshop toolbar) produces an SQL dialog area similar to that of the SQL Filter. In
this case, it facilitates the construction of an SQL UPDATE SET operation to calculate new values for a field as a function of a mathematical or logical equation. In the SET input box, select the field to be calculated. Then enter the equation
into the main input box after the "=" sign using the tabbed options as an aid. As with Filter, any valid SQL clause can be
entered—you are not restricted to the options specified in the tabbed control.
Finding Specific Records
The Find Next option of the Query Menu (also accessed by clicking the Find Next icon on the Database Workshop tool
bar) provides a simple way to search for records. The "=" option looks for the next exact match while the "like" option
looks for approximate matches. This latter option is only valid on text string fields and makes use of wildcard characters.
Here are some examples:
like "*ks"
finds next record ending with "ks"
like "ks*"
finds next record beginning with "ks"
like "*ks*"
finds next record with "ks" anywhere within the text string
like "k?s"
allows any character in second position
like "k#s"
allows any digit (but not a letter) in second position
like "[a-d]*"
finds next record starting with a letter from a through d
like "[!a-d]*"
finds next record not starting with a letter from a through d
Sorting
To sort the records of the database according to the values of a particular field, click the mouse into any record of that
field then click either the ascending or descending sort button on the Database Workshop tool bar.
Entering or Modifying Data
You will specifically need to enter Edit Mode before any cell value in the database can be entered or modified. This
guards against accidental changes to the data values. Enter edit mode by choosing the option from the Edit menu, or by
clicking onto the Edit Mode status button. The grid changes color when you are in edit mode. Several tool bar options are
disbled until edit mode is turned off. You should therefore exit edit mode as soon as you have finished entering data. To
do so, choose the option under the Edit menu, or click onto the Edit Mode status button.
Modifying the Table Structure
The table structure can be modified (i.e., add, rename or remove fields, add or delete records) from the Edit menu. However, this cannot be done if other "users" have access to the table. Any map windows that are linked to this database are
considered to be users. Thus you will need to close down all of these map windows before the table structure can be
altered.
Chapter 7 Database Workshop
75
Assigning Data To and Extracting Data From Raster Layers
Database Workshop provides a very simple means of assigning field data to a raster layer, or extracting data from a raster
layer into a database field. To assign field data to a raster layer, create a values file (".avl') using the Export Values File
option of the File menu. Then use the ASSIGN operation in IDRISI to assign those values to the feature definition raster
image. When you export the values file you will specify both a link field and a data field. The link field identifiers must
match the values in the raster feature definition file to which the values file will be assigned. To extract data from a raster
image, first run EXTRACT in IDRISI to create a values file. Then choose the Import Values File option from the File
Menu. EXTRACT creates a values file with the first column representing the IDs in the raster feature definition image
that was used in the extraction. To import this to the database, those identifiers must match one field (the link field) in the
database.
When a filter is in place and a values file is exported, only the records in the filtered set are written to the values file. When
you then assign this values file to a feature definition image, all features that are not part of the filtered set will be given the
value 0.
Export and Import
Database Workshop also provides selected import and export options under the File menu. These are principally oriented
to xBase (dBase or other system that uses its structure).
IDRISI Guide to GIS and Image Processing Volume 1
76
Idrisi Modules
The purpose of this chapter is to give a brief overview of the functionality of each of the IDRISI modules and their typical application, and also to outline the logic of the IDRISI menu structure. For much more detailed information about the
modules and their operation, please refer to the on-line Help System.
The IDRISI Main Menu is divided into seven headings, each of which is described below. Menu entries followed by an
arrow lead to submenus with further choices. Because some modules are used in different contexts, they may appear in
more than one place in the menu. The Reference Guide also includes a Menu Map showing the entire menu structure in
one figure.
The names of IDRISI modules that can be used in macro mode are shown in the menu in all capital letters. All other
interfaces or modules are written with only first letters capitalized.
The File Menu
Data Paths
Idrisi File Explorer
Metadata
Collection Editor
Run Macro
Shortcut (on/off)
User Preferences
Import >
Export >
The Data Paths command allows you to set the paths from which input data will
commonly be selected and output data will commonly be written. You may set a main
working folder as well as a set of resource folders. Access to these folders is facilitated
in the pick list for input and output filename boxes in all module dialog boxes. At any
time, however, the user may easily browse beyond these specified paths to access the
entire folder structure.
The Idrisi File Explorer provides a familiar interface for managing and viewing
IDRISI-specific files. Use Idrisi File Explorer for regular file maintenance activities,
such as copying or renaming files. The contents of IDRISI files may be viewed in a
variety of formats from this utility as well.
The Metadata command provides the ability to create, view, and edit reference sysIdrisi File Conversion (16/32) tem parameter files and documentation files for raster, vector, and attribute values
files.
Exit
The Collection Editor facilitates the creation of raster group files and vector link
files. Both are ASCII text files that define sets of layers to be treated as a collection for some processes, such as cursor
query.
Run Macro allows you to run an IDRISI macro. Macros contain all the information needed to run many IDRISI modules in a specific order. See the chapter on Creating Macros in this volume for detailed information about the construction and execution of IDRISI macros.
Toggle the Shortcut command on the File menu to open an alphabetical listing of IDRISI command modules. Type in
the name of the module you wish to use, or select it from the list. The Shortcut dialog will remain open at the bottom of
the IDRISI screen until it is toggled off in the menu.
Through User Preferences, you may customize your IDRISI working environment. Here you can choose the language
with which you wish to work. The IDRISI interface and on-line Help System have been translated into several languages.
Check the Clark Labs Web Site or call Clark Labs Customer Support to inquire about the availability of specific language
files. Overwrite protection may be toggled on and off in the User Preferences dialog box. If overwrite protection is
enabled, you will be prompted to cancel or continue the operation each time you provide an output filename that already
exists. If overwrite protection is disabled, the existing file will be automatically overwritten without warning. You may also
Chapter 8 Idrisi Modules
77
specify a prefix for automatically-generated output file names. Several options governing the nature of the default display
choices are also set in the User Preferences dialog box.
The Import and Export entries of the File menu activate sub-menus. The Import tools are organized into four groups.
General Conversion Tools leads to a list of modules that may be used alone or in combination to change files to an
IDRISI format, or to convert them to a format that is compatible with a specific IDRISI import module. The Government/ Data Provider Formats group includes modules that import the most commonly used government and agency
data formats. Similarly, the Desktop Publishing Formats group includes modules that import graphic exchange data
formats typically used in desktop publishing software. Finally, the Software-Specific Formats group provides the modules used to import files from many GIS and related software packages. For detailed information on specific import procedures, see the on-line Help System. For an overview of the issues surrounding database development, including data
import, see the chapter Database Development in this volume.
The Export tools are organized into three groups: General Conversions, Desktop Publishing Formats and SoftwareSpecific Formats. The specific commands of both the import and export submenus are listed below.
General Conversion Tools Submenu
IMPORT AND EXPORT
CRLF adds or removes carriage returns or line feeds, and is also commonly used with USGS DEM and DLG files, or in
converting data files between DOS or Windows and UNIX or Macintosh systems.
XYZIDRIS is used to import X,Y,Z coordinate data such as might be collected by a GPS unit or might be entered by
hand into a spreadsheet or text file. The imported result is a vector point file with the z value as the point identifier. A
vector point file may also be exported to an ASCII XYZ file with this module. The vector X,Y coordinates are written as
the X and Y values and the point identifiers are written as the Z values in the output file.
IMPORT ONLY
PARE creates a proper IDRISI documentation file for band-sequential (BSQ) files. It may also be used to remove (pare
off) a header from the beginning of the file or a trailer from the end of the file.
BILIDRIS and BIPIDRIS pull apart band-interleaved-by-line and band-interleaved-by-pixel files into separate images,
respectively. Both have the optional ability to pare off a header. BILIDRIS may be used with SPOT satellite data, for
example, and BIPIDRIS might be used to read older LANDSAT BIP files or uncompressed 24-bit TIF files.
FLIP is used when the byte order of two-byte integer files needs to be reversed. This is often the case when data are
transferred from UNIX or Macintosh systems to DOS or Windows.
VAR2FIX changes variable-length ASCII files to fixed-length files, with or without end-of-record markers. This module
is often used to prepare DLG files (that are not in the standard optional format) for import with the DLG module.
Finally, SSTIDRIS is used to import Lotus-compatible spreadsheet data when the cells of the spreadsheet are to be
interpreted as cells in the resulting image.
Government/Data Provider Formats Submenu (Import Only)
Currently supported Government and Data Provider file types are:
Landsat Enhanced Thematic Mapper satellite data in NLAPS, FAST,
GeoTIFF or HDF format
imported with Landsat ETM;
SPOT satellite data in GeoTIFF, SPOT Scene (CAP),
or GeoSPOT - SPOTView formats
imported with SPOT;
GeoTIFF (geo-referenced TIFF) format
imported with GeoTIFF;
IDRISI Guide to GIS and Image Processing Volume 1
78
RADARSAT International data
imported with RADARSAT;
Global AVHRR 10-day composite data from the USGS NASA DAAC in Goodes Homosoline projection, is imported
and converted to lat/long with GOODE2LL
USGS Raster Spatial Data Transfer Standard (SDTS)
imported with SDTS;
USGS Digital Line Graphs (Optional Format)
imported with the module DLG;
USGS Composite Theme Grid data
imported with CTG;
and USGS Digital Elevation Models
imported with DEMIDRIS.
Desktop Publishing Formats Submenus
These submenus include modules that import and/or export some of the most popular file types used in Desktop Publishing for graphic data exchange. These include:
Windows Bitmap files (BMP)
imported / exported with BMPIDRIS;
CAD DXF files
imported / exported with DXFIDRIS;
Tagged Information File Format files (TIF)
imported / exported with TIFIDRIS;
JPEG files
imported / exported with JPGIDRIS;
PCX files
imported / exported with PCXIDRIS;
and TGA (Targa) files
imported / exported with TGAIDRIS.
Note that the ability to save the currently-displayed map to a Windows Metafile (WMF) or an Enhanced Windows Metafile format (EMF) is available through the Display System Composer's Save Composition dialog box.
Software-Specific Formats Submenus (Import and Export)
IDRISI facilitates import and/or export of many common GIS and related software file structures. Several of these are
ESRI format files and these are grouped under the submenu ESRI Formats. The other modules are listed in the menus
alphabetically, and are a mixture of raster and vector file types. IDRISI imports raster files to the IDRISI image file format and vector files to the vector file format. The following formats are supported:
ESRI Formats Submenu
ArcView Shape files
imported/exported with SHAPEIDR;
Arc/Info Raster Exchange Format files
imported / exported with ARCRASTER;
Arc/Info GENERATE/UNGEN files
imported/exported with ARCIDRIS;
Atlas*GIS BNA files
imported/exported with ATLIDRIS;
Erdas LAN and GIS files
imported/exported with ERDIDRIS;
GRASS raster files
imported/exported with GRASSIDR;
Map Analysis Package files
imported/exported with MAPIDRIS;
MapInfo Interchange Format files
imported/exported with MIFIDRIS;
Surfer GRD files
imported/exported with SRFIDRIS;
Statistica files (to and from IDRISI images or values files)
imported/exported with STATIDRIS;
Chapter 8 Idrisi Modules
79
Palette files may be imported and exported using the module PALIDRIS. Import is supported for Idrisi for Windows
(.smp), Idrisi for DOS (.pal), ERDAS (.rnb), ILWIS (.col), IAX (.iax), IMDISP (.pal) and Adobe palettes. Export is supported from the current version of Idrisi to Idrisi for DOS, ERDAS, ILWIS and IAX.
Idrisi Vector Export Format files (.vxp) may be imported/exported with Vector Export Format utility.
Use the Idrisi File Conersion (16/32) to convert between earlier 16-bit versions of IDRISI files and the current 32-bit
version.
Finally, choosing Exit will close IDRISI.
The Display Menu
Display Launcher
ORTHO
Media Viewer
Symbol Workshop
COMPOSITE
SEPARATE
ILLUMINATE
HISTO
STRETCH
DISPLAY Launcher is the entry point into IDRISI's extensive display and map composition
system. It allows you to either redisplay an existing composition or launch a new composition
by displaying either a raster image or a vector layer. Whenever a map composition is displayed,
Composer, an interactive facility for the development and modification of map compositions,
automatically appears. The operation of Composer is detailed in the chapter Display System
in this volume.
ORTHO is a facility that creates orthographic perspective (3-D) displays of digital elevation
models (DEMs) or any continuous raster image. It also provides the ability to drape an image
on top of the perspective model, yielding a perspective view of the draped image. The view
direction, height above horizon (azimuth) and the vertical exaggeration are user-defined.
Media Viewer is a presentation utility that can play Windows video (AVI) files as well as
WAV audio files. In addition to playing video and sound, the Media Viewer can create AVI video files from a sequence of
IDRISI images. This can facilitate the demonstration of time series data.
Symbol Workshop allows one to create and modify symbol and palette files. Symbol files for vector point, line, polygon,
and text files indicate how features should be graphically rendered (e.g., their size, color, shape, etc.). For raster image palettes, the color scheme as well as the endpoints used in autoscaling are specified. In all cases, utilities to copy settings and
to blend (interpolate) attributes such as size and color across multiple symbol indices are provided.
COMPOSITE produces a color composite image from three bands of byte binary imagery. Options are available to create 24-bit composites (primarily for visualization) or 8-bit composites (primarily as input to the CLUSTER unsupervised
image classification routine). A 24-bit composite is displayed in true 24-bit color and Composer provides independent onthe-fly adjustment of the three input bands for such images. An 8-bit composite image should be displayed with the color
composite palette.
SEPARATE performs color separation of palette images into RGB components. This is a handy utility for the preparation of print-ready material.
ILLUMINATE is a hill-shading merge facility that uses the color separation process performed in SEPARATE, and the
color space transformations of COLSPACE for a full merge of hill-shading to any palette color image. It creates a truly
dramatic effect for display.
HISTO provides a frequency histogram of the cell values within an image. In addition, HISTO calculates the mean and
standard deviation for the entire image and the specified data range. Two forms of output are supported, graphic and
numeric. Line, bar, and area graph types are offered, as well as cumulative and noncumulative options.
STRETCH is located in this menu as it is used to prepare images for display. It may be used to increase the contrast in
an image and thereby enhance visual interpretation. Three types of stretches are offered: linear, linear with saturation, and
IDRISI Guide to GIS and Image Processing Volume 1
80
histogram equalization. Note that using STRETCH creates a new image with stretched values. A linear stretch of the display is available in Composer for any currently-displayed image, but no new image is created with this process.
The GIS Analysis Menu
Database Query >
Mathematical Operators >
Distance Operators >
At the very heart of GIS is the ability to perform analyses based on geographic location. Indeed, no other type of software can provide this. IDRISI offers a wealth of
analytical tools for geographic analysis.
The GIS Analysis Menu contains eight submenus. The first four are organized by
toolset type, following the logic presented in the Introduction to GIS chapter in this
volume. The remaining submenus are organized according to the particular type of
analysis to be performed. These groupings are primarily for organizational convenience. Most analyses will require the use of tools from multiple submenus.
Context Operators >
Statistics >
Decision Support >
Change / Time Series >
This section contains, in addition to the general descriptions of the individual modules
and their typical usage, further information about the operation and application of the
modules to particular problems.
Surface Analysis >
The Database Query Submenu
RECLASS
OVERLAY
CROSSTAB
Edit
ASSIGN
EXTRACT
HISTO
AREA
RERIM
PROFILE
QUERY
PCLASS
Database Workshop
Image Calculator
Database Query is the most fundamental of GIS operations. Following the module descriptions is a section called Performing Database Query with IDRISI, which further describes
this procedure.
RECLASS produces a new map image by reclassifying the values of an input image. Ranges
as well as individual values may be specified from the original image, while the output values
must be integers and must be given individually. Any original values not previously mentioned will be rounded to integers, but will otherwise remain unchanged.
OVERLAY performs operations between two images, producing a single output image.
There are nine options for OVERLAY, but in the case of database query, the multiply and
maximum options are most often used. In cases where each attribute of interest is represented by a Boolean image, the multiply option will give the Boolean AND, or INTERSECTION result, while the maximum option will give the Boolean OR or UNION result.
CROSSTAB compares two maps with qualitative data. The resulting image contains a
unique value for each unique combination of input values. Statistics about the similarity of
the two input maps may also be generated.
Edit allows access to the IDRISI text editor and may be used to create a values file for use
with ASSIGN to assign new values to an image. The values file lists the values from the original image on the left and the new values to be assigned on the right. Any old value not present in the values file will be
automatically assigned the new value zero. The original values must be integers, but real numbers may be assigned as new
values. The values specified in the values file are assigned to the original values in a feature definition image. The equivalent of ASSIGN may also be used from Database Workshop to assign the values of a database field to an image.
EXTRACT develops summary statistics for the features in a feature definition image, given an input image. The minimum, maximum, sum, average, range, population standard deviation, and sample standard deviation may be calculated.
The results may be saved as an attribute values file, or as a simple text table. The equivalent of EXTRACT may also be
accessed from Database Workshop to fill a field with values summarized from an image.
Chapter 8 Idrisi Modules
81
HISTO is used to display a histogram of the values in an image file. The histogram may be graphic or numeric. Some
summary statistics are also calculated and displayed.
AREA creates a new image by giving each output pixel the value of the area of the class to which the input pixel
belonged. Output can also be produced as a table or an attribute values file in a range of measurement units.
PERIM creates a new image by giving each output pixel the value of the perimeter of the class to which the input pixel
belonged. Output can also be produced as a table or an attribute values file in a range of measurement units.
PROFILE creates profiles over space by querying the values along a linear transect across an image, or over time by querying the value of the same location in several monthly images. The values in the displayed profile may be saved in an
attribute values file.
QUERY extracts pixels designated by an independent mask into a sequential file for subsequent statistical analysis. The
resulting file has the image file format, complete with documentation file, but cannot be displayed as an image in a meaningful way. The file is meant to be used with non-spatial statistical analyses modules, such as HISTO, or to be moved into
a statistical package or spreadsheet software.
PCLASS may be used as an alternative to RECLASS when information about the level of uncertainty in an image is
recorded in the Value Error field of the documentation file. In running PCLASS, you are asked to indicate a threshold,
and whether you would like to evaluate the probability that each pixel exceeds the threshold or is exceeded by the threshold. The resulting image is one of probabilities, ranging between 0 and 1. RECLASS may then be used on the image to
identify those areas that have at most (or at least) a given probability.
Database Workshop may be used for database query on attributes in database files. The results may be linked to a vector
file for display. The equivalent of ASSIGN may be accessed from Database Workshop to apply the results of queries to
image files, and the equivalent of EXTRACT may also be accessed to move extracted statistics into the database. In addition, the Calculate function of Database Workshop may be used to create new attribute fields in the database. The Filter
function may be used to perform the equivalent of RECLASS and OVERLAY in the database. Both the Calculate and
Filter operations are supported through the use of Structured Query Language (SQL) in Database Workshop. For more
information, see the chapter on Database Workshop in this volume.
Finally, Image Calculator is an interactive mathematical modeling tool that allows you to enter a model as a full algebraic
equation using a calculator-like interface. It incorporates the functionality found in the modules OVERLAY, SCALAR,
TRANSFORM, and RECLASS with considerably less work and without the use of macros. Image Calculator allows for
two types of expressions, Mathematical Expressions and Logical Queries, the latter of which is used in database query.
Performing Database Query in IDRISI
Perhaps the most fundamental of analytical operations undertaken in GIS is simple database query, in which we ask questions of the database and examine the results as a map. With a spatial database, two types of questions may be posed—
"What locations have this attribute?" and "What is the attribute at this location?" The first is known as query by attribute,
while the second is called query by location.
Query by attribute may be performed several ways, depending on the geography of the layers. If you are working with a
single geography (e.g. farm fields, provinces) defined by a vector file for which you have multiple attributes in a database,
the database query may be accomplished entirely in Database Workshop using an SQL filter. The results may then be
linked to a vector file for display or they may be assigned to a raster feature definition image for subsequent display.
For example, if you had a map of the countries of the world, and multiple attributes for each country stored in a database,
then you could perform a query such as, "Find all the countries where the median per capita annual income is less than
$5000, but the literacy rate is higher than 60%." The query conditions could be used in an SQL filter in Database Workshop, and the result linked for display to the original vector feature definition file in DISPLAY Launcher. If a new raster
image file should be made, the attributes may be assigned to a raster feature definition image from Database Workshop.
IDRISI Guide to GIS and Image Processing Volume 1
82
However, if the geographies of the attributes of interest are not the same, or if the attributes exist only as image layers,
then two steps are involved. First, the features meeting the conditions specified are selected in each layer. This normally
involves the use of RECLASS or ASSIGN. Then those selected data are used in an overlay operation, provided through
the module OVERLAY, to find the locations that meet all the conditions. (Both steps may be carried out with a single
command in Image Calculator, but behind the interface, the individual steps are still carried out in sequence.)
For example, you might ask, "Where are all the locations that have residential land use and are within a half mile of the
primary path of planes taking off from the proposed airport?" In this case, the geography of land use and that of the flight
paths for the airport are not the same. A Boolean image (zeros and ones only) would be made for each condition using
RECLASS or ASSIGN, then these would be combined in an OVERLAY multiply operation. The resulting image would
have the value of one only where both conditions are found:
Input
Image 1
X
Input
Image 2
=
Output
Image
0
0
0
1
0
0
0
1
0
1
1
1
Reclassification and overlay are fundamental to query by attribute in GIS. In IDRISI, RECLASS and ASSIGN are the
tools used to perform database queries on single attributes, and may be used to produce Boolean images either directly or
through the Image Calculator.
While RECLASS and ASSIGN may be used to produce similar results, there are several important differences between
these modules. Even in cases where either may be used, generally one will be easier to use than the other. The choice will
become more apparent as you become familiar with the characteristics of the two modules.
RECLASS works on an image file. The original image may have byte, integer or real values. However, the new values
assigned may only be byte or integer. Original values may be specified as individual values, or as ranges of values. This
information is entered in the RECLASS dialog box. Any values left out of the specified reclassification ranges will remain
unchanged, except that real values will automatically be rounded to the nearest whole number.
With ASSIGN, a feature definition image file and an attribute values file are required. The latter is commonly created with
Edit or imported from a spreadsheet or statistical software package. The data values in the feature definition image must
be byte or integer. However, the new value to be assigned may be byte, integer or real. Both old and new values must be
specified as single numbers, not as ranges. The old and new values are entered in a values file, rather than in the ASSIGN
dialog box. Any original values not specified in the values file will automatically be assigned the new value zero in the output image.
Whenever the query involves more than one attribute, it is necessary to use OVERLAY. (Again, the user may choose to
use OVERLAY directly, or through the Image Calculator.) For example, to find all agricultural land on soil type 6 requires
that we first isolate soil type 6 as a Boolean image from the soils layer, and the agricultural land as a Boolean image from
the land use layer. These two Boolean images are then overlaid, using the multiplication operation, to find all cases where
it is soil type 6 AND agricultural.
Chapter 8 Idrisi Modules
83
Similarly, the maximum option of OVERLAY may be used to produce the Boolean OR result:
Input Image 1
Max
Input Image 2
=
Output Image
0.00
0.00
0.00
1.00
0.00
1.00
0.00
1.00
1.00
1.00
1.00
1.00
All of the logical operations can be achieved similarly through simple operations on Boolean images. For example, the
Boolean XOR (exclusive OR) operation can be performed with an addition operation, followed by a reclassification of all
values not equal to 1 to 0. In developing models that require this kind of logic, it is often helpful to construct a table such
as the ones above in order to determine the type of IDRISI operations needed.
In Image Calculator, these analyses are built as logical expressions. While Image Calculator often provides a faster and
easier interface, there are advantages, particularly to those new to GIS, to using the modules directly and performing each
step individually. Doing so allows each step in the process to be evaluated so that errors in logic may be detected more
easily. Using the modules individually also allows the user to become more familiar with their operation, facilitating the
use of these modules outside the limits of database query.
Query by location is most easily accomplished in IDRISI with the cursor inquiry tool in the Display System. Select the
cursor inquiry icon and with your cursor placed on the location in question, click the left mouse button. The underlying
data value for that location will be displayed on the screen.
Query by location can be extended to include query across multiple raster files by simply creating a raster image group file
(.rgf) that contains all files pertaining to a particular group. A query by location in any of the grouped images will bring up
information about the pixel value at that location for all the images in the group. Similarly, query by location in a vector
file that has associated database and vector links files (.vlx) will bring up all the linked database field values for the queried
object. Group and link files are created with the Collection Editor, under the File menu.
Other tools for database query by location include PROFILE, QUERY, WINDOW and EXTRACT. Each of these gives
results based on the attributes found at the location of input features.
The Mathematical Operators Submenu
OVERLAY
SCALAR
TRANSFORM
Image Calculator
IDRISI, like most raster geographic analysis systems, provides a set of mathematical tools necessary for complete map algebra.
OVERLAY performs mathematical operations between two input images to produce a single
output image. The options in OVERLAY are:
1.
2.
3.
4.
5.
6.
7.
8.
9.
Add
Subtract
Multiply
Ratio (Divide)
Normalized Ratio
Exponentiate
Minimize
Maximize
Cover
IDRISI Guide to GIS and Image Processing Volume 1
Image 1 + Image 2
Image 1 - Image 2
Image 1 x Image 2
Image 1 / Image 2
(Image 1 + Image 2) / (Image 1 - Image 2)
Each pixel of Image 1, raised to the power in Image 2
The minimum of Image 1 and Image 2
The maximum of Image 1 and Image 2
Image 1 covers Image 2, except where zero
84
All OVERLAY operations work on a cell-by-cell basis. The following illustrates the result of adding two images.
3
2
5
4
6
7
6
5
9
+
5
4
1
4
3
3
2
1
0
=
8
6
6
8
9
10
8
6
9
While OVERLAY performs mathematical operations between two images, SCALAR undertakes arithmetic operations
between a constant and a single image. Its options are: add, subtract, multiply, divide, and exponentiate. In all cases, all
cells in the image are identically operated upon by a single value. For example, exponentiating an image by two will produce a new image where each cell is the square of its previous value. This module is very useful in evaluating image-wise
regression equations.
TRANSFORM undertakes mathematical transformations on the attributes of a single image. Options include natural
logarithms and antilogs, a logit transformation, reciprocal, square and square root, absolute value, and all of the trigonometric operations. The log transform, for example, produces a new image where each cell is the natural logarithm of the
corresponding cell value in the input image.
In summary, OVERLAY operates between two images, SCALAR operates between an image and a constant value, and
TRANSFORM operates only on a single image.
While database query is the most fundamental of GIS operations, map algebra probably represents the second level.
Indeed, map algebra is the core of derivative mapping. For example, from a set of input layers including soil type, land
use, and slopes, we can derive through various mathematical and database query operations a new map of soil erosion
potential.
As with Database Query, these fundamental mathematical operations are also available in Image Calculator. In its application to map algebra, the mathematical expression option of Image Calculator is used.
The Distance Operators Submenu
Display Launcher
ORTHO
Media Viewer
Symbol Workshop
COMPOSITE
SEPARATE
ILLUMINATE
HISTO
STRETCH
COMPOSITE
SEPARATE
ILLUMINATE
The third submenu of analytical tools consists of those that may be called distance operators.
DISTANCE calculates the true Euclidean distance of each cell to the nearest of a set of target
cells as specified in a separate image. Distances are output in the reference units specified in the
target feature image documentation file.
SPDIST is the equivalent of the DISTANCE module, except that it accommodates the special
case of spherical distance units (degrees, radians). SPDIST calculates spherical distances on the
surface of the earth from the designated target features using spherical trigonometry.
COST calculates a distance/proximity surface where distance is measured as the least cost distance in moving over a friction surface. For example, the cost distance from a given location
can be calculated based on a slope image to gauge the difficulty of crossing terrain. COST
requires a target feature image and a friction image as input. IDRISI provides two choices of
algorithm to use in COST distance calculations. The COSTPUSH algorithm uses a pushbroom
logic and is suitable for friction surfaces that are not overly complex. COSTGROW can
accommodate complex friction surfaces, as well as absolute barriers to movement, but requires
a longer processing time.
Chapter 8 Idrisi Modules
85
BUFFER creates buffers around any set of specified features in an image. The user specifies the buffer width from the
features in reference units and indicates the output values for the original features, the buffer zones and areas outside the
buffer zones.
The next set of four modules are used when frictions act with different strengths depending on the direction of movement. For a detailed discussion, consult the chapter on Anisotropic Cost Analysis in the IDRISI Guide to GIS and
Image Processing Volume 2.
VARCOST computes an anisotropic cost surface for movement having motive energy behind it. An example might be a
person walking across a landscape where going up a slope acts as an impediment to movement, while going down the
same slope acts as a facilitation to movement. Since the apparent friction experienced depends on the direction of travel
through it, the friction surface required for input is in the form of a pair of images—a friction magnitude image, and a
friction direction image. In addition, a target feature image is required, and a user-defined function relating direction to
friction may be entered if the default function is not desired.
DISPERSE is similar to VARCOST except in this case, the phenomena for which we are modeling movement has no
motive force behind it. Rather, it is subject to anisotropic forces that cause it to move. An example might be the dispersion
of plant pollen across the landscape, which depends on the force and direction of the wind. A force magnitude and direction image pair is required as input, as well as a target feature image. Again, if the default function relating direction to
force is not desired, a user-defined function may be input.
RESULTANT computes the resultant force vector (as a magnitude and direction image pair) from two input force vector image pairs. This may be used to incorporate multiple sources of anisotropic frictions or forces for subsequent use
with VARCOST or DISPERSE.
DECOMP decomposes a force vector (as a magnitude and direction image pair) into X and Y component images. It also
takes X and Y component images and produces a force vector image pair. This module is most commonly used in developing friction image inputs for VARCOST or DISPERSE. However, it is also useful in applications such as change vector analysis, as outlined in the chapter Change/Time Series Analysis in the IDRISI Guide to GIS and Image
Processing Volume 2.
PATHWAY calculates the route of least cost distance between one or more points and the lowest point on an accumulated cost distance surface produced with COST, VARCOST or DISPERSE. This may be used to identify the least cost
pathway across a landscape, given a cost surface.
ALLOCATE performs spatial allocation based on a distance or cost-distance image derived from DISTANCE, COST,
VARCOST or DISPERSE. For example, if the distance surface were calculated from a set of health centers, ALLOCATE could be used to assign each cell to its nearest health center. The final result of ALLOCATE would be an image
with identifiers indicating to which health center each cell is assigned. The result from ALLOCATE is identical to that of
THIESSEN where simple Euclidean distance is used. However, if cost distance is used, the result may be quite different.
RELOCATE moves non-zero features in an image to a target set of features in another image based on minimum distance. This might be used, for example, to locate villages at the nearest position on a road network.
THIESSEN produces Thiessen polygons around a set of irregularly distributed points. The polygons are derived by
drawing divisions exactly halfway between the original points, such that each pixel is assigned to its nearest source point.
A division of space into polygons of this nature is also known as a Voronoi Tessellation.
IDRISI Guide to GIS and Image Processing Volume 1
86
The Context Operators Submenu
SURFACE
FILTER
The fourth toolset in the Analysis menu contains context operators (also known as local or neighborhood operators). With context operators, each cell in the output image is assigned a value
based on its value in the original image and the values of its surrounding neighbors.
PATTERN
SURFACE calculates either the slope or the aspect of surface cells from a given input image of
terrain heights (a DEM) or any quantitative and continuous variable. SURFACE makes this calGROUP
culation by comparing the heights of cells to those of their neighbors. From this information,
VIEWSHED
the gradient of the slope can be determined for the middle cell, as well as the aspect of the slope
(the direction of maximum descent). The latter is expressed as an azimuth (an angle measured
WATERSHED
clockwise from north). SURFACE is also able to use this information to create a shaded relief
HINTERLAND
image through a technique called "analytical hillshading." Slope calculation is not limited to terPIXEL LOCATION
rain models. It can be quite useful in highlighting areas of rapid change in any quantitative
image. A slope image derived from incidence of disease data, for example, would indicate areas
of relatively homogenous incidence (low slope) and areas of rapid change in incidence (high slope).
TEXTURE
FILTER is most often thought of as an Image Processing function, and in fact appears in that submenu as well. However, it also has application in geographic analysis. FILTER creates a new image by calculating new values using a mathematical operation on the original cell value and its neighbors. The nature of this operation is determined by the values
stored in a 3 by 3, 5 by 5, 7 by 7, or a variable sized template or kernel, that is centered over each pixel as it is processed.
The following filters are available in IDRISI: mean, Gaussian, median, standard deviation, adaptive box, mode, Laplacian
edge enhancement, high pass, Sobel edge detection, and user-defined.
The simplest filter is a mean filter in which the new value is the average of the original value and that of its neighbors. The
result is an image that is "smoother" than the original. Mean and Gaussian filters are often used to generalize an image or
in smoothing terrain data after interpolation by means of INTERCON. The mode filter, which assigns the most common
value to the center pixel, is also commonly used to remove very small areas from a qualitative image, or slivers left after
rasterizing vector polygons. A median filter is useful for random noise removal in quantitative images. The adaptive box
filter is good for grainy random noise and also for data where pixel brightness is related to the image scene but with an
additive or multiplicative noise factor. Edge enhancement filters, such as the Laplacian, accentuate areas of change in continuous surfaces. High pass filters emphasize areas of abrupt change relative to those of gradual change. The Sobel edge
detector extracts edges between features or areas of abrupt change. Finally, the user-defined filter option allows the user
to specify any kernel size as well as the mathematical operation, and is useful for simulation modeling.
PATTERN computes various numerical pattern indices using a 3 by 3, 5 by 5, or 7 by 7 template. It computes measures
that are commonly used in fields such as landscape ecology, based on the values of the template. PATTERN is also useful
for processing of remotely sensed imagery, especially RADARSAT data. Relative richness, diversity, and fragmentation
are just a few examples of the measures provided.
TEXTURE provides options to calculate several measures of variability, fractional dimension, class frequency, and edge
analysis. All these are calculated based on a 3 by 3, 5 by 5, or 7 by 7 template.
GROUP finds polygons in an image by looking for contiguous groups of pixels sharing the same attribute. The groups
are numbered consecutively as they are found, such that each contiguous group of pixels is assigned a unique attribute.
This could, for example, be used to isolate distinct stands of vegetation, where all stands have the same value in the original image. The user may designate whether pixels touching only at a corner should be considered neighbors.
VIEWSHED calculates all cells directly in view of a set of target cells specified on a separate image. To do so,
VIEWSHED extends visual rays in all directions and traces the line of sight to the height of cells to determine whether or
not they are in view.
WATERSHED calculates all cells belonging to the watersheds of one or more target cells. It does so by working from
the target cells outward (i.e. uphill), evaluating each pixel to determine whether or not it receives flow from a neighboring
Chapter 8 Idrisi Modules
87
cell. If a pixel receives flow, it is classified as part of the watershed. This process is continued until a divide is reached.
HINTERLAND determines the supply area dominated by point demand centers. Given a supply map and a second
point demand map, HINTERLAND allows the demand centers to use up supplies beginning at the demand centers and
moving outward, until either their demands are met, or there are no more supplies available. Positive values in the output
image indicate pixels with a surplus, negative values indicate a deficit, and zeros indicate a balance between supply and
demand.
PIXEL LOCATION creates new images representing the X and/or Y coordinate of each cell center. For example, if the
input image were in the Lat/Long reference system, one might use PIXEL LOCATION to create an image of the longitude of each cell center.
The Statistics Submenu
HISTO
EXTRACT
PATTERN
COUNT
REGRESS
MULTIREG
LOGISTICREG
TREND
AUTOCORR
QUADRAT
CENTER
CRATIO
CROSSTAB
VALIDATE
ROC
SAMPLE
RANDOM
STANDARD
Statistics is a field that provides tools for describing groups of numbers. In IDRISI, the Statistics
Submenu provides a series of tools for performing both traditional statistical analysis and specialized spatial statistics routines.
HISTO is perhaps the most often used module of this set. It provides a frequency histogram of
the cell values within an image. In addition, HISTO calculates the mean and standard deviation for
the entire image and the specified data range. Two forms of output are supported, graphic and
numeric. There are several options for the graphic histogram including line, bar and area graphs,
and cumulative and non-cumulative options. The mode may be easily determined from the graphic
display. The numeric display provides a summary of frequencies within each class along with cumulative and proportional frequencies. From this, the median can easily be determined, as well as any
other percentile range.
EXTRACT calculates data summaries for specific features. The features are described in an image
known as a feature definition file, which contains integer identifiers. The summaries are then output
either as a table or as an attribute values file. Summary statistics include minimum, maximum, total,
average, range, mean, and standard deviation.
PATTERN performs pattern analysis based on a comparison of cell values with their surrounding
cells in any of a 3 by 3, 5 by 5, or 7 by 7 pattern. Pattern measures include relative richness, diversity, the dominance index, the fragmentation index, the number of different classes (NDC), center
versus neighbor (CVN), and a binary comparison matrix (BCM). These measures are very commonly used in landscape ecology studies.
COUNT calculates a relative frequency probability image derived from a set of input Boolean images. The probability is
based on the occurrence of non-zero values over multiple images.
REGRESS undertakes linear regression analysis on either two images or two attribute values files. Output consists of a
graphic scatter diagram and trend line, a tabular summary of the regression equation, the correlation coefficient, and several tests of significance.
Note that extreme care should be exercised in the use of the REGRESS module with image data. The regression model
assumes that the observations are independent of one another. With image data, this is rare. As a result, the degrees of
freedom may not accurately reflect the true number of independent observations, leading to a tendency to commit a Type
1 error (accepting the hypothesis when it is not true). A possible approach here is to use SAMPLE or CONTRACT with
the thinning option, to extract a sample of cells that are far enough apart to be considered independent prior to doing the
regression. AUTOCORR can be used to help verify this independence.
MULTIREG performs a multivariate regression analysis between one dependent variable and two or more independent
variables. The analysis can be performed on either images or values files. The same assumptions of independence mentioned in REGRESS above also apply. This analysis is useful for identifying the contribution of factors for any particular
IDRISI Guide to GIS and Image Processing Volume 1
88
outcome. MULTIREG outputs a table of regression coefficients and statistics as well as predicted and residual images.
LOGISTICREG is a logistical regression analysis performed on images or values files. Unlike that in REGRESS or
MULTIREG, the independent variable expresses the probability that an underlying dichotomous variable is true. This
analysis can be useful in explaining the "occurrence" or "nonoccurrence" of phenomena.
TREND is perhaps the most obviously spatial statistical routine. TREND can be considered the spatial equivalent of
REGRESS. TREND calculates the relationship between pixel values and their positions within the image. Three types of
polynomial regression are supported—linear, quadratic, and cubic.
TREND is most often used to determine whether a significant spatial trend can be found in the attribute values of an
image over space. For example, the number of cases reported of a particular disease may be available for villages over a
region. TREND might then be used to determine whether a trend occurs, in the hopes of isolating the spatial direction of
the progress of that disease. Given this relationship, TREND is then able to interpolate a full surface. TREND can also
be used as a generalization routine, decomposing complex surfaces into simpler underlying trends. Additionally, TREND
may be used to interpolate a surface from point data.
AUTOCORR is also distinctively spatial in character. AUTOCORR calculates the first-lag autocorrelation coefficient of
an image using Moran's "I" statistic. The area of analysis may be limited by a mask if desired. The user may choose to calculate autocorrelation based on a pixel's four immediate neighbors (rook's case) or on the eight surrounding neighbors
(king's case). The autocorrelation statistic is essential for the examination of the spatial dependence of attribute values.
While the routine itself can only calculate first-lag (i.e., immediate neighbor) autocorrelation, CONTRACT can be used
with its thinning option to create images representing other lags. Thus, several runs can be used to examine how the autocorrelation changes with distance. This can then be used, for example, to determine the distance necessary for spatial
independence to be assumed. The geostatistics tools available through the Surface Analysis submenu also provide means
for assessing spatial correlation and independence.
QUADRAT performs quadrat analysis, a useful technique for determining the character of a point set's pattern. Point
patterns may be distributed, random or clustered. One of the outputs of QUADRAT is the variance/mean ratio which
can be interpreted to indicate one of these three alternative patterns. QUADRAT also indicates the density of the point
set.
CENTER calculates the mean center (weighted or unweighted) and standard radius for a set of points. The mean center
can be considered the "center of gravity" for a point set while the standard radius is directly analogous to the standard
deviation for non-spatial data. It is used as a measure of the dispersion of points from a most probable location.
CRATIO measures the compactness ratio of defined polygons. This statistic is useful in many habitat studies that wish to
examine the spatial compactness of land cover units such as forest stands. The measure compares the area/perimeter
ratio to that of a circle (the most compact polygon possible).
CROSSTAB provides a comparative technique for qualitative data (e.g. landcover class images). The most fundamental
output of CROSSTAB is a crosstabulation table which lists the frequency with which each possible combination of the
categories on the two images occur. Marginal totals are also provided. A Chi Square statistic (to judge the likelihood of
association), degrees of freedom, and Cramer's "V" statistic (to measure the degree of association) are reported. In cases
where the categories on the two images are identical, CROSSTAB also outputs a Kappa Index of Agreement (KIA—also
known as "KHAT") measure to compare with Cramer's "V". The KIA measure can be evaluated both as an overall value
and on a per category basis—an excellent means of examining change in qualitative data sets.
In addition to this table, CROSSTAB can also create a crosscorrelation image. A crosscorrelation image has new categories to illustrate all existing combinations of the categories on the two input maps. It is thus directly related to the
crosstabulation table. In essence, the crosscorrelation image can be considered as a multiple AND overlay showing all possible combinations of the categories on one image AND those on the other. CROSSTAB also appears in the Change/
Time Series submenu.
VALIDATE provides several statistics for measuring the similarity between two qualitative images. These include spe-
Chapter 8 Idrisi Modules
89
cialized Kappa measures that discriminate between errors of quantity and errors of location.
ROC, the Relative Operating Characteristic, provides a measure of the correspondence between a quantitative modeled
image showing the likelihood that a particular class exists and a Boolean image of that class as it actually occurs (i.e., “reality”).
SAMPLE produces a vector file of points according to a systematic, random or stratified random sampling scheme. It is
useful in many accuracy assessment studies and also provides a general technique for creating samples to be submitted to
statistical analyses. Once a sample is created, variables for those points may be extracted from raster images with
EXTRACT, either directly or through Database Workshop.
RANDOM is useful for simulation studies. It creates a new image of specified dimensions with random values that obey
either a rectilinear, normal, or lognormal distribution, according to a user-specified mean and standard deviation. This
might be used, for example, to produce a probability surface to determine the likelihood of certain events. RANDOM
also has application in error analysis and thus is also found under the Decision Support submenu.
STANDARD is used to convert the values in an image to standard scores. To accomplish this, it subtracts the image
mean from each pixel, then divides by the image standard deviation. Converting images to standard scores may be desirable before comparing them.
The Decision Support Submenu
Decision Making Wizard
WEIGHT
MCE
RANK
MOLA
STANDARD
FUZZY
COUNT
MDCHOICE
PCLASS
BAYES
Belief
RANDOM
SAMPLE
ERRMAT
One of the most important applications of GIS is that of decision support. In fact, many
of the analyses performed with the modules in the other menus of IDRISI are intended
to support decision making. The modules in this menu are unique in that they specifically address multi-objective, multi-criteria resource allocation decision problems, as
well as problems of assessing and incorporating uncertainty in the decision making process. For an in-depth discussion of these issues, refer to the chapter on Decision Support in the IDRISI Guide to GIS and Image Processing Volume 2.
The Decision Making Wizard is an automated assistant that steps you through singleor multi-objective multi-criteria evaluation problems. The Wizard facilitates your use of
WEIGHT, MCE, RANK and MOLA. Your decision rules are recorded at each step so
you may return to change parameters at any time.
WEIGHT computes a best-fit set of weights by calculation of the principal eigenvector
of a pairwise reciprocal comparison matrix in which each factor in a multi-criteria evaluation is compared to every other factor. Information on consensus and procedures for
resolving lack of consensus are provided. Weights derived in this manner sum to one.
The weights are then used with MCE to create a multi-criteria suitability image.
MCE computes a multi-criteria evaluation image by means of either a Boolean analysis,
Weighted Linear Combination (WLC) or Ordered Weighted Averaging (OWA) of factor images. Using WLC, each standardized factor image is multiplied by its weight, then the results are summed. OWA
also works with standardized factor images and employs a variant of the WLC. It takes into account the risk associated
with the decision and degree of tradeoff associated with the variables in the analysis. In both cases, Boolean constraint
maps can be applied to limit areas under consideration in the final analysis. Finally, a strictly Boolean analysis can be
employed using Boolean factor maps to produce a crisp result.
RANK orders the cells in a raster image. Ties may optionally be resolved by using the rank order of a second image. Both
primary and secondary ranks may be in ascending or descending order. The results of this procedure are used extensively
in optimization problems such as with RECLASS for single objective decisions or MOLA for multi-objective decisions.
MOLA is an iterative multi-objective land allocation routine. Input maps are ranked suitability maps such as would be
produced by MCE, followed by RANK. The procedure uses a decision heuristic to resolve conflicts and is suitable for use
IDRISI Guide to GIS and Image Processing Volume 1
90
with massive data sets. Weights for the objectives are defined by the user.
STANDARD converts an image to standard scores. This type of standardization may be one option for standardizing
multiple criteria images to a common scale prior to scaling them to the byte range for combination with MCE or
MDCHOICE.
FUZZY evaluates the fuzzy set membership values (possibilities) of data cells based on any of three membership functions: sigmoidal, j-shaped, and linear. A fourth function allows for a user-defined membership. Monotonically increasing,
monotonically decreasing, symmetric, and asymmetric variants are supported. FUZZY is used to model transitional
changes in class membership in an image and is also used to standardize factor images for MCE using subjective knowledge of suitability. Other Fuzzy Set operations such as CON (concentration), DIL (dilution), AND and OR are covered
by the standard modules TRANSFORM and OVERLAY.
COUNT calculates a relative frequency probability image derived from a set of input Boolean images. The probability is
based on the occurrence of non-zero values over multiple images.
MDCHOICE resolves conflicts between competing objectives by means of a multiple ideal-point procedure. Axes in the
multidimensional decision space can be differentially weighted and minimum suitabilities set for each. Input maps should
be standardized either with STANDARD or by means of the histogram equalization procedure in STRETCH.
The remaining modules in this submenu are used in the evaluation and handling of error in geographic analysis.
PCLASS evaluates the probability with which data cells exceed or are exceeded by a specified threshold based on the
stated RMS error for the input map. In Bayesian terms, this procedure evaluates the probability of the evidence given a
hypothesis in the form of a threshold—p(e|h).
BAYES evaluates Bayes' Theorem. Multiple evidence maps (such as those produced by PCLASS) are permitted as long as
they are conditionally independent. Prior probabilities may be input in map form (and thus, like the evidence maps, may
vary continuously over space). This is an extension to what is sometimes called a Bayesian Weight-of-Evidence Approach.
The user may also specify the confidence in the decision rule, i.e., the belief that the evidence supportive of the hypothesis
is truly reflected in the evidence at hand—p(e|e').
Belief evaluates the degree to which evidence provides concrete support for a hypothesis (belief) and the degree to which
that evidence does not refute the hypothesis (plausibility). Belief employs the Dempster-Shafer Weight-of-evidence Procedure in the evaluation. Unlike Bayesian Probability Analysis, Belief explicitly recognizes the possibility of ignorance in
the evaluation, i.e., the incompleteness of knowledge or evidence in the hypothesis.
RANDOM creates random images according to rectilinear, normal or log-normal models. This module is particularly
important in the development of Monte Carlo simulations of error propagation.
SAMPLE creates systematic, random, and stratified random point sampling schemes.
ERRMAT produces an error matrix analysis of categorical map data compared to ground truth information. It tabulates
errors of omission and commission, marginal and total errors, and selected confidence intervals. Per-category Kappa
Index of Agreement figures are also provided.
Chapter 8 Idrisi Modules
91
The Change / Time Series Submenu
IMAGEDIFF
IMAGERATIO
CVA
CALIBRATE
CROSSTAB
PROFILE
TSA
CORRELATE
Media Viewer
MARKOV
STCHOICE
DISAGGREGATE
NORMALIZE
CELLATOM
CA_MARKOV
Change and time series analysis is an important application area for GIS and Image Processing.
There is an ongoing need to identify and quantify change, as well as to predict the effects of
change on the environment, at scales ranging from local to global. For more information about
this application area, see the chapter on Change and Time Series Analysis in the IDRISI
Guide to GIS and Image Processing Volume 2.
The most simple type of change analysis is a comparison between images from two dates.
IMAGEDIFF compares two quantitative images of the same variable for different dates. Output may be a simple difference image, a percentage change ((2nd image - 1st image) / 1st
image), a standardized difference image showing z values or a standardized class image in which
z values have been reclassified into 6 classes. A mask file may be specified for each image to
limit the area under analysis.
IMAGERATIO compares two quantitative images of the same variable for different dates
through ratioing. Options are offered to create either a ratio image (later image / earlier image)
or a log ratio image (ln(later image / earlier image).
CVA (Change Vector Analysis) provides a means for comparing two-band sets of images for
two dates. Both the magnitude and direction (character) of change are derived.
VALIDATE
CALIBRATE may be used to address problems of non-comparability due to sensor changes
between two images from different dates. Three types of information may be used to calibrate
the input image: regression with a reference image, user-defined offset and gain, and userdefined mean and standard deviation. Regression with a reference image is the most common, with the earlier image as
the input file and the later image as the reference file. The module computes a linear regression equation relating the two
images. This equation is then used with the earlier image to produce a predicted later image. This predicted later image is
really an adjusted earlier image, accounting for systematic sensor differences. It can then be compared with the actual later
image through quantitative change analysis techniques.
ROC
CROSSTAB may be used to compare two qualitative images, such as land cover images, from two dates. A new image
may be produced in which a unique identifier is assigned to every combination of original values. RECLASS or ASSIGN
might then be used to create an image with change and no change categories. A cross-classification matrix and statistics
on the similarity of the two images may also be generated. In cases where the categories of both images are the same, percategory Kappa Index of Agreement figures will be calculated. This provides important information about the types of
change occurring in the image. The overall Kappa, as well as the Chi Square and Cramer's V give information about the
degree of change the entire image has experienced.
To analyze change over multiple dates, the following four modules may be used.
PROFILE, when used in the context of change and time series analysis, is used to extract statistics for specific areas of
an image over multiple dates. For example, one might create a polygon of a forested area, then extract the average NDVI
(a measure of the amount of biomass present) over several dates. The trend of these data would give some indication of
the type of change occurring in the area. PROFILE will plot the data as a graph, and can also write the values to a file for
subsequent analysis.
TSA stands for Time Series Analysis, and provides standardized Principal Components Analysis for time series data. The
input to TSA is in the form of an IDRISI time series file, which simply lists, in order, the images to be analyzed. TSA will
accept up to 256 input images and can produce up to an equal number of component images. Loadings graphs are output
either as values files or as DIF-format data files that can be used with virtually all spreadsheet software systems. Standardized Principal Components Analysis has been shown to be quite effective in extracting elements of change from time
series data. The first component produced is the typical condition over the entire time series, then subsequent images represent different change events. Graphs of the loadings yield information about the temporal patterns of change events.
IDRISI Guide to GIS and Image Processing Volume 1
92
Input a mask image to limit the area to be analyzed.
CORRELATE calculates the Pearson Product Moment Coefficient of Correlation between a set of values in an attribute
values file and the values through a time series of images for each pixel of an image. The main use of the module is to
identify areas that correlate well with a particular temporal pattern of interest. The pattern of interest is defined by the values in the attribute values file. In the resulting image, high values show areas that better match the pattern of the values
file.
Media Viewer may be used to create video clips of the images in a time series. This is a powerful device for visual change
detection.
The following six modules are used in modeling future change.
MARKOV analyzes two qualitative land cover images from different dates and produces a transition matrix, a transition
areas matrix, and a set of conditional probability images. The transition matrix records the probability that each land cover
category will change to every other category while the transition areas matrix records the number of pixels that are
expected to change from each land cover type to each other land cover type over the specified number of time units. The
conditional probability images report the probability that each land cover type would be found at each pixel after the
specified number of time units and can be used as prior probability images in Maximum Likelihood Classification of
remotely sensed imagery.
STCHOICE (Stochastic Choice) creates a stochastic land cover map by evaluating the conditional probabilities that each
land cover can exist at each pixel location against a rectilinear random distribution of probabilities. It does so by generating a random number. The conditional probabilities for each class are then summed beginning with the first image listed
in the group file. The class represented by the image that makes the sum exceed the random threshold is assigned. The
resulting image will appear to be quite pock-marked. Because the random element is introduced, a different output image
may result each time the module is run.
DISAGGREGATE generates a sequence of calls to other IDRISI modules in order to redistribute the conditional probabilities of a particular land cover type according to a designated pattern. The user provides a model for the disaggregation. Note that the output image is designed to have the same sum of probabilities as the original image (on a per-category
basis).
NORMALIZE linearly adjusts the values for a set of quantitative images so the values sum to 1.0 at each pixel. For
example, a set of image showing monthly rainfall for a year might be normalized so the resulting images show the percent
of annual rainfall accumulated each month.
CELLATOM provides for cellular automata in Idrisi. This is typically used in dynamic modeling where the future state
of a pixel depends upon its current state and the states of its neighbors. The rules for changing states are governed by a
filter file and a reclass file that may be modified. The user specifies the number of iterations to perform. With each iteration, the filter is applied to the image then the resulting image is reclassified according to the reclass file. This output
image is then used as input for the next iteration.
CA_MARKOV is a combined cellular automata / Markov change land cover prediction procedure that adds an element
of spatial contiguity as well as knowledge of the likely spatial distribution of transitions to Markov change analysis.
VALIDATE provides several statistics for measuring the similarity between two qualitative images. These include specialized Kappa measures that discriminate between errors of quantity and errors of location.
ROC, the Relative Operating Characteristic, provides a measure of the correspondence between a quantitative modeled
image showing the likelihood that a particular class exists and a Boolean image of that class as it actually occurs (i.e., “reality”).
Chapter 8 Idrisi Modules
93
The Surface Analysis Submenu
Interpolation >
Geostatistics >
Topographic Variables >
Feature Extraction >
The Surface Analysis submenu contains four headings, each leading to further submenus. While the descriptions of the modules in these submenus often refer to elevation data and digital elevation models as examples, the modules available in the Surface
Analysis submenu provide a powerful set of analytical techniques that can be applied to
any continuous quantitative data.
Interpolation Submenu
INTERPOL
INTERCON
The first of the Surface Analysis submenus is Interpolation. The issues surrounding surface
interpolation as well as the options available in IDRISI are discussed in detail in the chapter
Surface Interpolation in the IDRISI Guide to GIS and Image Processing Volume 2.
TIN Interpolation >
INTERPOL interpolates a surface, according to either a distance-weighted average or a
potential model, given a vector input file of points, and the values of the variable at those
THIESSEN
points. With both models, the exponent associated with the distance weight is user-defined.
TREND
For example, if the user specifies 2, the interpolation process uses 1/d2. A variable search
radius is used that aims to use the 6 closest known points from the vector point in undertaking
the interpolation of any grid cell, and ensures that no fewer than 4 and no more than 8 are ever used. The search radius
can be disabled, in which case all control points are used in the interpolation of each point.
Kriging >
INTERCON interpolates a surface from a set of digitized lines. This module was designed specifically to facilitate the
creation of Digital Elevation Models from contour lines. One starts with digitized contours such as might be created with
CartaLinx or purchased from a data vendor. Then the vector contours are rasterized (with the use of INITIAL and LINERAS). INTERCON is then run on the rasterized line file to produce the interpolated model. Finally, it is common to
run FILTER to smooth the surface (using a mean filter) since the linear interpolation procedure tends to create a slightly
faceted surface.
TIN Interpolation is discussed in detail in the chapter Triangulated Networks and Surface Generation in the
IDRISI Guide to GIS and Image Processing Volume 2. The modules used to prepare data for TIN generation, to
create and optimize the TIN, and to interpolate a full surface from a TIN are all found in this submenu.
TIN creates a triangulated irregular network from point data. The input data may be true point data or iso-line data. In
the latter case, the vertices of the lines are used in the triangulation. TIN offers the options to constrain the TIN to isolines and to optimize the TIN for features known as bridge and tunnel edges. TINSURF uses the TIN and the original
point attribute data to interpolate a full raster surface model. Selecting the create raster surface option on the TIN dialog
will automatically launch the TINSURF dialog, The other three modules in this submenu, LINTOPNT, TINPREP and
PNTGEN are used to prepare the input vector data for triangulation. LINTOPNT creates a point vector file from the
vertices of an input line file. This is useful in visually assessing the density of the points in the line file. TINPREP adds or
removes points along an iso-line given a user-specified tolerance distance. PNTGEN thins vector point data according to
a user-defined radial search distance.
The Kriging submenu leads to three interfaces to the Gstat geostatistical modeling software package.47 The chapter Geostatistics in the IDRISI Guide to GIS and Image Processing Volume 2 gives background on the field of geostatistics
and the functions provided through these interfaces. In the Spatial Dependence Modeler interface, the user employs a
wide range of tools to learn about the patterns of spatial dependence in the sample data set. In the Model Fitting inter47. Idrisi32 provides a graphical user interface to Gstat, a program for geostatistical modeling, prediction and simulation written by Edzer J. Pebesma
(Department of Physical Geography, Utrecht University). Gstat is freely available under the GNU General Public License from http://
www.geog.uu.nl/gstat/. Clark Labs' modifications of the Gstat code are available from the downloads section of the Clark Labs website at http://
www.clarklabs.org.
IDRISI Guide to GIS and Image Processing Volume 1
94
face, the user defines mathematical models to describe the covariance relationships among sample data. In the Kriging
and Simulation interface, full raster surfaces may be created from sample data and the models developed through the
other interfaces. These interfaces are presented here as tools for surface interpolation. The use of these geostatistical techniques is broader than interpolation and they are therefore also available through the Geostatistics submenu.
THIESSEN creates Thiessen polygons around a set of irregularly distributed points, such that every pixel in the resulting
image is assigned the identifier of the point to which it is nearest. THIESSEN does not produce a smooth surface as the
other modules in this submenu do, but rather a stepped surface.
TREND calculates linear, quadratic, and cubic trend surfaces from an irregular set of points, and produces an interpolated surface as a result. TREND may be used as an interpolator in cases where it is known that some error exists in the
control points and the degree of variation in the surface is small. TREND also appears in the Analysis/Statistics submenu.
Geostatistics Submenu
Spatial Dependence Modeler
Model Fitting
Kriging and Simulation
The second submenu in the Surface Analysis submenu is Geostatistics. The field of
geostatistics has a broad range of application to many types of data and analyses.
The chapter Geostatistics in the IDRISI Guide to GIS and Image Processing
Volume 2 gives a broad overview of Geostatistics and the functions available in
IDRISI through the three interfaces to Gstat (see footnote on previous page).
In the Spatial Dependence Modeler interface, the user employs a wide range of tools to learn about the patterns of spatial dependence in the sample data set. In the Model Fitting interface, the user defines mathematical models to describe
the covariance relationships among sample data. In the Kriging and Conditional Simulation interface, full raster surfaces may be created from sample data and the models developed through the other interfaces.
Topographic Variables Submenu
SLOPE
APSECT
HILLSHADE
CURVATURE
FRACTAL
The Topographic Variables submenu is the third submenu of the Surface Analysis group and contains modules that operate on surface images to calculate a variety of measures. The SLOPE,
ASPECT and HILLSHADE entries all open the SURFACE module interface. SURFACE calculates these variables by comparing the heights of cells to those of their neighbors (thus, it is also
found in the Context Operators Submenu). From this information, the gradient of the slope can be
determined for the middle cell, as well as the aspect of the slope (the direction of maximum
descent). The latter is expressed as an azimuth (an angle measured clockwise from north). SURFACE is also able to use this information to create a shaded relief image through a technique called
"analytical hillshading."
The CURVATURE module calculates the maximum rate of change of a curve fit through a pixel in both the direction of
aspect and also in the direction orthogonal to aspect. The values of the pixel and its immediate neighbors are used in the
calculation.
FRACTAL calculates the fractal dimension of a surface using a 3 by 3 neighborhood. This calculation is also available
through the TEXTURE module.
Chapter 8 Idrisi Modules
95
Feature Extraction Submenu
CONTOUR
TOPOSHAPE
PIT REMOVAL
RUNOFF
FLOW
WATERSHED
The final submenu in the Surface Analysis group, Feature Extraction, includes two modules.
CONTOUR creates vector iso-lines (i.e., lines of constant value) from a continuous surface
raster image. The user specifies the contour interval to use. Options are available for cartographic generalization of the output vector file. The resulting vector file is most typically used
to enhance visualization of continuous surfaces or for export to other software packages.
TOPOSHAPE classifies a surface into eleven different features: peak, ridge, saddle, flat,
ravine, pit, convex hillside, saddle hillside, slope hillside, concave hillside, and inflection hillside. Any pixels not assigned to these classes are assigned to the “unclassified” class.
PIT REMOVAL creates an adjusted "depressionless" DEM in which the cells contained in depressions are raised to the
lowest elevation value on the rim of the depression. This is an important preparation step in many flow models.
RUNOFF calculates the accumulation of rainfall units per pixel as if one unit of rainfall was dropped on every location.
Using the RECLASS module, a threshold can be applied to the output to produce drainage networks.
FLOW calculates the flowdirection from each pixel into its next “downhill” neighbor. Unlike ASPECT, which gives a
continuous output of directions, the output of FLOW is restricted to the eight neighboring cells for each pixel. This output is useful in many flow and hydrological applications.
WATERSHED identifies areas that contribute flow to a part of the drainage network. The user either indicates the minimum size of watershed desired and the module breaks the entire image into watersheds of this size or larger or the user
indicates one or more “seed” features for which to identify watersheds.
The Image Processing Menu
Restoration >
Enhancement >
Transformation >
Fourier Analysis >
Signature Development >
Alongside the geographic analytical operators found in IDRISI, the Image
Processing capabilities round out a full suite of tools for the processing of
spatial data. The Image Processing functions fall into ten categories: restoration, enhancement, transformation, Fourier analysis, signature development, hard classifiers, soft classifiers, hardeners, hyperspectral analysis and
accuracy assessment.
For background information, consult the Introduction to Remote Sensing and Image Processing chapter in this volume. Also, several chapters
in the IDRISI Guide to GIS and Image Processing Volume 2 provide
Hardeners >
in-depth discussions of particular image processing tasks. The Image ResHyperspectral Image Analysis >
toration chapter discusses issues of geometric and radiometric correction
Accuracy Assessment >
and the IDRISI modules designed for these purposes. The Classification
of Remotely Sensed Imagery chapter provides more detailed information
on the classification process and a tour of the IDRISI classification operators. The Fourier Analysis chapter gives
detailed information about the use and function of FOURIER and its companion modules in the Fourier Analysis submenu. Finally, the Tutorial includes an extensive set of exercises covering many aspects of image processing.
Hard Classifiers >
Soft Classifiers / Mixture Analysis >
IDRISI Guide to GIS and Image Processing Volume 1
96
Restoration Submenu
RESAMPLE
LOCAL AFFINE
MOSAIC
DESTRIPE
Image Restoration is the manipulation of remotely sensed images in an attempt to remove
known value distortions. Restoration can be geometric or radiometric. The first module performs geometric corrections which are used to reduce distortion at the edges of the image and
to register the image to a coordinate system. The other three modules perform radiometric restoration for the removal or diminishment of distortions in the data values of the images.
RESAMPLE performs geometric restoration of images, especially in the case of remotely
sensed imagery which, in its raw state, is without real world coordinates. RESAMPLE is used to
georegister an image or vector file to a reference system or to another file. It takes the coordinates of a set of control points in an existing file, and in the desired new reference system, and
converts the file to the new reference system by means of either a linear, quadratic or cubic polynomial mapping function.
(In the case of satellite imagery, a simple linear resampling is sufficient in most instances.) With raster images, cells in the
new grid rarely match up in any simple way with the original grid. Therefore, new cell values are estimated by resampling
the old grid (hence the name of the module). When an image or vector file is already georeferenced to a known reference
system but must be transformed to another reference system, the PROJECT module in the Reformat menu is more
appropriate.
RADIANCE
NDVICOMP
LOCAL AFFINE is used to rectify images that have embedded in them a grid of control points with precise known
locations.
MOSAIC automates color balancing when adjacent overlapping images are joined into a single larger image. The areas of
overlap are used to determine the color adjustments.
DESTRIPE removes the striping caused by variable detector output in scanned imagery. It works by calculating a mean
and standard deviation for the entire image and for each detector separately. The output from each detector is scaled to
match the mean and standard deviation of the entire image.
RADIANCE converts raw LANDSAT data values to calibrated radiance using lookup tables of gain and offset values.
Conversion to radiance is used to facilitate comparisons between images from different dates.
NDVICOMP is intended for the production of temporal composite images of NDVI imagery, most typically for purposes of cloud removal. Two formats are provided. The first produces maximum value composites in which the final output value of each cell represents the maximum of the values in a set of input images. The other produces output cells
equal to the quadratic mean of the input cells.
Simple haze removal can be accomplished with the SCALAR module, which is found in the Analysis/Mathematical
Operators submenu. The linear with saturation option in STRETCH may often be used to produce the same result.
When DESTRIPE is not applicable or does not perform well, Principal Components Analysis (PCA) or Fourier Analysis
may provide solutions for destriping satellite imagery. See the chapter on Image Restoration in the IDRISI Guide to
GIS and Image Processing Volume 2 for details.
Enhancement Submenu
STRETCH
COMPOSITE
FILTER
Image enhancement is the modification of image values to highlight information within the image.
Most often these enhanced images are used in visual analysis only, while the original images are
used for automated analyses. The IDRISI display system includes some facilities for enhancement
of the screen display. These include the ability to interactively set the endpoints used in applying the
color palette to images. No new files are created with the Display tools, however. To create new
images that are enhanced, the following three modules are often used.
STRETCH is a contrast stretch utility used to visually enhance images. Options include linear stretch, linear stretch with
saturation, and histogram equalization. A linear stretch takes the maximum and minimum values for the image as a whole
Chapter 8 Idrisi Modules
97
and assigns the intervening values to a user specified number of classes. A linear stretch with saturation uses a user-specified percent saturation, and forces that percent of the distribution at the upper end to have the maximum value, and that
percent at the lower end to have the minimum value. The intervening values are assigned to the user-specified number of
classes. A histogram equalization stretch divides the histogram into classes containing an equal number of pixels (or as
close to that as possible), given the condition that pixels with the same original value may not be broken into different
new classes.
COMPOSITE produces a color composite image from three bands of byte binary imagery. To create a composite image,
you must have three bands of imagery that are registered to each other. Options are available to create 24-bit composites
(primarily to visualization) or 8-bit composites (primarily as input to the CLUSTER unsupervised image classification
routine). A 24-bit composite is displayed in true 24-bit color and COMPOSER provides independent on-the-fly adjustment of the three input bands for such images. An 8-bit composite image should be displayed with the color composite
palette.
FILTER changes the values of all pixels in an image, based on each pixel's original value and those of its neighbors. The
nature of this operation is determined by the values stored in a variable sized template, or kernel. The pixel and its neighbors are multiplied by the values stored in the corresponding positions of the template, and the resulting values are
summed to arrive at a new value for the center pixel. The following filters are available in IDRISI: mean, Gaussian,
median, standard deviation, adaptive box, mode, Laplacian edge enhancement, high pass, Sobel edge detection, and userdefined.
The simplest and the most commonly used filter in Image Processing is a mean filter. In operation, the new value is the
average of the original value and those of its neighbors. The result is an image that is "smoother" than the original. Mean
and Gaussian filters are often used to generalize or "smooth" an image and are effective for random noise removal. The
mode filter assigns the most common value to the center pixel of the kernel. A median filter is most commonly used for
random noise removal. The adaptive box filter is good for grainy random noise and also for data where pixel brightness is
related to the image scene but with an additive or multiplicative noise factor. Edge enhancement filters, such as the Laplacian, accentuate areas of change in continuous surfaces. High pass filters emphasize areas of abrupt change relative to
those of gradual change. The Sobel edge detector extracts edges between features or areas of abrupt change. Finally, the
user-defined filter option allows the user to specify any kernel size as well as the mathematical operation, and is useful for
simulation modeling.
Transformation Submenu
PCA
TSA
COLSPACE
TEXTURE
THERMAL
VEGINDEX
TASSCAP
PCA provides both standardized and unstandardized principal components analysis. Principal components analysis is a mathematical technique closely related to factor analysis that transforms an
original set of image bands, typically from LANDSAT or SPOT, into a new set of components that are
uncorrelated and are ordered in terms of the amount of the original variance explained. Often the
first two or three components explain virtually all of the variance because of substantial correlation
between the bands. As a result, unstandardized PCA is commonly used to determine which bands
are carrying the most information in a data set and for data compaction. PCA can also be used to
remove image correlation. Standardized PCA has been shown to be a powerful tool for time series
analysis, and is offered for that purpose in the module TSA. Using TSA is the same as the standardized option in PCA, except that PCA is limited to 12 input images while TSA can work with up to
84 input images.
COLSPACE performs Hue/Lightness/Saturation (HLS) to Red/Green/Blue (RGB) and vice versa color space transformations. This is especially useful for merging high resolution panchromatic imagery with lower resolution multispectral imagery. Three bands of the original multispectral imagery are input as RGB and output as HLS. The Lightness band
is then discarded and the (often enhanced) panchromatic imagery is substituted. The three bands are registered to each
other, then transformed from HLS to RGB. The resulting three images may then be combined in a color composite
image. The resulting image retains much of its spectral resolution and has a higher spatial resolution than the original
multi-spectral bands. COLSPACE may also be used to integrate a hillshaded model into a composite image following the
IDRISI Guide to GIS and Image Processing Volume 1
98
same procedure as outlined above, substituting a hillshade model rather than a panchromatic image.
TEXTURE performs texture analysis of an image and is useful for the processing of remotely sensed imagery. The
TEXTURE analysis provided includes variability, fractal dimension, and convolution filtering.
THERMAL converts LANDSAT TM Band 6 raw data values to blackbody temperatures, either Celsius, Kelvin or Fahrenheit.
VEGINDEX computes a large variety of vegetation indices from remotely sensed images. See the Chapter on Vegetation Indices in the IDRISI Guide to GIS and Image Processing Volume 2 manual.
TASSCAP performs the Tasseled Cap transformation. This transformation yields a variety of useful indices for assessing
vegetation biomass and moisture conditions.
Fourier Analysis Submenu
FOURIER
ZEROPAD
FILTERFQ
FREQDIST
DRAWFILT
The modules in the Fourier Analysis submenu support the application of Fourier Analysis, a transformation between spatial and frequency domains. The chapter Fourier Analysis in the IDRISI
Guide to GIS and Image Processing Volume 2 manual provides detailed information on the use
of FOURIER and its companion modules as well as the interpretation of results.
FOURIER allows for the transformation of images from the spatial domain to the frequency
domain and back again.
FOURIER requires that the number of rows and columns in an image be a power of two. ZEROPAD is used to prepare images that do not meet these conditions so they may be used with FOURIER.
FILTERFQ, FREQDIST and DRAWFILT all facilitate the creation of filters to be applied to frequency domain
images to enhance, suppress or remove particular frequencies prior to performing a reverse Fourier Transform. FILTERFQ offers 26 types of filters, each with several user-defined options. FREQDIST creates a frequency distance image
that may then be manipulated with RECLASS or FUZZY. DRAWFILT provides an interactive display utility in which
the user may trace with the cursor particular frequencies to be masked out.
Signature Development Submenu
MAKESIG
ENDSIG
FUZSIG
HYPERSIG
HYPERAUTOSIG
SIGCOMP
SEPSIG
SCATTER
Signature Development is most often associated with the first stages of supervised classification
and typically requires two steps—the creation of training sites and the creation of signature files
from the training sites. Training sites are examples of informational classes, e.g., forests, urban
or rangeland, which can be characterized across all bands of imagery. These characterizations
are then used to create signatures or spectral response patterns for each informational class.
Delineation of training sites is often accomplished through on-screen digitizing in the IDRISI
Display System or through the import of GPS data collected in the field. The second step, signature development, is accomplished with the use of the modules in this submenu.
MAKESIG creates signature files for each informational class for which you have created training sites. It does so by extracting and analyzing the pixels identified in the training sites for each
informational class for each band being used. Statistics such as the mean, minimum, maximum,
varianc,e and covariance between the pixels in the separate bands are developed.
ENDSIG is used to create end-member (i.e., pure) signatures for use with the UNMIX module. These endmember signatures contain the spectral mean for the class on each band. These values are input by the user, and the vaiance information in the signature file is set to zero.
FUZSIG is a variant of MAKESIG. Unlike MAKESIG which assumes the signature data are from areas that are pure
examples of any particular class, FUZSIG produces signatures from data that are assumed to be inherently fuzzy or
Chapter 8 Idrisi Modules
99
ambiguous in character. This is often used in the case of mixed pixel analysis.
HYPERSIG extends the logic of signature development to the special case of hyperspectral data. HYPERSIG creates
and displays hyperspectral signatures either from training site data or from spectral curve library files.
HYPERAUTOSIG automatically develops signatures for hyperspectral image data based on the Linear Spectral Unmixing logic. Given a raster group file of hyperspectral images, a set of hyperspectral signature files (.hsg) and a signature
group file (.sgf) is produced. The number of signatures to develop is set by the user and must be less than or equal to the
number of bands in the hyperspectral image set.
SIGCOMP graphically displays the signatures created with MAKESIG for comparison purposes. You may choose which
signatures to display and whether to display the mean, minimum and maximum, or all three. Overlapping signatures may
indicate heterogeneous training sites. With a parallelepiped classification, at least one of the bands should show completely separated minimum/maximum ranges. If not, ambiguity in the classification exists. Likewise, signatures that are
nearly coincident in their means will be difficult to separate with the Minimum Distance to Means and Maximum Likelihood classifiers.
SEPSIG provides statistical measures of the separability of signatures over a given set of bands. It is useful in determining
which bands should be used to best differentiate particular cover types.
SCATTER is used as a further means to compare the relationship of signatures in relation to their bands. SCATTER creates a scattergram of the band space between two images. The signatures can be plotted in the band space to further evaluate signature spacing.
Hard Classifiers Submenu
PIPED
MINDIST
MAXLIKE
FISHER (LDA)
HYPERSAM
HYPERMIN
CLUSTER
ISOCLUST
MAXSET
There are two basic approaches to the classification process: supervised and unsupervised classification. IDRISI provides the tools for both processes with both hard and soft classifiers available.
This submenu lists the hard classifiers available for both supervised and unsupervised classification.
PIPED is a Parallelepiped classifier. It is generally considered to be the fastest but least accurate
of the classifiers available in the IDRISI set. For parallelepiped classification, pixels are assigned
to a given informational class if either the reflectance values on each band fall within the minimum and maximum values recorded for that signature or the values fall within a specified number
of standard deviations. This procedure provides no assurance that a pixel will not fall into several
classes, which often happens. The class assigned to a pixel is the last signature encountered that
meets these criteria. To make use of this, specify your signatures in reverse order of importance or
likelihood, starting with the least important or likely ones, and ending with those that are most
important or likely.
MINDIST is a Minimum Distance to Means classifier in which pixels are assigned to the class they are closest to in band
space, a coordinate system defined by orthogonal axes representing each band. It is the second fastest routine, and produces the best results when training sites are of less than perfect quality. The procedure will, by default, classify each pixel
with no ambiguity. You can specify a maximum distance, either in raw Dn units (0-255) or in standard deviations, so that
pixels beyond that radius are left unclassified. The use of standard deviations allows the system to accommodate differences in spectral variance between informational classes.
MAXLIKE is a Maximum Likelihood classifier. When using MAXLIKE, a full multidimensional probability function is
evaluated to determine the likelihood that any pixel belongs to a given class. By default, each pixel is assigned to the most
likely class, regardless of how likely or unlikely that maximum may be. This is the most advanced technique, but also the
slowest, for image classification in IDRISI. This technique produces the best result when training sites are well sampled,
i.e., each site has 50-100 times as many pixels as there are bands in the data set, and the site is also strongly homogeneous.
As with MINDIST, you have the opportunity of leaving any pixel unclassified where the probability that it belongs to any
IDRISI Guide to GIS and Image Processing Volume 1
100
particular class falls too low. With this preocedure, you have the option to enter prior knowledge of the likelihood with
which each class occurs in the image. These prior probabilities can either be entered as values, as images, or as a combination of values and images for each signature. Without a priori knowledge, the system assumes all classes are equally likely
for the purposes of classification.
FISHER provides image classification based on discriminant analysis.
HYPERSAM is a spectral angle mapper hard classifier for hyperspectral data. The spectral angle mapper algorithm is a
minimum-angle procedure specifically designed for use with spectral curve library data.
HYPERMIN is a minimum-distance hyperspectral hard classifier specifically intended for use with image-based signatures developed using training sites. It employs the same logic as MINDIST using standardized distances.
CLUSTER performs an unsupervised classification and uses a variant of the histogram peak technique to create a new
image of like clusters from an 8-bit composite image created with COMPOSIT. You have a choice of broad or fine classifications. The broad classification gives a general picture of spectral classes while the fine classification gives intricate
details. The clusters are ranked in terms of how much of the image they describe. The resulting cluster image must then
be interpreted by the analyst, matching the clusters to their appropriate land cover categories.
ISOCLUST is an iterative self-organizing cluster analysis procedure and is a variant on the ISODATA routine. Like
CLUSTER, ISOCLUST is an unsupervised classifier. However, in contrast to CLUSTER, input to ISOCLUST may
include up to seven image bands. With ISOCLUST, the user determines, or seeds, the number of clusters to be uncovered
in the data. Through an iterative process, cells are assigned to their nearest cluster in band space.
MAXSET is a hard classifier that assigns to each pixel the class with the greatest degree of commitment based on a full
Dempster-Shafer hierarchy describing all classes and their hierarchical combination. MAXSET is an unsupervised classification in that it can assign a pixel to a class for which no exclusive training data have been supplied. MAXSET can be
instrumental in identifying mixed-class pixels.
Soft Classifiers / Mixture Analysis Submenu
BAYCLASS
BELCLASS
FUZCLASS
UNMIX
HYPERUSP
HYPEROSP
HYPERUNMIX
HYPERABSORB
MIXCALC
Belief
Unlike hard classifiers, soft classifiers defer making a definitive judgment about the class membership of any pixel and instead make groups of statements about the degree of membership of
any given pixel in each of all possible classes. Furthermore, with soft classifiers, the result is not
a single image or classification but a set of images (one per class) that express the degree of
membership each pixel possesses in a particular class. These modules are very important in
developing robust signatures and evaluating classification techniques, and are also used for subpixel classification and mixture analysis. Several modules are available, each employing a different set membership metric to express the degree of membership of any pixel to any class. Modules for use with multispectral and hyperspectral image sets are provided.
BAYCLASS employs Bayesian probability theory to express the degree of membership of a
pixel to any class.
BELCLASS employs Dempster-Shafer theory to express the degree of membership of a pixel
to any class.
FUZCLASS employs Fuzzy Set theory to express the degree of membership of a pixel to any class.
UNMIX is used to classify remotely-sensed images using Linear Spectral Unmixing (LSU—also called Linear Mixture
Modeling). The approach assumes that a pixel value is a combination of the means of the signatures of all the classes
present in the pixel. It thus provides information about mixed pixels and is used for sub-pixel classification. This
approach is most effective when the cover types being considered are few and are distinct, e.g., for target detection.
HYPERUSP provides unsupervised classification for hyperspectral image data. The module operates using a combined
Chapter 8 Idrisi Modules
101
process of signature development using the same logic as HYPERAUTOSIG and linear spectral unmixing using the same
logic as HYPERUNMIX.
HYPEROSP provides for hyperspectral image classification through an orthogonal subspace projection approach.
HYPERUNMIX extends the capabilities of UNMIX to hyperspectral data sets.
HYPERABSORB provides for hyperspectral image classification based on library spectra and continuum removal of
absorption areas and the correlation of these areas in terms of fit and depth between the library spectrum and the spectra
from an imaging data set.
MIXCALC extends the functionality of BELCLASS in the statement of degree of membership that each pixel exhibits
for each of the classes for which training data has been provided. Using the logic of Dempster-Shafer theory, a whole
hierarchy of classes can then be recognized, made up of the indistinguishable combinations of these classes. MIXCALC
can be used to deconstruct or calculate the possible mixture of any one of these non-unique classes.
Belief performs a Dempster-Shafer Weight-of-Evidence classification and extends the logic of mixture analysis, allowing
for the ability to combine new evidence with existing knowledge. The outputs of both BELCLASS and MIXCALC can
be used as inputs into Belief.
Hardeners Submenu
MAXBAY
MAXBEL
MAXFUZ
Whether Bayesian probability, Dempster-Shafer belief, or Fuzzy Set soft classification analysis has
been applied to a multispectral image set, the soft results can be reevaluated to produce a hard classification by using one of three hardeners.
Given a set of probability images, MAXBAY determines the class possessing the maximum posterior
probability for each cell. Up to four levels of abstraction can be produced. The first is the most likely
class, just described. The second outputs the class of the second highest posterior probability, and so on, up to the fourth
highest probability.
MAXBEL is essentially identical to MAXBAY, except that it is designed for use with Dempster-Shafer beliefs.
MAXFUZ is essentially identical to MAXBAY, except that it is designed for use with Fuzzy Sets.
Hyperspectral Image Analysis Submenu
HYPERSIG
HYPERAUTOSIG
HYPERSAM
HYPERMIN
HYPERUSP
HYPEROSP
HYPERUNMIX
HYPERABSORB
HYPERSIG extends the logic of signature development to the special case of hyperspectral
data. HYPERSIG creates and displays hyperspectral signatures either from training site data or
from spectral curve library files.
HYPERAUTOSIG automatically develops signatures for hyperspectral image data based on
the Linear Spectral Unmixing logic. Given a raster group file of hyperspectral images, a set of
hyperspectral signature files (.hsg) and a signature group file (.sgf) is produced. The number of
signatures to develop is set by the user and must be less than or equal to the number of bands
in the hyperspectral image set.
HYPERSAM is a spectral angle mapper hard classifier for hyperspectral data. The spectral
angle mapper algorithm is a minimum-angle procedure specifically designed for use with spectral curve library data.
HYPERMIN is a minimum-distance hyperspectral hard classifier specifically intended for use with image-based signatures developed using training sites. It employs the same logic as MINDIST using standardized distances.
HYPERUSP provides unsupervised classification for hyperspectral image data. The module operates using a combined
process of signature development using the same logic as HYPERAUTOSIG and linear spectral unmixing using the same
IDRISI Guide to GIS and Image Processing Volume 1
102
logic as HYPERUNMIX.
HYPEROSP provides for hyperspectral image classification through an orthogonal subspace projection approach.
HYPERUNMIX extends the capabilities of UNMIX
HYPERABSORB provides for hyperspectral image classification based on library spectra and continuum removal of
absorption areas and the correlation of these areas in terms of fit and depth between the library spectrum and the spectra
from an imaging data set.
Accuracy Assessment Submenu
SAMPLE
ERRMAT
Accuracy assessment is an important final step in both unsupervised and supervised classifications. Its
purpose is to quantify the likelihood that what you mapped is what you will find on the ground. This is
useful in comparing classification techniques, and determining the level of error that might be contributed by the land cover image in further analyses in which it is incorporated.
SAMPLE generates a random set of locations to be used for ground verification and the assessment of the level of error
associated with the classification process result. SAMPLE produces a vector file of points according to a systematic, random or stratified random sampling scheme. It is useful in many other accuracy assessment studies as well as in providing
a general technique for sampling over space.
ERRMAT produces an error matrix analysis of categorical map data compared to ground truth information. It tabulates
errors of omission and commission, marginal and total errors, and selected confidence intervals. Per-category Kappa
Index of Agreement figures are also provided. This information is used to assess the accuracy of the classification procedure undertaken and may suggest ways to improve the classification.
The Reformat Menu
CONVERT
PROJECT
RESAMPLE
WINDOW
EXPAND
CONTRACT
CONCAT
TRANSPOSE
Items in the Reformat Menu allow you to change the data and file type of a file,
reorient an image or vector file, change the extent of the study area, change resolution, generalize the level of detail in the file, join files together, and convert files
from raster to vector and vice versa.
CONVERT may be used to change the data type or file type of an image or vector file. Valid data types are byte, integer, and real, while valid file types are ASCII,
binary, and packed binary.
PROJECT is used to change the reference system of image or vector files. Supported projections in this version are: Albers Equal Area Conic, Hammer Aitoff,
Raster/Vector Conversion >
Lambert Azimuthal Equal Area (North and South Polar Aspects, Transverse and
PNTGEN
Oblique), Lambert Conformal Conic, Mercator, Plate Carree, Stereographic
(North and South Polar Aspects, Transverse and Oblique), Transverse Mercator,
LINEGEN
and Gauss-Kruger. By using reference system parameter files, a limitless number
LINTOPNT
of reference system conversions can be undertaken. PROJECT takes a file currently in one of the supported systems, back projects to spherical or geodetic coordinates, then undertakes the forward
projection to the desired referencing system. For a detailed discussion see the chapter on Georeferencing in this volume.
RESAMPLE is used to georegister an image or vector file to a reference system or to another file. It takes the coordinates of a set of control points in an existing file, and in the desired new reference system, and converts the file to the new
reference system by means of either a linear, quadratic or cubic polynomial mapping function. In the case of raster
images, cells in the new grid rarely match up in any simple way with the original grid. Therefore new cell values are esti-
Chapter 8 Idrisi Modules
103
mated by resampling the old grid (hence the name of the module). When an image or vector file is already georeferenced
to a known reference system but must be transformed to another reference system, the PROJECT module in the Reformat menu is more appropriate. For a detailed discussion see the chapter on Georeferencing in this volume.
WINDOW extracts a rectangular sub-area of a larger image to create a new smaller image. The same area, defined by
either row and column positions or reference system coordinates, may be extracted from multiple images in one operation. A similar function which allows the currently-displayed window to be saved to a new image is available through the
Save Map Composition dialog box of COMPOSER in the Display System.
EXPAND and CONTRACT may be used to alter the resolution of raster images (by increasing or decreasing the frequency of cells) as long as the change is by some integer multiple or division (e.g., by 3 times, but not by 3.2 times—for
that you need RESAMPLE). EXPAND works by pixel duplication while CONTRACT offers the options of contraction
by pixel thinning or by pixel aggregation. With thinning, pixel values are dropped out, while with aggregation, the resulting
value is the mean of the original values. CONTRACT may thus be used to generalize the information in an image.
CONCAT joins together multiple images or multiple vector files into a single image or vector file. Images may overlap
when concatenated, and you may choose an opaque process or a transparent process. All files must be in the same reference system. For vector files, the object type must be the same for all files.
TRANSPOSE can be used if a rotation is required. TRANSPOSE can rotate an image by 90 degrees in either direction.
It can also reverse the order of rows or columns—a process known as reflection.
PNTGEN is used to reduce the number of points in a vector point file. A radial tolerance is used. Such thinning may be
desired for display purposes or to limit the processing time required for an analysis.
LINEGEN generalizes vector line data using any of three techniques—point selection, low-pass filtering or tolerance
band generalization. The first of these is included only for pedagogic purposes—it rarely produces a good generalization.
Low-pass filtering is excellent for smoothing of lines but does not offer data set reduction. Tolerance band generalization,
however, offers excellent context-sensitive data reduction although the result is not so smooth.
LINTOPNT creates a vector point file from the vertices of a vector line file. It is useful in assessing the density of vertices in a line file, which is commonly a part of TIN surface generation (see the Data Entry menu).
The Raster / Vector Conversion Submenu
POINTRAS
LINERAS
POLYRAS
POINTVEC
LINEVEC
POLYVEC
It is often necessary to convert data between raster and vector formats. Data must often be generated
through digitizing, which produces vector files. If the subsequent analysis is to be done in raster format, the data must be converted. There may be a need to convert from raster to vector as well, in
order to transfer data between systems, or to produce vector layers to display on top of other raster
layers.
The first three modules of this submenu convert vector data to raster data. In all three cases, an existing image, such as one created with INITIAL, is updated with the vector information. Normally, the
image to be updated is a blank image, but this is not required. Any image may be updated, and the
original values will be altered only where they are covered by a vector feature.
POINTRAS converts point vector data to raster format by assigning a new value to all cells in which a vector point is
located. This new value may be the identifier of the vector point, a Boolean one or zero to simply indicate the presence or
absence of one or more points, or an identifier equal to the number of points falling in the pixel.
LINERAS converts line vector data to raster format. All pixels intersected by a vector line are assigned the identifier of
the vector line.
POLYRAS converts polygon vector data to raster format by assigning all pixels covered by a vector polygon the identifier of the polygon.
IDRISI Guide to GIS and Image Processing Volume 1
104
The next three modules convert raster data to vector.
POINTVEC converts all non-zero raster pixels into vector point features, where the vector point is assigned an identifier equal to the value of the original pixel. The points are given coordinates corresponding to the center of the pixel.
LINEVEC traces non-zero raster pixels to create a vector file of line features.
POLYVEC is used to convert areas of pixels into vector polygons. The boundaries of the resulting vector polygons have
a "stair-step" character, because they are created by tracing around the edges of the raster cells. This effect can be mitigated with the use of LINEGEN, choosing a three point low pass filter and the default weighting scheme.
The Data Entry Menu
CartaLinx
Edit
ASSIGN
INITIAL
UPDATE
UTMRef
Surface Interpolation >
Database Workshop
Collection Editor
Data entry is often a very time consuming element when working with GIS. In addition
to the data entry modules in this menu, you will find conversion utilities for existing data
that are in non-IDRISI formats in the File/Import submenu. The chapter on Database
Development in this volume also discusses issues of data entry.
CartaLinx is a spatial database development tool also developed and distributed by Clark
Labs. It provides tablet as well as on-screen digitizing capabilities and a wide range of
data editing tools. If CartaLinx is installed, choosing this menu item will launch
CartaLinx. If it is not installed, choosing this menu item will display information about
CartaLinx.
Edit is a very basic text editor that may be used to create or alter any ASCII file. Its primary function in data entry is in the creation of attribute values files for use with the
ASSIGN module. Edit may also be used to create a variety of other IDRISI files, such as correspondence files and group
files.
ASSIGN is used to change the attributes of a raster image. It does this by using an attribute values file, in which the first
column contains the original value and the second column contains the new values to be assigned. The equivalent of
ASSIGN may also be accessed from Database Workshop.
INITIAL is used to create an image containing a single value. While there are some analytical operations where a single
value image might be needed, more commonly it will be used to set up a blank image initialized with zeros for the rasterization modules (of the Reformat Menu) to update. The parameters of the new file, such as rows and columns, reference
coordinates and so forth, may be entered interactively, or may be copied from a specified existing image.
UPDATE assigns single values to specific cells or rectangular groups of cells.
UTMRef is used to facilitate the creation of reference system parameter files based on the Universal Transverse Mercator
system, for subsequent use with PROJECT.
The options in the Surface Interpolation submenu are identical to those of the Analysis / Surface Analysis / Interpolation submenu and are described in that section above. The issues surrounding surface interpolation as well as the options
available in IDRISI are discussed in greater detail in the chapter Surface Interpolation in the IDRISI Guide to GIS and
Image Processing Volume 2.
Database Workshop provides full access to dBase, Access, and FoxPro database files. Edit mode may be activated so
the user may update or enter database values directly.
The Collection Editor is used to define members of a raster group file collection or a linked-table collection.
Chapter 8 Idrisi Modules
105
Window List
The Window List menu item provides a listing of all open windows. Open dialogs are listed with the module name (e.g.,
Composer) and open map windows are listed with the filename that appears in the banner of the map display window.
Clicking on a window name in the list will bring that window into focus.
The Help Menu
Contents
Using Help
ArcView / ArcInfo Quick Start >
The Help menu gives you access to the IDRISI on-line Help System.
Contents leads you directly to the IDRISI Help System. Here you will find
Contents, Index and Search tabs.
Using Help describes the IDRISI Help System. Here you will find out how to
access and navigate through the Help System. Also provided are general information on the Help System screen and functions as well as how a typical program module's entry is organized within the
Help System.
About Idrisi32
ArcView/ArcInfo Quick Start Submenu
Help Using ArcView with IDRISI
SHAPEIDR (import/Export Shape Files)
ARCRASTER (Import/Export GRID Files)
ARCIDRIS (Import/Export ArcInfo GEN Files)
TIFIDRIS (Import/Export ArcView TIF Image to/from IDRISI)
JPGIDRIS (Import/Export ArcView JPG Image to/from IDRISI)
The ArcView/ArcInfo Quick Start section
of the help menu gives immediate access to
information about Using ArcView with
IDRISI and to the Help System for modules
commonly used to transfer data between the
two systems.
POINTRAS (Point Vector to Raster)
LINERAS (Line Vector to Raster)
POLYRAS (Polygon Vector to Raster)
POINTVEC (Raster to Point Vector)
LINEVEC (Raster to Line Vector)
POLYVEC (Raster to Polygon Vector)
About Idrisi32 provides licensing and copyright information as well as general information, including contact addresses
and telephone numbers.
IDRISI Guide to GIS and Image Processing Volume 1
106
Idrisi Modeling Tools
One of the most fundamental roles for GIS is in the development, testing and utilization of models—suitability models,
soil erosion models, urban growth models, and the like. IDRISI provides an extensive set of tools for modeling, accommodating a range of levels of expertise. The most fundamental is Macro Modeler, a graphical modeling environment that
combines the strengths of extensive capabilities and ease of use. For simpler equation-based modeling using GIS layers,
Image Calculator provides rapid equation entry using a familiar calculator interface. A third facility is a macro scripting
(.iml) language that is provided largely for legacy applications (this was the original form of modeling tool provided in an
early release of IDRISI). Finally, there is the IDRISI32 API, an industry standard COM interface that provides access to
the internals of the IDRISI system for the most demanding applications and interface development. The API requires a
COM-compliant programming environment such as Visual C++, Delphi, or Visual Basic.
Idrisi Macro Modeler
The Idrisi Macro Modeler is a graphic environment in which you may assemble and run multi-step analyses. Input files,
such as raster images, vector layers, and attribute values files, are linked with Idrisi modules that in turn link to output data
files. The result is a graphic model, much like the cartographic models described in the Idrisi Tutorial Introductory GIS
Exercises section. A model may be as simple or complex as desired. Figure 9-1 shows a simple suitability mapping model.
Figure 9-1
Model Construction
To work with Macro Modeler, either click on its icon on the tool bar, or select it from the Modeling Menu. This yields a
special workspace in the form of a graphic page along with a separate menu and tool bar. Constructing a model involves
placing symbols for data files and modules on the graphic page, then linking these model elements with connectors. To
place an element, click on its icon (shown in Figure 9-2) or choose the element type from the Macro Modeler menu, then
choose the specific file or operation from the supplied pick list. (Note that not all Idrisi modules are available for use in
the Macro Modeler.48) To connect elements, click on the connector icon, then click on one of the elements and drag the
cursor onto the other element and release. It is necessary to connect elements in the proper order because the Macro
Modeler assumes that the process flows from the element you first clicked to the element upon which you released the cur-
48. The modules of the image processing menu as well as any module that does not create an output file, such as REGRESS, do not work in the Macro
Modeler.
Chapter 9 Idrisi Modeling Tools
107
sor. To delete any element, first click on that element, then click the delete icon or press the delete key on the keyboard..
Data Elements
Raster
Layer
Collection
Vector
Layer
Command Elements
Attribute
Values File
Module
Submodel
Connector
Delete
Selected
Element
Figure 9-2
Whenever a module is placed in the model, it is placed with the appropriate output data element for that operation already
linked. Default temporary filenames are assigned to output files. Right click on a data file symbol to change the filename.
Long filenames are left-justified. To see the entire filename, pause the cursor over the element graphic to cause a balloon
to appear with the entire element name. Filenames are stored without paths. Input files may exist in any folder of the current project (specified with the Data Paths module) while intermediate and final output files are always written to the
Working Folder.
To specify the parameters for the module, right click on the module symbol. This opens the module parameters dialog.
Click on any parameter to see and choose other options. Access the Help System for detailed information about the various options by clicking the Help button on the parameters dialog.
Submodels are user-constructed models that are subsequently encapsulated into a single command element. When a
model is saved as a submodel, a parameters dialog is created for the submodel in which the user provides captions for all
the model inputs and outputs. A submodel consists of a submodel parameter file (.ims) and its macro model (.imm) file.
When a submodel is placed as a command element in a model, the Macro Modeler knows how many and what types of
input files are required to run the submodel and what types of output files should be produced. For example, you might
create a submodel that stretches a single image then exports it to a JPG image. The model would include both STRETCH
and JPGIDRIS, but the submodel would only require the input image and the output image. The user may change the
input and output data elements of a submodel through the parameters dialog, but settings for module commands that are
part of a submodel must be changed by opening the model from which the submodel was created, making the changes,
then resaving the submodel. Submodels not only simplify multi-step processes, but are important elements when models
have sections that should loop. This is discussed in further detail below.
Models are saved to an Idrisi Macro Model file (.imm). This file preserves all aspects of the model, including the graphic
layout. The file may be opened and modified with the Macro Modeler. The graphic description of the model may be copied (as a .bmp file) to the operating system clipboard then pasted into other word processing and graphics software. The
graphic may also be printed. The toolbar icons for Macro Modeler file management are shown in Figure 9-3.
New
Open
Save
Print
Copy to
Clipboard
Figure 9-3
Models may be run at any stage of completion. The Macro Modeler first checks to see if any of the output images already
exist. If so, it shows a message indicating which file exists and asks if you wish to continue. Answering Yes (or Yes to All)
will cause the existing output file to be deleted before the model runs. Answering No will prevent the Macro Modeler
from running the model. As the model runs, the module element that is currently being processed turns green. To stop a
model at any point while it is running, click the stop icon. The model will finish the process it is currently doing and will
then stop.
IDRISI Guide to GIS and Image Processing Volume 1
108
The last output created by the model will be automatically displayed. Intermediate images may be displayed by clicking on
the display icon on the Macro Modeler toolbar, then on the data layer symbol. The metadata may be accessed for any data
layer by clicking the metadata icon on the Macro Modeler toolbar, then the data layer symbol. Toolbar icons for these
functions are shown in Figure 9-4.
Run
Stop
Display
Metadata
Figure 9-4
DynaGroups
Two distinctive capabilities of the Macro Modeler are its use of DynaGroups to facilitate running the same model on multiple data layers and DynaLinks to construct iterative models in which the output of a process becomes an input for the
next iteration of the process.
A DynaGroup is a raster group file or time series file that is used in the Macro Modeler so each member of the group is
used to produce an output. Contrast this with the use of a group file as a regular data input to the Macro Modeler in which
a group file used as input produces a single output as with the module COUNT. More than one DynaGroup may be used
in a process. The module OVERLAY, for example, requires two raster images as input and produces a single raster image
output. If DynaGroups are used as the two inputs, then Macro Modeler will first verify that the same number of members
is in each group, then it will run OVERLAY using corresponding images from the two group files. If a single raster layer
is used as one input and a DynaGroup is used as the other input to OVERLAY, then the modeler will run OVERLAY
with the single raster image and each group member in turn. In all cases the number of output images is equal to the number of members in the DynaGroup file. Several options for naming the output files are described in the Help System. In
addition, a group file of the output files is automatically constructed. In the model shown in Figure 9-5, two Dynagroups
are used as input to an overlay operation. The output will be the number of raster images contained in the input DynaGroups and a raster group file.
Figure 9-5
DynaLinks
DynaLinks are used to run a model iteratively. An output data element is joined to an initial input data element with a
DynaLink. This indicates that when the model performs the second iteration, the linked output filename should be substituted in place of the linked input file name. A dynalink may only be connected to an initial data layer, i.e., one which has
no connector joining to its left-hand side. Macro Modeler asks how many iterations are to be performed and whether each
terminal output of the model or only the terminal output of the final iteration should be displayed. The final output of a
model that includes a DynaLink is a single raster layer. The model shown in figure 9-6 uses a DynaLink. When the model
Chapter 9 Idrisi Modeling Tools
109
is run, the Macro Modeler asks for the desired number of iterations. Let us assume that in this case we want 5 iterations.
The input file called Band3 is filtered to create an output image that is called filtered x5_01. On the second iteration, the
file filtered x5_01 is filtered to produce the file filtered x5_02. This file is then substituted back as the input for the next
filter operation and so forth. At the end of the 5th iteration, the final image is named filtered x5. Intermediate images (e.g.,
filtered x5_03) are not automatically deleted until the model is run again.
Figure 9-6
Note that the entire model is run during each iteration, even if the substitution occurs near the end of the model. To run
just a portion of a model iteratively, use a submodel that contains a DynaLink (see below).
Using DynaGroups, DynaLinks and Submodels Together
When DynaGroups and DynaLinks are used in the same model, the number of members in the DynaGroup defines the
number of iterations to be run. Each iteration of the model feeds in a new DynaGroup member. The output of a model
that contains both elements is a single data layer. Figure 9-7 shows a simple model in which a group of raster files is
summed through the use of a DynaGroup and a DynaLink. In the first iteration, the initial layer BLANK (which is simply
an image with value 0 everywhere) is used with the first DynaGroup member in an Overlay Addition operation. This produces the temporary file SUM_01. SUM_01 is then substituted back into the place of BLANK for the second iteration.
The number of iterations equals the number of images in the DynaGroup.
Figure 9-7
To iterate or loop only a portion of the model, create a submodel for the portion that should loop. When you run the
model, a dialog box for each submodel will appear asking for the number of iterations for that submodel.
Image Calculator
Image Calculator is a calculator tool for quickly evaluating equations using raster image layers. Equations can be saved and
edited at a later time. It is extremely easy to use since it works exactly like a scientific calculator. The only special requirement is that all image files must be specified using square brackets as delimiters. For example, in the example shown in
Figure 9-8, a calibrated image band is created by multiplying its values by a gain of 0.9421 and then adding an offset of
0.0168. Image Calculator offers the functionality typically associated with calculators. It can be access either through its
icon on the tool bar, or from its menu entry under the Mathematical Operators section of the GIS Analysis menu.
IDRISI Guide to GIS and Image Processing Volume 1
110
Figure 9-8
Command Line Macros
Macro Modeler and Image Calculator provide very direct and simple interfaces for creating models. However, command
line macro scripts are also supported, primarily to support legacy applications from early versions of IDRISI. A command
line macro is an ASCII file containing the module names and parameters for the sequence of commands to be performed.
It provides no automatic batch capability (though you may quickly copy, paste and edit to achieve this) nor looping.
Specific instructions on the macro command format for each module can be found in the on-line Help System. These
instructions can then be placed into an ASCII file with an ".IML" (an acronym for "Idrisi Macro Language") extension.
Typically, this will be done with the Edit module. Choosing to save as type Macro file will automatically add the correct
file name extension.
Note that some modules do not have a macro command version. These are typically modules that do not produce a
resulting file (e.g., IDRISI File Explorer) or modules that require interaction from the user (e.g., Edit). In the menu, any
module written in all upper-case letters may be used in a macro.
IDRISI records all the commands you execute in a text file located in the working folder. This file is called a LOG file.
The commands are recorded in a similar format to the macro command format. It may sometimes be more efficient to
edit a LOG file to have the macro format rather than typing in macro commands from scratch. To do so, open the LOG
file in Edit and alter it to have the macro file format. Save it as a macro file.
Note also that whether from a dialog box or a macro, the command line used to generate each output image is recorded
in that image's Lineage field in its documentation file. This may be viewed with the Metadata utility and may be copied
and pasted into a macro file using the CTL+C keyboard sequence to copy highlighted text in Metadata and the CTL+V
keyboard sequence to paste it into the macro file in Edit.
The Run Macro dialog box, under the File menu, will ask for the name of the macro to execute along with any command
Chapter 9 Idrisi Modeling Tools
111
line parameters that should be passed to the macro file itself. This latter box can be left blank if there are none.
Each line in a macro is completed in sequence because the resulting image of one command is often used as input to a
later command. In addition, if more than one macro is launched, the first macro launched will be completed before the
second is begun.
Macro File Structure
IDRISI Macro files support the following syntax.
1. IDRISI modules in command line mode. The following is an example of a valid line in an IML file:
OVERLAY x 3*soilsuit*slopsuit*suitland
The x after the module name indicates that command line mode is invoked. All parameters after the x are separated by asterisks.
Short file names may be given, as in the example above. In this case, the macro will look for the appropriate file type first
in the working folder, then in each resource folder in the order in which they are listed in the current project file. Long file
names, as well as full paths, may also be given in the command line, e.g.:
OVERLAY x 3*c:\data\soil.rst*c:\data\slopes.rst*c:\output\suitable_land.rst
2. The macro processor supports command line variables using the %# format (familiar to DOS users). For example, the
above line could be modified as such:
OVERLAY x 3*%1*%2*%3
In this modification, %1, %2 and %3 are replaced with the first three command line parameters specified in the Macro
Parameters text box of the Macro dialog box.
3. External Applications. Any non-IDRISI application can be called from an IML file. The syntax is as follows:
CALL [Application Name] [Command Line Parameters]
where [Application Name] should specify the full path to the application. For example, to call an application named SOILEROD.EXE in a folder named MODELS on drive C, and pass it the name of an input file named SOILTYPE.RST, the
command would be:
CALL c:\models\soilerod.exe soiltype.rst
4. Other IML files. It is possible to execute other IML files from within a main IML file by using the BRANCH command. The syntax for such an operation is:
BRANCH [IML File Name] [Command Line Parameters]
The IML file name is the short name of the IML file to call (i.e., without extension). Command line parameters are
optional.
The IDRISI32 API
IDRISI32 has been designed as an OLE Automation Server using COM Object technology (i.e., a COM Server). As a
consequence, it is possible to use high-level development languages, such as Delphi, Visual C++, Visual Basic, or Visual
Basic for Applications (VBA) as macro languages for controlling the operation of IDRISI32. In addition, you can create
sophisticated OLE Automation Controller applications that have complete control over the operation of IDRISI32. Thus
you would use the API in instances where you wish to:
IDRISI Guide to GIS and Image Processing Volume 1
112
—create complex models that require the extensive control structures of a high level computer language (such as
many dynamic model);
—you wish to create custom applications, perhaps to be distributed to others;
—you wish to create custom interfaces.
The OLE Automation Server feature of IDRISI32 is automatically registered with Windows when IDRISI32 is first run
on your system. Thus, if you have installed IDRISI32 and run it at least once, you automatically have access to the full
API. The IDRISI32 OLE Automation server provides a wide range of functions for controlling IDRISI, including running modules, displaying files, and manipulating IDRISI environment variables. With visual programming environments
such as Delphi, Visual C++ and Visual Basic, access to these functions is very easy. Specific instructions as well as a complete reference manual can be accessed from the Modeling menu.
While the API provides a wide range of tools, most users only require a few - typically the RunModule operation that can
run any module, the Display File operator (for construction automatically displayed outputs), and a couple of procedures
for accessing the project folder structure.
Instructions are also provided in the API help file (under the Modeling menu) for how to change the scripts for the
IDRISI menu system so that you can develop applications and incorporate them as direct extensions to IDRISI itself.
Chapter 9 Idrisi Modeling Tools
113
IDRISI Guide to GIS and Image Processing Volume 1
114
Database Development
Introduction
As illustrated in the Introduction to GIS chapter in this volume, the database is at the center of the Geographic Information System. In fact, it can be seen as providing the fuel for the GIS. The tasks of finding, creating, assembling and integrating these data may collectively be termed database development. While it is seldom the most interesting part, database
development can easily consume 80 to 90 percent of the time and resources allocated to any project. This chapter discusses many of the issues encountered in database development and also presents a summary of the database development tools included in IDRISI.
Collecting Data
The first stage in database development is typically the identification of the data layers necessary for the project. While it
is tempting to acquire every available data layer that coincides with the study area, it is usually more efficient to identify
what is needed prior to assessing what is available. This helps to avoid the problems of data-driven project definitions and
encourages a question-driven approach.
In addition to identifying the themes that are necessary (e.g., elevation, landcover), it is also important to determine what
resolution (precision) and what level of accuracy are required. These issues will be discussed more thoroughly below.
Once the necessary data requirements for the project have been specified, the search for data can begin. There are five
main ways to get data into the database:
1. Find data in digital format and import it.
2. Find data in hard-copy format and digitize it.
3. Collect data yourself in the field, then enter it.
4. Substitute an existing data layer as a surrogate.
5. Derive new data from existing data.
Find Data In Digital Format and Import It
Where To Find Data
In many countries, governmental organizations provide data as a service. In the United States, these data are often free for
electronic download or are available for a small fee. The United States Geological Survey, the National Oceanic and
Atmospheric Administration and the Census Bureau are good examples of governmental agencies in the United States
that have a mandate to create and provide digital data. All these agencies have sites on the World Wide Web that provide
information about the availability of digital data. In addition, it is becoming more and more common for states (or provinces) to have a particular agency or department that is responsible for creating, maintaining and distributing GIS data layers. In Massachusetts, for example, the state office MASSGIS has this mandate. For many, the World Wide Web is fast
becoming the preferred method for finding and acquiring government-provided data.
Commercial companies also provide digital data. These data might be generic sets of commonly-used base layers, such as
transportation networks, or they might be customized to the needs of specific users. Commercial companies often begin
Chapter 10 Database Development
115
with free government data and "add value" by converting it to a specific software format or updating it.
With the proliferation of GIS and image processing technologies, many non-governmental and academic institutions may
also possess data layers that would be useful to your project. While these organizations seldom create data specifically to
share with others, they will often make available what they have collected. Therefore, it is usually worthwhile to identify
other researchers or institutions working in the same study area to inquire about their data holdings. Electronic discussion
forums, such as IDRISI-L, also provide a venue to request particular data layers. The Clark Labs Web Site includes a page
of links to GIS and image processing related sites, many of which provide data.
Once you have located electronic data useful to your project, you need to import it into the software system you are using.
Sound simple? Ideally, these format conversions are trivial. In reality, it is not uncommon to find that getting from what
you have to what you want is not straightforward.
Physical Media
Data might come on a CD-ROM, a 9-track tape, a Zip disk or any number of less common media. When acquiring data,
be sure to ask about the medium on which it will arrive and make sure you have the ability to read that medium. The ability to download large data files electronically has lessened this problem substantially, but even with network download,
access to the network and appropriate software, such as an FTP client, are necessary.
Data Formats
You must also determine the file format of the data. In general terms, you should know which category it falls within. For
example, a data layer of superfund sites could be stored as a raster, vector, database or spreadsheet data file. More specifically, you will need to know if it is in a particular software format, an agency format, or some sort of data interchange format. IDRISI provides for direct import of a number of standard raster, vector and attribute data formats.
If the particular file format you have is not directly supported by IDRISI import routines, then you may still be able to use
it. Raster data tends to have a simpler structure than vector data and it is often the case that a raster file can be imported
into IDRISI using low-level tools, such as PARE or BILIDRIS, to re-order the pixels. Vector data are more complex, and
unless the structure is extremely simplistic, it is unlikely that you will be able to import vector data using low-level tools.
For both raster and vector data, translation software can provide the bridge between the current data format and IDRISI.
Translation software for GIS file formats vary in the number of formats they support and in price. Your goal in using a
translation software is to turn the file you have into something that you can import with IDRISI.
You must also consider data compression when acquiring data. Compression allows files to be stored using less memory
than they would normally require. There are two types of compression of which to be aware. The first is compression that
is used by the GIS software itself. IDRISI, for example, has a run-length encoded packed binary file type. While IDRISI
can use these files analytically, other GIS software will not be able to use them. It is best to avoid sharing files that are
compressed with software-specific routines between different GIS software systems.
The second type of compression, sometimes referred to as "zipping", is used for files that are not currently being used.
This compression is used to make the transfer and storage of files easier. The files must be uncompressed or "unzipped"
prior to use.
Always inquire about the compression software used. If you don't have the same software, get the file in an uncompressed format, or as a self-extracting executable file (a compressed file that will uncompress itself without the compression software installed on your computer).
Information about importing specific file formats is available in the IDRISI on-line Help System.
Find Data In Hard Copy Format and Digitize It
If the data you need is in a hard-copy format, you will need to digitize it (i.e., make it digital) to bring it into your GIS data-
IDRISI Guide to GIS and Image Processing Volume 1
116
base. Some hard-copy data might be digitized by simply typing it into an ASCII editor or a database table. For example,
you might have tabular field notes from sample survey sites that can be typed directly into Database Workshop.
More commonly, hard-copy data is in the form of a map, an ortho-photo, or an aerial photograph. If you want to extract
particular features from a map, such as elevation contours, well-head locations or park boundaries, then you will digitize
these as vector features using a digitizing tablet. (Alternatively, features plainly visible on a digital ortho-photo can be captured with on-screen digitizing, which is discussed below.)
A digitizing tablet (or digitizing board) contains a fine mesh of wires that define a cartesian coordinate system for the
board. Most boards have 1000 wires per inch, which results in a maximum resolution of 1/1000 inch. The user attaches
the hard copy map to the digitizing board, then traces features on the map with a digitizing puck (a mouse-like device) or
stylus (a pen-like device). The digitizing tablet senses the x,y positions of the puck as the features are traced and communicates these to the digitizing software.
Digitizing software packages vary tremendously in ease of use and capability. CartaLinx, also a product of the Clark Labs,
combines an easy user interface with flexible digitizing and post-digitizing editing, and is recommended for use with
IDRISI.
Most digitizing software will allow the user to "register" the map on the digitizing tablet. This process establishes the relationship between the tablet's cartesian coordinates and the coordinate system of the paper map. The software compares
the tablet coordinates and the map coordinates for a set of control points and then derives a best-fit translation function.
This function is then applied to all the coordinates sent from the board to the software. If your digitizing software does
not provide this translation, the RESAMPLE module in IDRISI may be used to transform the tablet coordinates to map
coordinates after digitizing.
In addition to digitizing using a digitizing tablet, scanners can be used to digitize hard copy images such as maps or aerial
photos. Unlike digitizing tablets that produce vector data, scanners produce raster data. Also, scanners do not capture distinct features but measure the relative reflectance of light across a document according to a user-specified resolution, normally specified as dots per inch. Each dot becomes a pixel in the resulting image. Sensors detect the reflection of red,
green, and blue (or gray level) and record these as digital values for each pixel. The scanned image is then imported to the
GIS software as a raster image. Scanned images are normally imported to IDRISI through either the .BMP or .TIF file
formats.
Once scanned, features such as roads may be extracted into a vector file format using specialized software. However, this
process is often too costly or inaccurate, and it requires very clean hard copy sources for scanning. Another method to
extract vector features from scanned raster images is with on-screen digitizing.
With on-screen digitizing, sometimes referred to as "heads-up" digitizing, the scanned source map or photo is displayed
on screen and features are digitized using a standard mouse. The RESAMPLE module may be used to georegister the
image before digitizing. Both IDRISI and CartaLinx allow you to capture features as vector format files from a displayed
raster image through on-screen digitizing.
Collect Data Yourself In The Field Then Enter It
For many research projects, it is necessary to go into the field and collect data. When collecting data in the field for use in
a GIS, it is imperative to know the location of each data point collected. Depending upon the nature of the project and
level of accuracy required, paper maps may be used in conjunction with physical landmarks (e.g., roads and buildings) to
determine locations. However, for many projects, traditional surveying instruments or Global Positioning System (GPS)
devices are necessary to accurately locate data points.
Locational coordinates from traditional surveys are typically processed in Coordinate Geometry (COGO) software and
may then be transformed into a GIS-compatible vector file type. CartaLinx is able to perform this transformation.
For many GIS applications, however, GPS devices may provide a less expensive alternative. The Global Positioning System is composed of 24 US Department of Defense satellites orbiting the Earth at approximately 20,000 kilometers. Each
Chapter 10 Database Development
117
satellite continuously transmits a time and location signal. A GPS receiver processes the satellites' signals and calculates its
position.49 The level of error in the position depends on the quality (and price) of the receiver, atmospheric conditions
and other variables. Some units provide display of locations only, while others record locational information electronically for later download to a computer.
IDRISI includes a GPS link. When this is active, the position recorded by the GPS receiver is displayed on the active display window and is updated as the position changes. The geodetic coordinates produced by the receiver are automatically
projected to the reference system of the display. This GPS link is intended primarily to display and save waypoints and
routes. CartaLinx also includes a GPS link and allows for more flexibility in creating data layers from the GPS data.
Most GPS data processing software supports several GIS export formats. Shape files and DXF files are commonly used
as a bridge to IDRISI. If those formats are not available or if the data set is small enough to be entered by hand, creation
of an IDRISI vector file is easily accomplished through the use of a text editor like the EDIT module in IDRISI.
Additionally, GPS software typically exports to several types of ASCII text files. The XYZIDRIS module in IDRISI
imports one of the more common types. The input to XYZIDRIS must be a space- or comma-delimited ASCII file in
which numeric X, Y, and Z values (where Z = elevation or a numeric identifier) are separated by one or more spaces and
each line is terminated by a CR/LF pair. XYZIDRIS produces an IDRISI vector point file.
Substitute An Existing Data Layer As a Surrogate
At times, there is simply no way to find or create a particular data layer. In these cases, it may be possible to substitute
existing data as a surrogate. For example, suppose an analysis requires powerline location information, but a powerlines
data file is not available and you don't have time or funds to collect the data in the field. You know, however, that in your
study area, powerlines generally follow paved roads. For the purposes of your analysis, if the potential level of error introduced is acceptable, you may use a paved roads layer as a surrogate for powerlines.
Derive New Data From Existing Data
New data layers may also be derived from existing data. This is referred to as derivative mapping and is the primary way in
which a GIS database grows. With derivative mapping, some knowledge of relationships is combined with existing data
layers to create new data layers. For example, if an image of slopes is needed, but none exists, you could derive the slope
image (using the IDRISI SURFACE module) from a digital elevation model, if available. Similarly, if you need an image
of the relative amount of green biomass on the ground, you might derive such an image from the red and infrared bands
of satellite imagery.
Another common form of derivative mapping is the interpolation of a raster surface (e.g. an elevation model or temperature surface) from a set of discrete points or iso-lines. See the chapter Surface Interpolation in the IDRISI Guide to
GIS and Image Processing Volume 2 for more information about this form of derivative mapping.
Considerations
Resolution
In all database development tasks, no matter the source of data, there are several issues to consider. The first consideration is the resolution of the data. How much detail do you need? Resolution affects storage and processing time if too
fine and limits the questions you can ask of the data if too coarse.
Resolution may refer to the spatial, temporal or attribute components of the database. Spatial resolution refers to the size
of the pixel in raster data or the scale of the map that was digitized to produce vector data. Temporal resolution refers to
the currency of the data and whether the images in the time series are frequent enough and spaced appropriately.
49. Several GPS companies (e.g., Trimble, Magellan) have excellent web sites explaining how positions are calculated from the satellite signals.
IDRISI Guide to GIS and Image Processing Volume 1
118
Attribute resolution refers to the level of detail captured by the data values. This might be exemplified by the difference
between a landcover map with a single forest class, a landcover map with hardwood and softwood forest classes, and a
landcover map with many different forest classes.
Accuracy
Accuracy is the second consideration. While accuracy does have a relationship to resolution, it is also a function of the
methods and care with which the data were collected and digitized. Unfortunately, it is not always easy to evaluate the
accuracy of data layers. However, most government mapping agencies do have mapping standards that are available. The
IDRISI metadata (documentation) file structure includes fields for accuracy information, but many other file formats
carry no accuracy information. In addition, even when such information is reported, the intelligent use of accuracy information in GIS analysis is still an avenue of active research. The capabilities of IDRISI in this area are discussed in the
chapter on Decision Support in the IDRISI Guide to GIS and Image Processing Volume 2.
Georeferencing
The third consideration is georeferencing. Data layers from various sources will often be georeferenced using different
reference systems.50 For display and GIS analysis, all data layers that are to be used together must use the same reference
system. Providing that the details of the reference systems are known, they can usually be converted with the IDRISI
module PROJECT.
However, it may be possible to download graphic files from the Web, for example, that are not georeferenced. Also, while
IDRISI supports many projections, you may find data that are in an unsupported projection. In both cases, you will not
be able to use PROJECT. It is sometimes possible to georeference an unreferenced file or a file with an unsupported projection through resampling (the IDRISI module RESAMPLE) if points of known locations can be found on the unreferenced image. All users are encouraged to read the chapter on Georeferencing in this volume for more information about
this key issue. Also, see the section on Data Integration below.
Cost
Finally, the fourth consideration in database development is cost. This must be assessed in terms of both time and money.
Another consideration is whether the data will be used once or many times, the accuracy level necessary for the particular
(and future) uses of the data, and how often it must be updated to remain useful. Don't underestimate the amount of
labor required to develop a good database. As stated in the introduction, database development quite commonly consumes a large portion of the labor allocated to a GIS project.
General Import Tips
IDRISI requires files to be in Intel format. This typically is only an issue when integer data are acquired from Macintosh,
UNIX, workstation and mainframe platforms which use the Motorola format. The byte order is different between the
two formats. The FLIP module can be used to reverse the byte order of an integer file.
If you import data from certain media, the files will be written to your hard disk with a read-only attribute. This can be
removed with Windows Explorer in the Properties dialog for the file.
Tools for Import
Files are sometimes in an unspecified format. The following modules are useful for inspecting files and changing file
structures.
50. A reference system is defined by a projection, datum and grid system. See the chapter on Georeferencing in this volume for more information.
Chapter 10 Database Development
119
EDIT: Displays and edits an ASCII file. Vector data are often ASCII.
Idrisi File Explorer VIEW STRUCTURE: Displays byte-level contents of any file (in ASCII or binary format). Used to
view data files for presence and size of headers, as well as to check files for carriage returns (CR's) and line feed (LF's).
CRLF: Adds or removes carriage returns and line feeds. Used with import modules that require specific fixed record
lengths (e.g., DLG's, DEM's).
VAR2FIX: Converts variable length ASCII files to fixed length. Used in conjunction with CRLF.
METADATA: Creates and updates documentation files for data files already in IDRISI format.
Importing Satellite Imagery
IDRISI includes several special import routines for specific satellite image formats, such as LANDSAT and SPOT. If a
specific import routine isn’t available for the format of your data, the generic import tools available in IDRISI may be
used to bring the satellite imagery into IDRISI. When preparing to import satellite data using the generic tools, there are
two primary concerns: the location and contents of the header information (the information that describes the data) and
the format in which the data are stored.
The first task is to locate and read the header information. Commonly, the header will be stored as a separate file in ASCII
format or it will be distributed as paper documentation. If, as a third possibility, the header is attached to the data file, it
will be stored in byte binary format because satellite imagery is distributed in byte binary format. In this case, you can try
to read the information from the header with the VIEW STRUCTURE utility of the Idrisi File Explorer which displays
binary data as ASCII text. If you know the number of rows and columns in the image, but not the header size, there is a
way to figure this out. If the image is in byte binary format, the number of bytes in the image should equal the number of
pixels in the image. To find the number of pixels in the image, multiply the number of rows by the number of columns.
The extra bytes in the data file should be the size of the header. If the file contains 2-byte integer data, the number of
bytes in the image equals rows multiplied by columns multiplied by 2.
The contents of the header information will be presented in a variety of formats and with varying levels of detail. This
section will tell you what to look for in the header, but it cannot anticipate the format in which you will find that information.
There are three pieces of information you must find in the header (the answers to 1 and 2 will determine the import routine you will use):
1) the data file format;
2) whether a header file is attached to the data file, and its size; and
3) the number of rows and columns in the data file.
Most satellite imagery is distributed in one of two file formats: band sequential (BSQ) and band-interleaved-by-line (BIL).
Band sequential format is the same as IDRISI image format. The data values are stored as a single series of numbers and
each band of imagery is contained in a separate data file. SPOT panchromatic51 and LANDSAT TM and MSS satellite
imagery are commonly distributed in this format. With BSQ format files, the following two import routines apply:
1. If a header is attached to the data file, you will need to use the PARE module. PARE will remove the header, rename
the data file to have an .RST extension, and create a documentation file. PARE will ask for the number of rows and columns, the minimum information necessary to create the documentation file. PARE will ask for additional information as
well, such as the reference system parameters which you should be able to find in the header. When in doubt, you can try
51. The documentation for the SPOT panchromatic band will indicate that it is in BIL format, but it contains only one band, so there is no interleaving.
This is the same as BSQ format.
IDRISI Guide to GIS and Image Processing Volume 1
120
the following values: Minimum X=0, Maximum X=1 (or # of columns), Minimum Y=0, Maximum Y=1 (or # of rows),
Reference System=Plane.
2. If the header file is separate from the data file, then you merely need to rename the data file to have an .RST extension
(with the Windows Explorer rename command). Then run METADATA from the File menu to create the documentation file. Again, you must know the number of columns and rows in order to complete the documentation file, but you
should also enter any reference system information that is available from the header. If the values are unknown, try entering the values noted in the previous paragraph, with the addition of Reference Units=M and Unit Distance=1.
Band-interleaved-by-line (BIL) format stores all the bands of multispectral imagery in a single data file. The bands are
combined by storing the first row of data from the first band of imagery, followed by the first row of data from the second band and so on for every band of imagery stored in the data file. The data file then continues with the second row of
data from the first band, the second row of data from the second band and so on. SPOT multispectral imagery (SPOTXS) is commonly distributed in BIL format. For all BIL files, with or without a header, the following import routine is
used:
If a header is attached, determine its size (in bytes) and then use BILIDRIS. BILIDRIS will pull apart and create individual IDRISI format image and documentation files for each band of imagery.
Note: If the header size is incorrectly specified during PARE or BILIDRIS, the image will not be correctly imported.
Because too much or too little is taken out of the file as a header, the actual data left behind is either incomplete or contains extra information. This can lead to an error message informing you that the image is incorrectly sized, or the image
will display improperly (e.g., data from the right side may appear moved to the left side). You will need to determine the
correct header size and try again.
If an image seems to successfully import, but the display of the image is staggered or skewed to one side, it is possible that
you have incorrectly specified the number of rows and columns in the documentation file. Recheck your documentation
file for the correct information or use the method discussed above to determine the correct rows and columns based on
the image and header size. If an image seems to successfully import but the display of the image contains a venetian blind
striping effect, there are two possible causes. If the striping is vertical, try importing the image with BIPIDRIS instead. If
the striping is horizontal, try importing with BILIDRIS.
Importing GIS Data
Much of the information presented above regarding the import of satellite imagery also applies to the import of raster
GIS data. Specific import routines are also available in IDRISI for many common agency and software-specific GIS formats. The procedures for importing common GIS data formats are found in the on-line Help System. Search the Help
System index for the name of the format you would like to import.
Data Integration
Requirements for Data Integration
Data integration refers to the use of various data layers together in display or analysis. There are several parameters that
must match if data layers are to be used together. First, the reference systems must match.52 This is important because we
are often combining separate themes based on their spatial characteristics. A roads vector layer can be properly overlayed
on a raster satellite image, for example, only if both layers have the same reference system. A particular x,y coordinate pair
in the vector file must represent exactly the same place on the ground as that same x,y coordinate pair in the raster image.
52. The reference system projection, datum and grid system must all match for the reference system to match.
Chapter 10 Database Development
121
For raster layers to be used in image-to-image operations, such as OVERLAY, the extent of the images, the number of
rows and columns in the images, and the pixel resolution of the images must all match. Note that if any two of these three
conditions are met, the third is met automatically.
The steps used to integrate data layers, especially if data is transformed, should be recorded in the lineage field of the documentation file of each data layer. For example, it is trivial to make smaller pixels from larger pixels, and this is often done
to achieve data integration. However, it is important to note that the information carried by those smaller pixels still represents data gathered at a coarser resolution. Such information should be recorded in the new image's documentation file
so users will be aware that the apparent resolution exceeds that of the information.
Tools for Data Integration
The modules PROJECT and RESAMPLE are the primary tools available in IDRISI for georeferencing and changing reference systems. The chapter on Georeferencing details the differences between these two modules and when they
should be employed. In addition, the Tutorial includes exercises on the use of RESAMPLE and PROJECT.
The WINDOW module is used to change the extent of raster images. It works by saving a subset of a larger image as a
new image.
Changing the number of rows and columns in an image and/or changing the image pixel resolution can be accomplished
with the modules CONTRACT, EXPAND and PROJECT. CONTRACT and EXPAND work by thinning or duplicating pixels in an image by an integer multiple. For example, an image with 100 rows and 100 columns and a pixel resolution
of 30 meters could be contracted to an image of 25 rows and 25 columns by using a contraction factor of 4. Similarly, an
image of 100 rows and 100 columns could be transformed to an image of 1000 rows and 1000 columns by using
EXPAND and an expansion factor of 10.
If it is necessary to contract or expand an image by a non-integer factor (e.g., 1.5), CONTRACT and EXPAND cannot be
used. In these cases, the easiest method to use is PROJECT. Project to the same reference system that the file is currently
in, enter the new number of rows and columns, and choose the resampling technique that best suits your data and application. In general, if quantitative data are used, then the bilinear resampling type should be chosen, but if qualitative data
are used, the nearest neighbor resampling type should be chosen.
Conclusions
Database development is seldom quick and easy. The quality of the database, in many ways, limits the quality of any
resulting analysis to be carried out with the data. There are a significant number of technical pitfalls that are typically
encountered during database development and it is easy to become overwhelmed by these. However, despite these difficulties, it is important to maintain a project-driven approach to GIS analysis and to limit the influence data constraints can
have on the definition of the problems to be addressed.
IDRISI Guide to GIS and Image Processing Volume 1
122
Georeferencing
Georeferencing refers to the manner in which map locations are related to earth surface locations. Georeferencing
requires several ingredients:
•
a logic for referring to earth surface locations—a concern of the field of Geodesy;
•
a specific implementation of that logic, known as a Geodetic Datum—a concern of the field of Surveying;
•
a logic for referring locations to their graphic positions—a concern of the field of Cartography; and
•
an implementation of that logic, known as a data structure—a concern of GIS and Desktop Mapping software, and in this case, IDRISI.
Geodesy
Geodesy is that field of study which is concerned with the measurement of the size and shape of the earth and positions
upon it. The most fundamental problem that Geodesists face is the fact that the earth's surface is irregular in shape. For
example, imagine two locations (A and B) at either end of a thin straight beach bordered by a steep cliff (Figure 11-1).
Clearly the distance one would determine between the two locations would depend upon the route chosen to undertake
the measurement. Measuring the distance by the cliff route would clearly be longer than that determined along the coast.
Thus irregularities (hills and valleys) along the measuring surface can cause ambiguities in distance (and thereby, location).
To alleviate this situation, it has long been a common practice to reduce all measurements to a more regular measuring
surface—a reference surface.
Figure 11-1
Geoid
The oldest reference surface used for mapping is known as the geoid. The geoid can be thought of as mean sea level, or
where mean sea level would be if the oceans could flow under the continents. More technically, the geoid is an equipotential
surface of gravity defining all points in which the force of gravity is equivalent to that experienced at the ocean's surface.
Since the earth spins on its axis and causes gravity to be counteracted by centrifugal force progressively towards the equator, one would expect the shape of the geoid to be an oblate spheroid—a sphere-like object with a slightly fatter middle
and flattened poles. In other words, the geoid would have the nature of an ellipse of revolution—an ellipsoid.
As a reference surface, the geoid has several advantages—it has a simple physical interpretation (and an observable position along the coast), and it defines the horizontal for most traditional measuring instruments. Thus, for example, leveling
a theodolite or sextant is, by definition, a process of referring the instrument to the geoid.
Chapter 11 Georeferencing
123
Reference Ellipsoids
Unfortunately, as it turns out, the geoid is itself somewhat irregular. Because of broad differences in earth materials (such
as heavier ocean basin materials and lighter continental materials, irregular distributions such as mountains, and isostatic
imbalances), the geoid contains undulations that also introduce ambiguities of distance and location. As a result, it has
become the practice of modern geodetic surveys to use abstract reference surfaces that are close approximations to the
shape of the geoid, but which provide perfectly smooth reference ellipsoids (Figure 11-2). By choosing one that is as close
an approximation as possible, the difference between the level of a surveying instrument (defined by the irregular geoid)
and the horizontal of the reference ellipsoid is minimized. Moreover, by reducing all measurements to this idealized
shape, ambiguities of distance (and position) are removed.
geoid
earth surface
reference ellipsoid
Figure 11-2
There are many different ellipsoids in geodetic use (see Appendix 1:
Ellipsoid Parameters). They can be defined either by the length of the
major (a) and minor (b) semi-axes53 (Figure 11-3), or by the length of
the semi-major axis along with the degree of flattening [f = (a-b) / a].
The reason for having so many different ellipsoids is that different ones
give better fits to the shape of the geoid at different locations. The ellipsoid chosen for use is that which best fits the geoid for the particular
location of interest.
b
a
Figure 11-3
Geodetic Datums
Selecting a specific reference ellipsoid to use for a specific area and orienting it to the landscape, defines what is known in
Geodesy as a datum (note that the plural of datum in geodesy is datums, not data!). A datum thus defines an ellipsoid (itself
defined by the major and minor semi-axes), an initial location, an initial azimuth (a reference direction to define the direction of north), and the distance between the geoid and the ellipsoid at the initial location. Establishing a datum is the task
of geodetic surveyors, and is done in the context of the establishment of national or international geodetic control survey
networks. A datum is thus intended to establish a permanent reference surface, although recent advances in survey technology have led many nations to redefine their current datums.
Most datums only attempt to describe a limited portion of the earth (usually on a national or continental scale). For example, the North American Datum (NAD) and the European Datum each describe large portions of the earth, while the
Kandawala Datum is used for Sri Lanka alone. Regardless, these are called local datums since they do not try to describe
the entire earth. By contrast, we are now seeing the emergence of World Geodetic Systems (such as WGS84) that do try
53. The major and minor semi-axes are also commonly referred to as the semi-major and semi-minor axes.
IDRISI Guide to GIS and Image Processing Volume 1
124
to provide a single smooth reference surface for the entire globe. Such systems are particularly appropriate for measuring
systems that do not use gravity as a reference frame, such as Global Positioning Systems (GPS). However, presently they
are not very commonly found as a base for mapping. More typically one encounters local datums, of which several hundred are currently in use.
Datums and Geodetic Coordinates
Perhaps the most important thing to bear in mind about datums is that each defines a different concept of geodetic coordinates—latitude and longitude. Thus, in cases where more than one datum exists for a single location, more than one
concept of latitude and longitude exists. It can almost be thought of as a philosophical difference. It is common to assume
that latitude and longitude are fixed geographic concepts, but they are not. There are several hundred different concepts
of latitude and longitude currently in use (one for each datum). It might also be assumed that the differences between
them would be small. However, that is not necessarily the case. In North America, a change is being undertaken to convert all mapping to a recently defined datum called NAD83. At Clark University, for example, if one were to measure latitude and longitude according to NAD83 and compare it to the ground position of the same coordinates in the previous
system, NAD27, the difference is in excess of 40 meters! Other locations in the US experience differences in excess of
100 meters. Clearly, combining data from sources measured according to different datums can lead to significant discrepancies.
The possibility that more than one datum will be encountered in a mapping project is actually reasonably high. In recent
years, many countries have found the need to replace older datums with newer ones that provide a better fit to local geoidal characteristics. In addition, regional or international projects involving data from a variety of countries are very likely
to encounter the presence of multiple datums. As a result, it is imperative to be able to transform the geodetic coordinates
of one system to those of another. In IDRISI, the PROJECT option under the Reformat menu incorporates full datum
transformation as a part of its operation.
Cartographic Transformation
Once a logic has been established for referring to earth locations and a set of measurements has been made, a means of
storing and analyzing those positions is required. Traditionally, maps have been the preferred medium for both storage
and analysis, while today that format has been supplemented by digital storage and analysis. Both, however, share a common trait—they are most commonly flat! Just as flat maps are a more manageable medium than map globes, plane coordinates are a more workable medium than spherical (ellipsoidal) coordinates for digital applications. As a result, surveyed
locations are commonly transformed to a plane grid referencing system before use.
Projection
The process of transforming spheroidal geodetic coordinates to plane coordinate positions is known as projection, and
falls traditionally within the realm of cartography. Originally, the concern was only with a one-way projection of geodetic
coordinates to the plane coordinates of a map sheet. With the advent of GIS, however, this concern has now broadened
to include the need to undertake transformations in both directions in order to develop a unified database incorporating
maps that are all brought to a common projection. Thus, for example, a database developed on a Transverse Mercator
projection might need to incorporate direct survey data in geodetic coordinates along with map data in several different
projections. Back projecting digitized data from an existing projection to geodetic coordinates and subsequently using a forward projection to bring the data to the final projection is thus a very common activity in GIS. The PROJECT module in
IDRISI supports both kinds of transformation.
Projection of spheroidal positions to a flat plane simply cannot be done without some (and oftentimes considerable) distortion. The stretching and warping required to make the transformation work leads to continuous differences in scale
over the face of the map, which in turn leads to errors in distance, area and angular relationships. However, while distor-
Chapter 11 Georeferencing
125
tion is inevitable, it is possible to engineer that distortion such that errors are minimized and certain geometrical qualities
are maintained.
Of particular importance to GIS are the family of projections known as conformal or orthomorphic. These are projections in
which the distortion is engineered in such a manner that angular relationships are correctly preserved in the near vicinity
of any location. Traditionally, the field of surveying has relied upon the measurement of angular relationships (using
instruments such as a theodolite) in order to measure accurate distances over irregular terrain. As a result, conformal projections have been important for the correct transfer of field measurements to a map base, or for the integration of new
surveys with existing map data. For the same reasons, conformal projections are also essential to many navigation procedures. As a consequence, virtually all topographic map bases are produced on conformal projections, which form the
basis for all of the major grid referencing systems in use today.
Grid Referencing Systems
A grid referencing system can be thought of very simply as a systematic way in which the plane coordinates of the map sheet can be
related back to the geodetic coordinates of measured earth positions. Clearly, a grid referencing system requires a projection (most
commonly a conformal one). It also requires the definition of a
plane cartesian coordinate system to be superimposed on top of that
projection. This requires the identification of an initial position that
can be used to orient the grid to the projection, much like an initial
position is used to orient a datum to the geoid. This initial position is
called the true origin of the grid, and is commonly located at the position where distortion is least severe in the projection (Figure 11-4).
Then, like the process of orienting a datum, a direction is established
to represent grid north. Most commonly, this will coincide with the
direction of true north at the origin. However, because of distortion,
it is impossible for true north and grid north to coincide over many
other locations.
true origin
false origin
Once the grid has been oriented to the origin and true north, a numbering system and units of measure are determined. For example,
the UTM (Universal Transverse Mercator) system uses the equator
Figure 11-4
and the central meridian of a 6-degree wide zone as the true origin
for the northern hemisphere. The point is then given arbitrary coordinates of 500,000 meters East and 0 meters North. This then gives a false origin 500 kilometers to the west of the true origin (see figure above). In other words, the false origin marks the location where the numbering system is 0 in both axes.
In IDRISI, the numbering logic of a grid referencing system is always given by specifying the latitude and longitude of the
true origin and the arbitrary coordinates that exist at that point (in this example, 500,000 E and 0 N).
Georeferencing in IDRISI
The logic that IDRISI uses for georeferencing is quite simple, and is based on the issues previously described.
IDRISI Guide to GIS and Image Processing Volume 1
126
All geographic files are assumed to be stored according
to a grid reference system where grid north is aligned
with the edges of the raster image or vector file. As a
result, the minimum X of a raster image is always the
left-hand edge, the maximum Y is the top edge, and so
on (Figure 11-5).The georeferencing properties of an
IDRISI coverage (accessible through Metadata) include
an entry specifying the reference system used by that file
when referring to geographic locations. The particulars
of that reference system (e.g., projection, datum, origin,
etc.) are then contained in a reference system parameter file
(.ref) (see below). Whenever a projection or datum transformation is required, the PROJECT module refers to
this information to control the transformation process.
Maximum Y
Grid
North
Minimum Y
Minimum X
Maximum X
Figure 11-5
Every grid reference system must have a reference system parameter file. The only exception to this is a system identified
by the keyword "plane". Any coverage that indicates a plane coordinate referencing system is understood to use an arbitrary plane system for which geodetic and projection parameters are unknown, and for which a reference system parameter file is not provided. A coverage in a plane reference system cannot be projected.
Over 400 reference system parameter files are supplied with IDRISI. However, .ref files can be created for any grid referencing system using the METADATA or EDIT modules. Details on the structure of this simple ASCII text file format
are provided below and in the on-line Help System.
For simplicity, geodetic coordinates (lat/long) are recognized as a special type of grid referencing system. The true and
false origins are identical and occur at 0 degrees of longitude and 0 degrees of latitude. Units are in decimal degrees or
radians. Longitudes west of the prime (Greenwich) meridian and latitudes south of the equator are expressed as negative
numbers. Thus, for example, Clark University has approximate coordinates of -71.80,+42.27 (71°48' W, 42°16' N).
Although geodetic coordinates are truly spheroidal, they are logically treated here as a plane coordinate system. Thus, they
are implicitly projected according to a Plate Carrée projection,54 although the projection will actually be listed as "none".
Be aware that there are many possible interpretations of the concept of latitude and longitude, depending upon the datum
in use. A single .ref file has been supplied (called LATLONG) for the case where the datum is WGS84. This should be
copied and modified for other interpretations.
In IDRISI, the registration point for referencing raster images in the lower left corner of any cell. Thus, for example, with
an image of 10 columns by 5 rows, using a plane reference system of cells 10 meters by 10 meters, the lower left-most corner of the lower left-most cell has the coordinates 0,0 wile the upper right-most corner of the upper right-most cell has
the position 100.50. Note also that this lower left-most cell is considered to be in column 0 and row 4 while the upper
right-most cell is in column 0, row 0 (Figure 11-6).
IDRISI image and vector documentation files contain several fields that describe the reference system of the file. The min
X, max X, min Y, and max Y entries give the edge coordinates of an image or bounding rectangle of a vector file in reference system coordinates. For image files, the number of rows and columns is also given.
The reference system name (ref. system) identifies the reference system of the file, and may be “plane”, “lat/long”, or a specific reference system described by a reference system parameter file. In the latter case, the ref. system entry in the documentation file should match exactly the name of the corresponding reference system parameter file, without the .ref
extension.
54. Note, however, that geodetic coordinates are not considered to be identical to a system based on the Plate Carrée since no new numbering scheme
has been undertaken. PROJECT supports transformations based on the Plate Carrée as well, where it is understood that a true plane coordinate system
with its own special numbering scheme has been superimposed onto the projected geodetic coordinates.
Chapter 11 Georeferencing
127
The reference units entry indicates the unit of measure used in the reference coordinate system (e.g., meters).
The unit distance refers to the ground distance spanned by a distance of one unit (measured in reference units) in the reference system. Thus, for example, if the hypothetical image in Figure 11-6 had a unit distance of 2.0 and reference units in
meters, it would imply that as we move one in the reference coordinates (e.g. from x=99 to x=100) we actually move 2.0
meters on the ground. Simply think of the unit distance parameter as a multiplier that should be applied to all coordinates
to yield the reference units indicated. The unit distance will be 1.0 in most cases. However, the unit distance should be
specified as a value other than 1.0 whenever you have reference units that are not among the standard units supported
(meters, feet, miles, kilometers, degrees or radians). For example, a data file where the coordinates were measured in minutes of arc (1/60 of a degree) should have reference units set to degrees and the unit distance set to 0.016667.
In IDRISI, as you move the cursor over a displayed image, the row and column position and the X and Y coordinates are
indicated on the status bar at the bottom of the screen. Vector files can also be overlaid (using Composer) onto the raster
image, provided they have the same reference system. There is no need for the bounding rectangle of the vector file to match that of the
raster image. IDRISI will correctly overlay the portion of the vector file that overlaps the displayed raster image. Onscreen
digitizing will always output coordinates in the same reference system as the raster image being displayed.
Reference System Parameter Files
As previously indicated, IDRISI comes with over 400 reference system parameter files. These include one for geodetic
coordinates (latitude/longitude) using the WGS84 datum, 160 for the UTM system (one each for the 60 zones, for both
the northern and southern hemispheres) using the WGS84 datum, 32 based on the Gauss-Kruger projection, 40 for the
UTM system covering North America using NAD27 and NAD83, and 253 for all US State Plane Coordinate (SPC) systems based on the Lambert Conformal Conic and Transverse Mercator projections. Appendix 3: Supplied Reference System Parameter Files lists each of these files by geographic location. In CartaLinx, the Georeferencing tab of the
Preferences/Properties/Options dialog provides a list box in which any specific .ref file can be selected. During program installation, these supplied files are installed in the \Georef subfolder of your IDRISI program foler. The contents of these
files may be viewed with METADATA or EDIT.
Where .ref Files Are Stored
As indicated above, supplied .ref files are stored in a subfolder of the IDRISI program folder called \Georef. If your
IDRISI program folder is C:\Idrisi32, then the supplied reference system parameter files are in c:\Idrisi32\Georef. When
you create new reference system parameter files, we recommend that you store them in this folder, always adding to your
master library of reference files. However, reference system parameter files may be stored anywhere. When a reference
system parameter file is given without its path in a dialog box or macro, IDRISI always looks first in the working folder,
followed by the resource folders in the order they are named in the current project. If the file is not found, IDRISI will
next look in the \Georef subfolder.
Creating New .ref Files
Although IDRISI supplies more than 400 reference system parameter files, many users will need to create new .ref files to
suit the needs of specific systems. There are two options available to do this:
1.
For cases where a different version of the UTM system is required, the module UTMREF can be used. This will
simply require that you enter the zone number, the hemisphere (northern or southern), and the name of the datum along
with its associated Molodensky constants. The constants can be found in Appendix 2: Mododensky Constants for
Selected Geodetic Datums and are used by PROJECT to undertake datum transformations. If you enter a datum that is
not included, you will need to supply the name of the reference ellipsoid and the length of its major and minor semi-axes.
These constants for many reference ellipsoids are given in Appendix 1: Reference Ellipsoids.
2.
For all other cases, METADATA or EDIT can be used to create the appropriate file. Often the easiest procedure is to use the copy function in the IDRISI File Explorer to duplicate an existing .ref file, and then modify it using
METADATA or EDIT. The section below indicates the structure and contents of a .ref file.
IDRISI Guide to GIS and Image Processing Volume 1
128
The .ref File Structure
Like all IDRISI documentation files, .ref files are simple ascii text files that can be modified by means of any ASCII text
editor.
The .ref file type follows the conventions of other documentation files by having the first 14 characters of each line
devoted to a field description. Here is an example of a reference system parameter file named UTM-19N :
ref. system
projection
datum
delta WGS84
ellipsoid
major s-ax
minor s-ax
origin long
origin lat
origin X
origin Y
scale fac
units
parameters
:
:
:
:
:
:
:
:
:
:
:
:
:
:
Universal Transverse Mercator Zone 19
Transverse Mercator
NAD27
-8 160 176
Clarke 1866
6378206.5
6356584
-69
0
500000
0
1
m
0
The first line is simply a title and has no analytical use. The second line indicates the projection. In the current release of
IDRISI, the following projections are supported:
Mercator
Transverse Mercator
Gauss-Kruger (Gauss-Krueger spelling also accepted.)
Lambert Conformal Conic
Plate Carrée
Hammer Aitoff
Lambert North Polar Azimuthal Equal Area
Lambert South Polar Azimuthal Equal Area
Lambert Transverse Azimuthal Equal Area
Lambert Oblique Azimuthal Equal Area
North Polar Stereographic
South Polar Stereographic
Transverse Stereographic
Oblique Stereographic
Albers Equal Area Conic
none (i.e., geodetic coordinates)
Note that each of the names above are keywords and must appear exactly in this form to be understood by the PROJECT module.
The next line lists the geodetic datum. The text here is informational only, with the exception of the keywords NAD27
and NAD83. These latter two cases are the keywords for the two existing implementations of the North American datum.
When PROJECT encounters these keywords as elements of the input and output systems, it uses a special (and more precise) datum transformation technique. For all other cases, the Molodensky transform is used.
Chapter 11 Georeferencing
129
The next line lists the differences between the center (i.e., earth center) of the datum in use and that of the WGS84 datum
(World Geodetic System, 1984). The values express, in meters, the differences in the X, Y and Z axes respectively, relative
to WGS84. These are known as the Molodensky constants. Appendix 2: Datum Parameters contains a list of these
constants for most datums worldwide. They are used by PROJECT to undertake datum transformations except in the
case of conversions between NAD27 and NAD83. These conversions use the US National Geodetic Survey's NADCON
procedure.
The next three lines give information about the reference ellipsoid. The line listing the name of the ellipsoid is for informational purposes only. The following two lines describe, in meters, the major and minor semi-axes of the ellipsoid used
by the datum. These entries are used analytically by the PROJECT module. Appendix 1: Reference Ellipsoids contains
a list of ellipsoids and the lengths of their semi-axes.
The next four lines describe the origin and numbering system of the reference system. The origin long and origin lat entries
indicate the longitude and latitude of the true origin. The origin X and origin Y entries then indicate what coordinate values
exist at that location in the reference system being described.
The scale fac entry indicates the scale factor to be applied at the center of the projection. A typical value would be 1.0,
although it is equally permissible to indicate "na" (an abbreviation for "not applicable"). The most likely case in which
users will encounter a need to specify a value other than 1.0 is with the UTM system and other systems based on the
Transverse Mercator projection. With the UTM system, the scale fac should read 0.9996.
The units entry duplicates the units information found in raster and vector documentation files and is included here for
confirmation by the system. Therefore, in both reference and documentation files, units should be considered as analytical
entries.
Finally, the parameters entry indicates the number of special parameters included with the .ref file information. Of the projections currently supported, only the Lambert Conformal Conic and Alber's Equal Area Conic require special parameters. All others should have a value of 0 for this entry. In cases using either the Lambert Conformal Conic or Alber's
Equal Area Conic projections, the parameters entry should read 2 and then include the latitudes (in decimal degrees) of the
two standard lines (lines of no distortion) on the following two lines. Here, for example, is the .ref file for the Massachusetts State Plane Coordinate System, Zone 1 (SPC83MA1), based on the NAD83 datum:
ref. system
projection
datum
delta WGS84
ellipsoid
major s-ax
minor s-ax
origin long
origin lat
origin X
origin Y
scale fac
units
parameters
stand ln 1
stand ln 2
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
Massachusetts State Plane Coordinate System Mainland Zone
Lambert Conformal Conic
NAD83
000
GRS80
6378137
6356752.31
-71.5
41
200000
750000
na
m
2
41.72
42.68
Note that the latitude of the "lower" standard line (i.e., that which is nearer the equator) is listed first and that of the lati-
IDRISI Guide to GIS and Image Processing Volume 1
130
tudinally "higher" standard line is listed second. Future additions of other projections may dictate the need for additional
special parameters. However, in this version, only those reference systems based on the Lambert Conformal Conic and
Alber's Equal Area Conic projections require special parameters.
Projection and Datum Transformations
The PROJECT option of the COVERAGE menu in CartaLinx provides full transformation between reference systems.
As a consequence, it undertakes transformation of both projection and datum characteristics.
Projection transformations for all supported projections are undertaken using ellipsoidal formulas accurate to 2 cm on the
ground. As CartaLinx uses double precision floating point numbers (15 significant figures) for the representation of coordinate data, this precision is maintained in all results. However, be aware that the precision of exported coordinates
depends on the numeric precision used by the export format being used.
For datum transformations, the primary method used is the Molodensky transformation described in DMA (1987).
Whenever the keywords "NAD27" and "NAD83" are encountered in a transformation, however, the more precise US
National Geodetic Survey NADCON procedure is used (see Dewhurst, 1990). This latter procedure should only be used within
the continental US. For other regions covered by the North American Datum, do not use the NAD27 or NAD83 keywords,
but rather some other spelling in the datum field, and indicate the correct Molodensky transformation constants from
Appendix 2: Datum Parameters.55
Algorithms Used by PROJECT
Forward and backward ellipsoidal projection formulae were taken from Snyder (1987). For datum transformations, the
Molodensky transform procedure (DMA, 1987) is used. However, for the specific case of NAD27/NAD83 transformations within the continental US, the US National Geodetic Survey's NADCON procedure is used. This procedure is
described in Dewhurst (1990) and is accompanied by relevant data files. Briefly, the procedure involves a matrix of corrections to longitude and latitude that are accessed as a two-dimensional look-up table. Final correction values are then
interpolated between the four nearest values in the matrix using a bilinear interpolation. To facilitate this, the NADCON
data files for the continental US were converted into IDRISI images. These images are contained in the GEOREF subdirectory under the IDRISI program directory. There are two of these files, NADUSLON and NADUSLAT, for the corrections to longitude and latitude respectively.
Further Reading
Geodesy is not a familiar topic for many in GIS. However, as one can see from the material in this chapter, it is essential
to effective database development. The following readings provide a good overview of the issues involved:
Burkard, R.K., (1964) Geodesy for the Layman, (St. Louis, MO: USAF Aeronautical Chart and Information Center).
Smith, J.R., (1988) Basic Geodesy, (Rancho Cordova, CA: Landmark Enterprises).
For projections, there are few texts that can match the scope and approachability of:
55. Note that while a set of constants exist for the North American Datum as a whole, Molodensky constants also exist for more specific portions of the
regions covered to enhance the precision of the transformation.
Chapter 11 Georeferencing
131
Maling, D.H., (1973) Coordinate Systems and Map Projections, (London: George Phillip and Son).
Snyder, J.P., (1987) Map Projections: A Working Manual. U.S. Geological Survey Professional Paper 1395, (Washington, DC: US Government Printing Office).
Finally, with respect to the procedures used for datum transformations, please refer to:
DMA, (1987) Department of Defense World Geodetic System 1984: Its Definition and Relationships with Local Geodetic Datums, DMA TR 8350.2, (Washington, DC: The Defense Mapping Agency).
Dewhurst, W.T., (1990) NADCON: The Application of Minimum Curvature-Derived Surfaces in the Transformation
of Positional Data from the North American Datum of 1927 to the North American Datum of 1983, NOAA
Technical Memorandum NOS NGS-50, (Rockville, MD: National Geodetic Information Center).
IDRISI Guide to GIS and Image Processing Volume 1
132
Appendix 1: Ellipsoid Parameters
ELLIPSOID NAME
MAJOR SEMI-AXIS
MINOR SEMI-AXIS
Airy
6377563.396
6356256.909
Modified Airy
6377340.189
6356034.448
Australian National
6378160.000
6356774.719
Average Terestrial System 1997
6378135
6356750.305
Bessel 1841 (Ethiopia, Indonesia, Japan, Korea)
6377397.155
6356078.963
Bessel 1841 (Namibia)
6377483.865
6356165.383
Bessel Modified
6377492.018
6356173.509
Clarke 1858
20926348
20855233
Clarke 1866
6378206.400
6356583.800
Clarke 1866 Michigan
20926631.531
20855688.674
Clarke 1880
6378249.145
6356514.870
Clarke 1880 Benoit
6378300.789
6356566.435
Clarke 1880 IGN
6378249.2
6356515
Clarke 1880 (SGA 1922)
6378249.2
6356514.997
Everest - India 1830
6377276.345
6356075.413
Everest - India 1956
6377301.24
6356100.23
Everest - Pakistan
6377309.61
6356108.57
Everest - Sabah and Sarawak
6377298.56
6356097.55
Everest - West Malaysia 1969
6377295.66
6356094.67
Everest - West Malaysia and Singapore 1948
6377304.063
6356103.039
Everest (1830 definition)
20922931.8
20853374.58
Everest 1830 (1962 definition)
6377301.243
6356100.23
Everest 1830 (1967 definition)
6377298.556
6356097.55
Everest 1830 (1975 definition)
6377299.151
6356098.145
Fischer 1960
6378166.000
6356784.280
Modified Fischer 1960
6378155.000
6356773.320
Fischer 1968
6378150.000
6356768.340
GEM 10C
6378137
6356752.314
GRS 1967
6378160
6356774.516
Appendix 1: Ellipsoid Parameters
133
ELLIPSOID NAME
MAJOR SEMI-AXIS
MINOR SEMI-AXIS
GRS 1980
6378137.000
6356752.314
Helmert 1906
6378200.000
6356818.170
Hough
6378270.000
6356794.343
Indonesian 1974
6378160.00
6356774.50
International 1924
6378388.000
6356911.946
Krassovsky 1940
6378245.000
6356863.019
NWL 9D
6378145
6356759.769
OSU86F
6378136.2
6356751.517
OSU91A
6378136.3
6356751.617
Plessis 1817
6376523
6355862.933
SGS 85
6378136.000
6356751.300
South American 1969
6378160.000
6356774.719
Sphere
6371000
6371000
Struve 1860
6378298.3
6356657.143
War Office
6378300.583
6356752.27
WGS 60
6378165.000
6356783.290
WGS 66
6378145.000
6356759.770
WGS 72
6378135.000
6356750.520
WGS 84
6378137.000
6356752.314
IDRISI Guide to GIS and Image Processing Volume 1
134
Appendix 2: Datum Parameters
The following table contains the constants required for the Molodensky Datum Transformation procedure. The DX, DY
and DZ values are the three values (in that order) that should be specified in the "Delta WGS84" field of the Reference
System Parameter File. These values represent the three-dimensional difference in position of the datum ellipsoid from
that of WGS84. The values listed here were taken from the European Petroleum Survey Group Database.
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
ABIDJAN 1987
Cote D’Ivoire (Ivory Coast)
Clarke 1880
-124.76
53
466.79
ADINDAN
MEAN FOR Ethiopia, Sudan
Clarke 1880
-166
-15
204
Burkina Faso
-118
-14
218
Cameroon
-134
-2
210
Ethiopia
-165
-11
206
Mali
-123
-20
220
Senegal
-128
-18
224
Sudan
-161
-14
205
AFGOOYE
Somalia
Krassovsky 1940
-43
-163
45
AIN EL ABD 1970
Bahrain Island
International 1924
-150
-250
-1
-143
-236
7
Saudi Arabia
AMERICAN SAMOA 1962
American Samoa Islands
Clarke 1866
-115
118
426
AMERSFOORT
Netherlands
Bessel 1841
593.16
26.15
478.54
ANNA 1 ASTRO 1965
Cocos Islands
Australian National
-491
-22
435
ANTIGUA ISLAND ASTRO 1943
Antigua (Leeward Islands)
Clarke 1880
-270
13
62
Appendix 2: Datum Parameters
135
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
ARC 1950
MEAN FOR Botswana, Lesotho, Malawi, Swaziland, Zaire, Zambia, Zimbabwe
Clarke 1880
-143
-90
-294
Botswana
-138
-105
-289
Burundi
-153
-5
-292
Lesotho
-125
-108
-295
Malawi
-161
-73
-317
Swaziland
-134
-105
-295
Zaire
-169
-19
-278
Zambia
-147
-74
-283
Zimbabwe
-142
-96
-293
ARC 1960
MEAN FOR Kenya, Tanzania
Clarke 1880
-160
-6
-302
ASCENSION ISLAND 1958
Ascension Island
International 1924
-205
107
53
ASTRO BEACON E 1945
Iwo Jima
International 1924
145
75
-272
ASTRO DOS 71/4
St. Helena Island
International 1924
-320
550
-494
ASTRO TERN ISLAND (FRIG) 1961
Tern Island
International 1924
114
-116
-333
ASTRONOMICAL STATION 1952
Marcus Island
International 1924
124
-234
-25
AUSTRALIAN GEODETIC 1966
Australia & Tasmania
Australian National
-133
-48
148
AUSTRALIAN GEODETIC 1984
Australia & Tasmania
Australian National
-134
-48
149
AYABELLE LIGHTHOUSE
Djibouti
Clarke 1880
-79
-129
145
BATAVIA
Indonesia (Sumatra)
Bessel 1841
-377
681
-50
BELLEVUE (IGN)
Efate & Erromango Islands
International 1924
-127
-769
472
BERMUDA 1957
Bermuda
Clarke 1866
-73
213
296
BISSAU
Guinea - Bissau
International 1924
-173
253
27
BOGOTA OBSERVATORY
Colombia
International 1924
307
304
-318
BUKIT RIMPAH
Indonesia (Bangka & Belitung Islands)
Bessel 1841
-384
664
-48
CAMP AREA ASTRO
Antarctica (McMurdo Camp Area)
International 1924
-104
-129
239
CAMPO INCHAUSPE
Argentina
International 1924
-148
136
90
CANTON ASTRO 1966
Phoenix Islands
International 1924
298
-304
-375
CAPE
South Africa
Clarke 1880
-136
-108
-292
CAPE CANAVERAL
Bahamas, Florida
Clarke 1866
-2
151
181
CARTHAGE
Tunisia
Clarke 1880
-263
6
431
CHATHAM ISLAND ASTRO 1971
New Zealand (Chatham Island)
International 1924
175
-38
113
CHTRF95
Liechtenstein, Switzerland
GRS1980
0
0
0
IDRISI Guide to GIS and Image Processing Volume 1
136
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
CHUA ASTRO
Paraguay
International 1924
-134
229
-29
CORREGO ALEGRE
Brazil
International 1924
-206
172
-6
DABOLA
Guinea
Clarke 1880
-83
37
124
DECEPTION ISLAND
Deception Island, Antarctica
Clarke 1880
260
12
-147
DJAKARTA (BATAVIA)
Indonesia (Sumatra)
Bessel 1841
-377
681
-50
DOMINICA 1945
Dominica
Clarke 1880
725
685
536
DOS 1968
New Georgia Islands (Gizo Island)
International 1924
230
-199
-752
EASTER ISLAND 1967
Easter Island
International 1924
211
147
111
EGYPT 1907
Egypt
Helmert 1906
-130
110
-13
EST92
Estonia
GRS80
0.055
-0.541
-0.185
ETRF89
Europe
GRS80
0
0
0
EUROPEAN 1950
MEAN FOR Austria, Belgium, Denmark, Finland, France, West Germany, Gibraltar,
Greece, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland
International 1924
-87
-98
-121
MEAN FOR Austria, Denmark, France, West
Germany, Netherlands, Switzerland
-87
-96
-120
MEAN FOR Iraq, Israel, Jordan, Lebanon,
Kuwait, Saudi Arabia, Syria
-103
-106
-141
Cyprus
-104
-101
-140
Egypt
-130
-117
-151
England, Channel Islands, Ireland, Scotland,
Shetland Islands
-86
-96
-120
Finland, Norway
-87
-95
-120
France
-84
-97
-117
Greece
-84
-95
-130
Iran
-117
-132
-164
Italy (Sardinia)
-97
-103
-120
Italy (Sicily)
-97
-88
-135
Malta
-107
-88
-149
Portugal, Spain
-84
-107
-120
Tunisia
-112
-77
-145
United Kingdom UKCS offshore east of 6 deg.
west
-89.5
-93.8
-123.1
Appendix 2: Datum Parameters
137
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
EUROPEAN 1979
MEAN FOR Austria, Finland, Netherlands,
Norway, Spain, Sweden, Switzerland
International 1924
-86
-98
-119
FD58
Iran (Kangan district)
Clarke 1880
-241.54
-163.64
396.06
FORT THOMAS 1955
Nevis, St. Kitts (Leeward Islands)
Clarke 1880
-7
215
225
GAN 1970
Republic of Maldives
International 1924
-133
-321
50
GEODETIC DATUM 1949
New Zealand
International 1924
84
-22
209
GGRS87
Greece
GRS 1980
-199.87
74.79
246.62
GRACIOSA BASE SW 1948
Azores (Faial, Graciosa, Pico, Sao Jorge, Terceira)
International 1924
-104
167
-38
GRENADA 1953
Grenada
Clarke 1880
72
213.7
93
GUAM 1963
Guam
Clarke 1866
-100
-248
259
GUNUNG SEGARA
Indonesia (Kalimantan)
Bessel 1841
-403
684
41
GUX 1 ASTRO
Guadalcanal Island
International 1924
252
-209
-751
HERAT NORTH
Afghanistan
International 1924
-333
-222
114
HJORSEY 1955
Iceland
International 1924
-73
46
-86
HONG KONG 1963
Hong Kong
International 1924
-156
-271
-189
HU-TZU-SHAN
Taiwan
International 1924
-637
-549
-203
INDIAN
Bangladesh
Everest 1830
282
726
254
INDIAN
India, Nepal
Everest 1956
295
736
257
INDIAN
Pakistan
Everest
283
682
231
INDIAN 1954
Thailand
Everest 1830
217
823
299
INDIAN 1960
Vietnam (near 16oN)
Everest 1830
198
881
317
182
915
344
Fahud
Con Son Island (Vietnam)
INDIAN 1975
Thailand
Everest 1830
209
818
290
INDONESIAN 1974
Indonesia
Indonesian 1974
-24
-15
5
IRELAND 1965
Ireland
Modified Airy
506
-122
611
ISTS 061 ASTRO 1968
South Georgia Islands
International 1924
-794
119
-298
ISTS 073 ASTRO 1969
Diego Garcia
International 1924
208
-435
-229
JOHNSTON ISLAND 1961
Johnston Island
International 1924
189
-79
-202
KANDAWALA
Sri Lanka
Everest 1830
-97
787
86
KERGUELEN ISLAND 1949
Kerguelen Island
International 1924
145
-187
103
KERTAU 1948
West Malaysia & Singapore
Everest 1948
-11
851
5
IDRISI Guide to GIS and Image Processing Volume 1
138
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
KUSAIE ASTRO 1951
Caroline Islands, Federal States of Micronesia
International 1924
647
1777
-1124
L. C. 5 ASTRO 1961
Cayman Brac Island
Clarke 1866
42
124
147
LEIGON
Ghana
Clarke 1880
-130
29
364
LIBERIA 1964
Liberia
Clarke 1880
-90
40
88
LUZON
Philippines (Excluding Mindanao)
Clarke 1866
-133
-77
-51
-133
-79
-72
Philippines (Mindanao)
MAHE 1971
Mahe Island
Clarke 1880
41
-220
-134
MASSAWA
Ethiopia (Eritrea)
Bessel 1841
639
405
60
MERCHICH
Morocco
Clarke 1880
31
146
47
MIDWAY ASTRO 1961
Midway Island
International 1924
912
-58
1227
MINNA
Cameroon
Clarke 1880
-81
-84
115
-92
-93
122
Nigeria
MONTSERRAT ISLAND ASTRO 1958
Montserrat (Leeward Islands)
Clarke 1880
174
359
365
M'PORALOKO
Gabon
Clarke 1880
-74
-130
42
NAHRWAN
Oman (Masirah Island)
Clarke 1880
-247
-148
369
Saudi Arabia
-243
-192
477
United Arab Emirates
-249
-156
381
-10
375
165
NAPARIMA BWI
Trinidad & Tobago
Appendix 2: Datum Parameters
International 1924
139
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
NORTH AMERICAN 1927
MEAN FOR Antigua, Barbados, Barbuda,
Caicos Islands, Cuba, Dominican Republic,
Grand Cayman, Jamaica, Turks Islands
Clarke 1866
-3
142
183
MEAN FOR Belize, Costa Rica, El Salvador,
Guatemala, Honduras, Nicaragua
0
125
194
MEAN FOR Canada
-10
158
187
MEAN FOR Continental US (CONUS)
-8
160
176
MEAN FOR CONUS (East of Mississippi
River) Including Louisiana, Missouri, Minnesota
-9
161
179
MEAN FOR CONUS (West of Mississippi
River)
-8
159
175
Alaska (Excluding Aleutian Islands)
-5
135
172
Aleutian Islands (East of 180oW)
-2
152
149
Aleutian Islands (West of 180oW)
2
204
105
Bahamas (Except San Salvador Island)
-4
154
178
Bahamas (San Salvador Island)
1
140
165
Canada (Alberta, British Columbia)
-7
162
188
Canada (Manitoba, Ontario)
-9
157
184
Canada (New Brunswick, Newfoundland, Nova
Scotia, Quebec)
-22
160
190
Canada
chewan)
4
159
188
Canada (Yukon)
-7
139
181
Canal Zone
0
125
201
Cuba
-9
152
178
Greenland (Hayes Peninsula)
11
114
195
Mexico
-12
130
190
0
0
0
Aleutian Islands
-2
0
4
Hawaii
1
1
-1
NORTH AMERICAN 1983
(Northwest
Territories,
Saskat-
Alaska (Excluding Aleutian Islands), Canada,
Central America, CONUS, Mexico
GRS 80
NORTH SAHARA 1959
Algeria
Clarke 1880
-186
-93
310
OBSERVATORIO METER- EO 1939
Azores (Corvo & Flores Islands)
International 1924
-425
-169
81
OLD EGYPTIAN 1907
Egypt
Helmert 1906
-130
110
-13
IDRISI Guide to GIS and Image Processing Volume 1
140
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
OLD HAWAIIAN
MEAN FOR Hawaii, Kauai, Maui, Oahu
Clarke 1866
61
-285
-181
Hawaii
89
-279
-183
Kauai
45
-290
-172
Maui
65
-290
-190
Oahu
58
-283
-182
OMAN
ORD.
1936
SURVEY
GREAT
BRITIAN
Oman
Clarke 1880
-346
-1
224
MEAN FOR England, Isle of Man, Scotland,
Shetland Islands, Wales
Airy
375
-111
431
England
371
-112
434
England, Isle of Man, Wales
371
-111
434
Scotland, Shetland Islands
384
-111
425
Wales
370
-108
434
PICO DE LAS NIEVES
Canary Islands
International 1924
-307
-92
127
PITCAIRN ASTRO 1967
Pitcairn Island
International 1924
185
165
42
POINT 58
MEAN FOR Burkina Faso & Niger
Clarke 1880
-106
-129
165
POINTE NOIRE 1948
Congo
Clarke 1880
-148
51
-291
PORTO SANTO 1936
Porto Santo, Madeira Islands
International 1924
-499
-249
314
PROVISIONAL S. AMERICAN 1956
MEAN FOR Bolivia, Chile, Colombia, Ecuador,
Guyana, Peru, Venezuela
International 1924
-288
175
-376
Bolivia
-270
188
-388
Chile (Northern, Near 19oS)
-270
183
-390
Chile (Southern, Near 43oS)
-305
243
-442
Colombia
-282
169
-371
Ecuador
-278
171
-367
Guyana
-298
159
-369
Peru
-279
175
-379
Venezuela
-295
173
-371
PROVISIONAL S. CHILEAN 1963
Chile (South, Near 53oS) (Hito XVIII)
International 1924
16
196
93
PUERTO RICO
Puerto Rico, Virgin Islands
Clarke 1866
11
72
-101
PULKOVO 1942
Russia
Krassovsky 1940
28
-130
-95
QATAR NATIONAL
Qatar
International 1924
-128
-283
22
QORNOQ
Greenland (South)
International 1924
164
138
-189
Appendix 2: Datum Parameters
141
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
REUNION
Mascarene Islands
International 1924
94
-948
-1262
ROME 1940
Italy (Sardinia)
International 1924
-225
-65
9
S-42 (PULKOVO 1942)
Hungary
Krassovsky 1940
28
-121
-77
SANTO (DOS) 1965
Espirito Santo Island
International 1924
170
42
84
SAO BRAZ
Azores (Sao Miguel, Santa Maria Islands)
International 1924
-203
141
53
SAPPER HILL 1943
East Falkland Island
International 1924
-355
21
72
SCHWARZECK
Namibia
Bessel
(Namibia)
616
97
-251
SELVAGEM GRANDE 1938
Salvage Island
International 1924
-289
-124
60
SGS 85
Soviet Geodetic System 1985
SGS 85
3
9
-9
S-JTSK
Czechoslavakia (prior to 1 Jan. 1993)
Bessel 1841
589
76
480
SOUTH AMERICAN 1969
MEAN FOR Argentina, Bolivia, Brazil, Chile,
Colombia, Ecuador, Guyana, Paraguay, Peru,
Trinidad & Tobago, Venezuela
South
1969
-57
1
-41
Argentina
-62
-1
-37
Bolivia
-61
2
-48
Brazil
-60
-2
-41
Chile
-75
-1
-44
Colombia
-44
6
-36
Ecuador (Excluding Galapagos Islands)
-48
3
-44
Ecuador (Baltra, Galapagos)
-47
26
-42
Guyana
-53
3
-47
Paraguay
-61
2
-33
Peru
-58
0
-44
Trinidad & Tobago
-45
12
-33
Venezuela
-45
8
-33
7
-10
-26
1841
American
SOUTH ASIA
Singapore
Modified
1960
TANANARIVE OBSERVATORY 1925
Madagascar
International 1924
-189
-242
-91
TIMBALAI 1948
Brunei, East Malaysia (Sabah, Sarawak)
Everest (Sabah &
Sarawak)
-679
669
-48
IDRISI Guide to GIS and Image Processing Volume 1
Fischer
142
DATUM
LOCATION
ELLIPSOID
DX
DY
DZ
TOKYO
MEAN FOR Japan, Okinawa, South Korea
Bessel 1841
-148
507
685
Japan
-148
507
685
Okinawa
-158
507
676
South Korea
-146
507
687
TRISTAN ASTRO 1968
Tristan da Cunha
International 1924
-632
438
-609
VOIROL 1960
Algeria
Clarke 1880
-123
-206
219
VITI LEVU 1916
Fiji (Viti Levu Island)
Clarke 1880
51
391
-36
WAKE-ENIWETOK 1960
Marshall Islands
Hough
102
52
-38
WAKE ISLAND ASTRO 1952
Wake Atoll
International 1924
276
-57
149
WGS 1972
Global Definition
WGS 72
0
0
0
YACARE
Uruguay
International 1924
-155
171
37
ZANDERIJ
Surinam
International 1924
-265
120
-358
Appendix 2: Datum Parameters
143
IDRISI Guide to GIS and Image Processing Volume 1
144
Appendix 3: Supplied Reference System
Parameter Files
Geodetic (Latitude/Longitude)
A single REF file named LATLONG is supplied for geodetic coordinates. This file is based on the WGS84 datum. For
other datums, use the COPY function in the IDRIS File Explorer to copy this file to another name and then use METADATA or EDIT along with the data in Appendices 1 and 2 to enter the new datum information.
Universal Transverse Mercator (UTM)
160 REF files are supplied for the UTM system—60 for the northern hemisphere using the WGS84 datum, 60 for the
southern hemisphere using the WGS84 datum, 20 for North America based on NAD27 and 20 for North America based
on NAD83. The northern WGS84 group has names ranging from UTM-01N to UTM-60N, while the southern WGS84
group has names ranging from UTM-01S to UTM-60S. The NAD27 group (covering zones 1-20) has names ranging
from US27TM01 to US27TM20 while those for NAD83 range from US83TM01 to US83TM20. Note that the North
American groups all use a North American mean value for the Molodensky constants.
For other datums, the module UTMREF may be used to crate the reference system parameter file. In addition, several
very specific instances of the UTM system are available in the Miscellaneous group listed later in this appendix.
US State Plane Coordinate System 1927
REF files are supplied for all US State Plane Coordinate Systems based on the Transverse Mercator and Lambert Conformal Conic projections for NAD27 and NAD83. The following table lists these files for the NAD27 datum. The Projection
column indicates the projection upon which the system is based (L=Lambert Conformal Conic / TM=Transverse Mercator).
State
File name
Title
Projection
Alabama
SPC27AL1
Alabama State Plane Coordinate System Eastern Zone
TM
SPC27AL2
Alabama State Plane Coordinate Western Zone
TM
SPC27AK0
Alaska State Plane Coordinate System Zone 10
TM
SPC27AK2
Alaska State Plane Coordinate System Zone 2
TM
SPC27AK3
Alaska State Plane Coordinate System Zone 3
TM
SPC27AK4
Alaska State Plane Coordinate System Zone 4
TM
SPC27AK5
Alaska State Plane Coordinate System Zone 5
TM
SPC27AK6
Alaska State Plane Coordinate System Zone 6
TM
Alaska
Appendix 3: Supplied Reference System Parameter Files
145
State
File name
Title
Projection
SPC27AK7
Alaska State Plane Coordinate System Zone 7
TM
SPC27AK8
Alaska State Plane Coordinate System Zone 8
TM
SPC27AK9
Alaska State Plane Coordinate System Zone 9
TM
SPC27AZ1
Arizona State Plane Coordinate System Eastern Zone
TM
SPC27AZ2
Arizona State Plane Coordinate System Central Zone
TM
SPC27AZ3
Arizona State Plane Coordinate System Western Zone
TM
SPC27AR1
Arkansas State Plane Coordinate System Northern Zone
L
SPC27AR2
Arkansas State Plane Coordinate System Southern Zone
L
SPC27CA1
California State Plane Coordinate System Zone I
L
SPC27CA2
California State Plane Coordinate System Zone II
L
SPC27CA3
California State Plane Coordinate System Zone III
L
SPC27CA4
California State Plane Coordinate System Zone IV
L
SPC27CA5
California State Plane Coordinate System Zone V
L
SPC27CA6
California State Plane Coordinate System Zone VI
L
SPC27CA7
California State Plane Coordinate System Zone VII
L
SPC27CO1
Colorado State Plane Coordinate System Northern Zone
L
SPC27CO2
Colorado State Plane Coordinate System Central Zone
L
SPC27CO3
Colorado State Plane Coordinate System Southern Zone
L
Connecticut
SPC27CT1
Connecticut State Plane Coordinate System Zone 1
L
Delaware
SPC27DE1
Delaware State Plane Coordinate System Zone 1
TM
Florida
SPC27FL1
Florida State Plane Coordinate System Eastern Zone
TM
SPC27FL2
Florida State Plane Coordinate System Western Zone
TM
SPC27FL3
Florida State Plane Coordinate System Northern Zone
TM
SPC27GA1
Georgia State Plane Coordinate System Eastern Zone
TM
SPC27GA2
Georgia State Plane Coordinate System Western Zone
TM
SPC27HI1
Hawaii State Plane Coordinate System Zone 1
TM
SPC27HI2
Hawaii State Plane Coordinate System Zone 2
TM
SPC27HI3
Hawaii State Plane Coordinate System Zone 3
TM
SPC27HI4
Hawaii State Plane Coordinate System Zone 4
TM
SPC27HI5
Hawaii State Plane Coordinate System Zone 5
TM
SPC27ID1
Idaho State Plane Coordinate System Eastern Zone
TM
SPC27ID2
Idaho State Plane Coordinate System Central Zone
TM
SPC27ID3
Idaho State Plane Coordinate System Western Zone
TM
Arizona
Arkansas
California
Colorado
Georgia
Hawaii
Idaho
IDRISI Guide to GIS and Image Processing Volume 1
146
State
File name
Title
Projection
Illinois
SPC27IL1
Illinois State Plane Coordinate System Eastern Zone
TM
SPC27IL2
Illinois State Plane Coordinate System Western Zone
TM
SPC27IN1
Indiana State Plane Coordinate System Eastern Zone
TM
SPC27IN2
Indiana State Plane Coordinate System Western Zone
TM
SPC27IA1
Iowa State Plane Coordinate System Northern Zone
L
SPC27IA2
Iowa State Plane Coordinate System Southern Zone
L
SPC27KS1
Kansas State Plane Coordinate System Northern Zone
L
SPC27KS2
Kansas State Plane Coordinate System Southern Zone
L
SPC27KY1
Kentucky State Plane Coordinate System Northern Zone
L
SPC27KY2
Kentucky State Plane Coordinate System Southern Zone
L
SPC27LA1
Louisiana State Plane Coordinate System Northern Zone
L
SPC27LA2
Louisiana State Plane Coordinate System Southern Zone
L
SPC27LA3
Louisiana State Plane Coordinate System Offshore Zone
L
SPC27ME1
Maine State Plane Coordinate System Eastern Zone
TM
SPC27ME2
Maine State Plane Coordinate System Western Zone
TM
Maryland
SPC27MD1
Maryland State Plane Coordinate System Zone 1
L
Massachusetts
SPC27MA1
Massachusetts State Plane Coordinate System Mainland Zone
L
SPC27MA2
Massachusetts State Plane Coordinate System Island Zone
L
SPC27MI1
Current Michigan State Plane Coordinate System Northern Zone
L
SPC27MI2
Current Michigan State Plane Coordinate System Central Zone
L
SPC27MI3
Current Michigan State Plane Coordinate System Southern Zone
L
SPC27MI4
Old Michigan State Plane Coordinate System Eastern Zone
TM
SPC27MI5
Old Michigan State Plane Coordinate System Central Zone
TM
SPC27MI6
Old Michigan State Plane Coordinate System Western Zone
TM
SPC27MN1
Minnesota State Plane Coordinate System Northern Zone
L
SPC27MN2
Minnesota State Plane Coordinate System Central Zone
L
SPC27MN3
Minnesota State Plane Coordinate System Southern Zone
L
SPC27MS1
Mississippi State Plane Coordinate System Eastern Zone
TM
SPC27MS2
Mississippi State Plane Coordinate System Western Zone
TM
SPC27MO1
Missouri State Plane Coordinate System Eastern Zone
TM
SPC27MO2
Missouri State Plane Coordinate System Central Zone
TM
SPC27MO3
Missouri State Plane Coordinate System Western Zone
TM
SPC27MT1
Montana State Plane Coordinate System Northern Zone
L
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Michigan
Minnesota
Mississippi
Missouri
Montana
Appendix 3: Supplied Reference System Parameter Files
147
State
File name
Title
Projection
SPC27MT2
Montana State Plane Coordinate System Central Zone
L
SPC27MT3
Montana State Plane Coordinate System Southern Zone
L
SPC27NE1
Nebraska State Plane Coordinate System Northern Zone
L
SPC27NE2
Nebraska State Plane Coordinate System Southern Zone
L
SPC27NV1
Nevada State Plane Coordinate System Eastern Zone
TM
SPC27NV2
Nevada State Plane Coordinate System Central Zone
TM
SPC27NV3
Nevada State Plane Coordinate System Western Zone
TM
New Hampshire
SPC27NH1
New Hampshire State Plane Coordinate System Zone 1
TM
New Jersey
SPC27NJ1
New Jersey State Plane Coordinate System Zone 1
TM
New Mexico
SPC27NM1
New Mexico State Plane Coordinate System Eastern Zone
TM
SPC27NM2
New Mexico State Plane Coordinate System Central Zone
TM
SPC27NM3
New Mexico State Plane Coordinate System Western Zone
TM
SPC27NY1
New York State Plane Coordinate System Eastern Zone
TM
SPC27NY2
New York State Plane Coordinate System Central Zone
TM
SPC27NY3
New York State Plane Coordinate System Western Zone
TM
SPC27NY4
New York State Plane Coordinate System Long Island Zone
L
North Carolina
SPC27NC1
North Carolina State Plane Coordinate System Zone 1
L
North Dakota
SPC27ND1
North Dakota State Plane Coordinate System Northern Zone
L
SPC27ND2
North Dakota State Plane Coordinate System Southern Zone
L
SPC27OH1
Ohio State Plane Coordinate System Northern Zone
L
SPC27OH2
Ohio State Plane Coordinate System Southern Zone
L
SPC27OK1
Oklahoma State Plane Coordinate System Northern Zone
L
SPC27OK2
Oklahoma State Plane Coordinate System Southern Zone
L
SPC27OR1
Oregon State Plane Coordinate System Northern Zone
L
SPC27OR2
Oregon State Plane Coordinate System Southern Zone
L
SPC27PA1
Pennsylvania State Plane Coordinate System Northern Zone
L
SPC27PA2
Pennsylvania State Plane Coordinate System Southern Zone
L
SPC27PR1
Puerto Rico & Virgin Islands State Plane Coordinate System Zone 1
L
SPC27PR2
Puerto Rico & Virgin Islands State Plane Coordinate System Zone 2 (St. Croix)
L
Rhode Island
SPC27RI1
Rhode Island State Plane Coordinate System Zone 1
TM
Samoa
-none-
-not supported-
L
South Carolina
SPC27SC1
South Carolina State Plane Coordinate System Northern Zone
L
SPC27SC2
South Carolina State Plane Coordinate System Southern Zone
L
Nebraska
Nevada
New York
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico & Virgin Islands
IDRISI Guide to GIS and Image Processing Volume 1
148
State
File name
Title
Projection
South Dakota
SPC27SD1
South Dakota State Plane Coordinate System Northern Zone
L
SPC27SD2
South Dakota State Plane Coordinate System Southern Zone
L
Tennessee
SPC27TN1
Tennessee State Plane Coordinate System Zone 1
L
Texas
SPC27TX1
Texas State Plane Coordinate System Northern Zone
L
SPC27TX2
Texas State Plane Coordinate System North Central Zone
L
SPC27TX3
Texas State Plane Coordinate System Central Zone
L
SPC27TX4
Texas State Plane Coordinate System South Central Zone
L
SPC27TX5
Texas State Plane Coordinate System Southern Zone
L
SPC27UT1
Utah State Plane Coordinate System Northern Zone
L
SPC27UT2
Utah State Plane Coordinate System Central Zone
L
SPC27UT3
Utah State Plane Coordinate System Southern Zone
L
Vermont
SPC27VT1
Vermont State Plane Coordinate System Zone 1
TM
Virginia
SPC27VA1
Virginia State Plane Coordinate System Northern Zone
L
SPC27VA2
Virginia State Plane Coordinate System Southern Zone
L
SPC27WA1
Washington State Plane Coordinate System Northern Zone
L
SPC27WA2
Washington State Plane Coordinate System Southern Zone
L
SPC27WV1
West Virginia State Plane Coordinate System Northern Zone
L
SPC27WV2
West Virginia State Plane Coordinate System Southern Zone
L
SPC27WI1
Wisconsin State Plane Coordinate System Northern Zone
L
SPC27WI2
Wisconsin State Plane Coordinate System Central Zone
L
SPC27WI3
Wisconsin State Plane Coordinate System Southern Zone
L
SPC27WY1
Wyoming State Plane Coordinate System Eastern Zone
TM
SPC27WY2
Wyoming State Plane Coordinate System East Central Zone
TM
SPC27WY3
Wyoming State Plane Coordinate System West Central Zone
TM
SPC27WY4
Wyoming State Plane Coordinate System Western Zone
TM
Utah
Washington
West Virginia
Wisconsin
Wyoming
US State Plane Coordinate System 1983
REF files are supplied for all US State Plane Coordinate Systems based on the Transverse Mercator and Lambert Conformal Conic projections for NAD27 and NAD83. These are located in the GEOREF subdirectory of your CartaLinx program directory. The following table lists these files for the NAD83 datum. The Proj column indicates the projection upon
which the system is based (L=Lambert Conformal Conic / TM=Transverse Mercator). Note that Samoa did not make
the change from NAD27 to NAD83.
Appendix 3: Supplied Reference System Parameter Files
149
State
File name
Title
Proj
Alabama
SPC83AL1
Alabama State Plane Coordinate System Eastern Zone
TM
SPC83AL2
Alabama State Plane Coordinate Western Zone
TM
SPC83AK0
Alaska State Plane Coordinate System Zone 10
TM
SPC83AK2
Alaska State Plane Coordinate System Zone 2
TM
SPC83AK3
Alaska State Plane Coordinate System Zone 3
TM
SPC83AK4
Alaska State Plane Coordinate System Zone 4
TM
SPC83AK5
Alaska State Plane Coordinate System Zone 5
TM
SPC83AK6
Alaska State Plane Coordinate System Zone 6
TM
SPC83AK7
Alaska State Plane Coordinate System Zone 7
TM
SPC83AK8
Alaska State Plane Coordinate System Zone 8
TM
SPC83AK9
Alaska State Plane Coordinate System Zone 9
TM
SPC83AZ1
Arizona State Plane Coordinate System Eastern Zone
TM
SPC83AZ2
Arizona State Plane Coordinate System Central Zone
TM
SPC83AZ3
Arizona State Plane Coordinate System Western Zone
TM
SPC83AR1
Arkansas State Plane Coordinate System Northern Zone
L
SPC83AR2
Arkansas State Plane Coordinate System Southern Zone
L
SPC83CA1
California State Plane Coordinate System Zone I
L
SPC83CA2
California State Plane Coordinate System Zone II
L
SPC83CA3
California State Plane Coordinate System Zone III
L
SPC83CA4
California State Plane Coordinate System Zone IV
L
SPC83CA5
California State Plane Coordinate System Zone V
L
SPC83CA6
California State Plane Coordinate System Zone VI
L
SPC83CO1
Colorado State Plane Coordinate System Northern Zone
L
SPC83CO2
Colorado State Plane Coordinate System Central Zone
L
SPC83CO3
Colorado State Plane Coordinate System Southern Zone
L
Connecticut
SPC83CT1
Connecticut State Plane Coordinate System Zone 1
L
Delaware
SPC83DE1
Delaware State Plane Coordinate System Zone 1
TM
Florida
SPC83FL1
Florida State Plane Coordinate System Eastern Zone
TM
SPC83FL2
Florida State Plane Coordinate System Western Zone
TM
SPC83FL3
Florida State Plane Coordinate System Northern Zone
TM
SPC83GA1
Georgia State Plane Coordinate System Eastern Zone
TM
SPC83GA2
Georgia State Plane Coordinate System Western Zone
TM
Alaska
Arizona
Arkansas
California
Colorado
Georgia
IDRISI Guide to GIS and Image Processing Volume 1
150
State
File name
Title
Proj
Hawaii
SPC83HI1
Hawaii State Plane Coordinate System Zone 1
TM
SPC83HI2
Hawaii State Plane Coordinate System Zone 2
TM
SPC83HI3
Hawaii State Plane Coordinate System Zone 3
TM
SPC83HI4
Hawaii State Plane Coordinate System Zone 4
TM
SPC83HI5
Hawaii State Plane Coordinate System Zone 5
TM
SPC83ID1
Idaho State Plane Coordinate System Eastern Zone
TM
SPC83ID2
Idaho State Plane Coordinate System Central Zone
TM
SPC83ID3
Idaho State Plane Coordinate System Western Zone
TM
SPC83IL1
Illinois State Plane Coordinate System Eastern Zone
TM
SPC83IL2
Illinois State Plane Coordinate System Western Zone
TM
SPC83IN1
Indiana State Plane Coordinate System Eastern Zone
TM
SPC83IN2
Indiana State Plane Coordinate System Western Zone
TM
SPC83IA1
Iowa State Plane Coordinate System Northern Zone
L
SPC83IA2
Iowa State Plane Coordinate System Southern Zone
L
SPC83KS1
Kansas State Plane Coordinate System Northern Zone
L
SPC83KS2
Kansas State Plane Coordinate System Southern Zone
L
SPC83KY1
Kentucky State Plane Coordinate System Northern Zone
L
SPC83KY2
Kentucky State Plane Coordinate System Southern Zone
L
SPC83LA1
Louisiana State Plane Coordinate System Northern Zone
L
SPC83LA2
Louisiana State Plane Coordinate System Southern Zone
L
SPC83LA3
Louisiana State Plane Coordinate System Offshore Zone
L
SPC83ME1
Maine State Plane Coordinate System Eastern Zone
TM
SPC83ME2
Maine State Plane Coordinate System Western Zone
TM
Maryland
SPC83MD1
Maryland State Plane Coordinate System Zone 1
L
Massachusetts
SPC83MA1
Massachusetts State Plane Coordinate System Mainland Zone
L
SPC83MA2
Massachusetts State Plane Coordinate System Island Zone
L
SPC83MI1
current Michigan State Plane Coordinate System Northern Zone
L
SPC83MI2
current Michigan State Plane Coordinate System Central Zone
L
SPC83MI3
current Michigan State Plane Coordinate System Southern Zone
L
SPC83MI4
old Michigan State Plane Coordinate System Eastern Zone
TM
SPC83MI5
old Michigan State Plane Coordinate System Central Zone
TM
SPC83MI6
old Michigan State Plane Coordinate System Western Zone
TM
SPC83MN1
Minnesota State Plane Coordinate System Northern Zone
L
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Michigan
Minnesota
Appendix 3: Supplied Reference System Parameter Files
151
State
File name
Title
Proj
SPC83MN2
Minnesota State Plane Coordinate System Central Zone
L
SPC83MN3
Minnesota State Plane Coordinate System Southern Zone
L
SPC83MS1
Mississippi State Plane Coordinate System Eastern Zone
TM
SPC83MS2
Mississippi State Plane Coordinate System Western Zone
TM
SPC83MO1
Missouri State Plane Coordinate System Eastern Zone
TM
SPC83MO2
Missouri State Plane Coordinate System Central Zone
TM
SPC83MO3
Missouri State Plane Coordinate System Western Zone
TM
Montana
SPC83MT1
Montana State Plane Coordinate System Single Zone
L
Nebraska
SPC83NE1
Nebraska State Plane Coordinate System Single Zone
L
Nevada
SPC83NV1
Nevada State Plane Coordinate System Eastern Zone
TM
SPC83NV2
Nevada State Plane Coordinate System Central Zone
TM
SPC83NV3
Nevada State Plane Coordinate System Western Zone
TM
New Hampshire
SPC83NH1
New Hampshire State Plane Coordinate System Zone 1
TM
New Jersey
SPC83NJ1
New Jersey State Plane Coordinate System Zone 1
TM
New Mexico
SPC83NM1
New Mexico State Plane Coordinate System Eastern Zone
TM
SPC83NM2
New Mexico State Plane Coordinate System Central Zone
TM
SPC83NM3
New Mexico State Plane Coordinate System Western Zone
TM
SPC83NY1
New York State Plane Coordinate System Eastern Zone
TM
SPC83NY2
New York State Plane Coordinate System Central Zone
TM
SPC83NY3
New York State Plane Coordinate System Western Zone
TM
SPC83NY4
New York State Plane Coordinate System Long Island Zone
L
North Carolina
SPC83NC1
North Carolina State Plane Coordinate System Zone 1
L
North Dakota
SPC83ND1
North Dakota State Plane Coordinate System Northern Zone
L
SPC83ND2
North Dakota State Plane Coordinate System Southern Zone
L
SPC83OH1
Ohio State Plane Coordinate System Northern Zone
L
SPC83OH2
Ohio State Plane Coordinate System Southern Zone
L
SPC83OK1
Oklahoma State Plane Coordinate System Northern Zone
L
SPC83OK2
Oklahoma State Plane Coordinate System Southern Zone
L
SPC83OR1
Oregon State Plane Coordinate System Northern Zone
L
SPC83OR2
Oregon State Plane Coordinate System Southern Zone
L
SPC83PA1
Pennsylvania State Plane Coordinate System Northern Zone
L
SPC83PA2
Pennsylvania State Plane Coordinate System Southern Zone
L
SPC83PR1
Puerto Rico & Virgin Islands State Plane Coordinate System Zone 1
L
Mississippi
Missouri
New York
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico & Virgin Islands
IDRISI Guide to GIS and Image Processing Volume 1
152
State
File name
Title
Proj
SPC83PR2
Puerto Rico & Virgin Islands State Plane Coordinate System Zone 2 (St. Croix)
L
Rhode Island
SPC83RI1
Rhode Island State Plane Coordinate System Zone 1
TM
Samoa
-none-
-not supported-
L
South Carolina
SPC83SC1
South Carolina State Plane Coordinate System, single zone
L
South Dakota
SPC83SD1
South Dakota State Plane Coordinate System Northern Zone
L
SPC83SD2
South Dakota State Plane Coordinate System Southern Zone
L
Tennessee
SPC83TN1
Tennessee State Plane Coordinate System Zone 1
L
Texas
SPC83TX1
Texas State Plane Coordinate System Northern Zone
L
SPC83TX2
Texas State Plane Coordinate System North Central Zone
L
SPC83TX3
Texas State Plane Coordinate System Central Zone
L
SPC83TX4
Texas State Plane Coordinate System South Central Zone
L
SPC83TX5
Texas State Plane Coordinate System Southern Zone
L
SPC83UT1
Utah State Plane Coordinate System Northern Zone
L
SPC83UT2
Utah State Plane Coordinate System Central Zone
L
SPC83UT3
Utah State Plane Coordinate System Southern Zone
L
Vermont
SPC83VT1
Vermont State Plane Coordinate System Zone 1
TM
Virginia
SPC83VA1
Virginia State Plane Coordinate System Northern Zone
L
SPC83VA2
Virginia State Plane Coordinate System Southern Zone
L
SPC83WA1
Washington State Plane Coordinate System Northern Zone
L
SPC83WA2
Washington State Plane Coordinate System Southern Zone
L
SPC83WV1
West Virginia State Plane Coordinate System Northern Zone
L
SPC83WV2
West Virginia State Plane Coordinate System Southern Zone
L
SPC83WI1
Wisconsin State Plane Coordinate System Northern Zone
L
SPC83WI2
Wisconsin State Plane Coordinate System Central Zone
L
SPC83WI3
Wisconsin State Plane Coordinate System Southern Zone
L
SPC83WY1
Wyoming State Plane Coordinate System Eastern Zone
TM
SPC83WY2
Wyoming State Plane Coordinate System East Central Zone
TM
SPC83WY3
Wyoming State Plane Coordinate System West Central Zone
TM
SPC83WY4
Wyoming State Plane Coordinate System Western Zone
TM
Utah
Washington
West Virginia
Wisconsin
Wyoming
Gauss-Kruger
The Gauss-Kruger reference system is used primarily in the countries of the former Soviet Union and Eastern Bloc. REF
Appendix 3: Supplied Reference System Parameter Files
153
files are included for zones 1-32. All use the Pulkovo 1942 datum and Krassovsky 1940 ellipsoid. The names of these files
are GK01_P42.ref through GK32_P42.ref. The Gauss-Kruger projection is identical to the Transverse Mercator projection. (Note that the alternate spelling “Gauss-Krueger” is also acceptable in a REF file.)
Miscellaneous
A small set of additional reference system parameter files has been provided, including:
Location
File Name
Title
United States
ALBERSUS
Alber's Equal Area Conic for Conterminous US
United States
ALBERSUS
Alber's Equal Area Conic for Conterminous US
United States
ALBERSUS
Alber's Equal Area Conic for Conterminous US
United States
ALBERSUS
Alber's Equal Area Conic for Conterminous US
United States
LAZEA
USGS Lambert Azimuthal Equal Area
South Africa
CAPE35
South Africa Universal Transverse Mercator Zone 35
Malawi
MALAWI36
Universal Transverse Mercator Zone 36 for Malawi
Africa
CLABSHA
Clark Labs Hammer-Aitoff Grid for Africa
Bangladesh
BANGTM46
Bangladesh Universal Transverse Mercator Zone 46
Thailand
THAITM47
Thailand Universal Transverse Mercator Zone 47
Vietnam
VIETTM48
Vietnam Universal Transverse Mercator Zone 48
IDRISI Guide to GIS and Image Processing Volume 1
154
Index
Atmosphere 18
Atmospheric Windows 18
Attribute Documentation Files 53
Attribute Files 44, 52
AUTOCORR 89
Automatic Display 58
Automatically-Generated Output File Name
Autoscaling 57
AVI Files 66, 80
AVIRIS 22, 27
Symbols
.ADC Files 53
.AVL Files 52
.MAP Files 55
.MDB Files 52
.RDC File 47
.RST Files 44
.SM0, .SM1, .SM2 Files
.SMP Files 55
.SMT Files 55
.VCT Files 49
.VDC Files 51
55
Numerics
3-D Display
80
A
Accuracy 119
Accuracy Assessment 32, 103
Active Sensors 18
Add Layer 60
Advanced Very High Resolution Radiometer (AVHRR)
21, 26, 33
Aerial Photography 23
Aerial Videography 24
Alber’s equal area conic projection 154
ALLOCATE 86
Allocation 13
Analysis Menu 81
Analytical Hillshading 95
API 112
Application Programming Interface 112
Arc/Info Files 79
ARCIDRIS 79
ArcView Files 79
AREA 82
ASCII File Type 46
ASPECT 95
ASSIGN 76, 81, 105
Atlas GIS Files 79
ATLIDRIS 79
Index
39
B
Background 64
Band Sequential 120
Band-Interleaved-by-Line 120
Band-Interleaved-by-Line (BIL) Files 78
Band-Interleaved-by-Pixel (BIP) Files 78
BAYCLASS 31, 101
BAYES 91
Bayesian Probability Theory 101
BELCLASS 31, 101
BELIEF 91, 102
Binary File Type 46
BIPIDRIS 78
BMP Files 64, 79
BMPIDRIS 79
Boolean Layer 11
Bounding Rectangle 11, 52
BRANCH 112
BUFFER 86
Buffer Zone 13
Byte Data Type 45
Byte Order 78
C
CA_MARKOV 93
Calculate 75
CALL 112
CartaLinx 6, 105, 117, 118
CELLATOM 93
Cellular automata 93
CENTER 89
Change / Time Series Submenu 92
Change Analysis 92
Change Resolution 122
Change Rows and Columns 122
Chlorophyll 18, 20, 21, 32
155
Clark Labs 1
Classification (of remotely-sensed images) 21,
CLUSTER 31, 101
Cluster Analysis 31
Collection 42
Collection Linked Zoom 68
Collections 73
Color 19, 20
COLSPACE 33, 98
COM Server 112
Comment Field 49
Comments Field 49
Composer 59, 60, 80
COMPOSITE 80, 98
Composite Generation 29
Computer Assisted Design (CAD) 6
CONCAT 104
Consistency Field 49
Context Operators 14
Context Operators Submenu 87
CONTOUR 96
CONTRACT 104
Contrast 29, 61
CONVERT 103
Coordinate Geometry (COGO) 6
CORRELATE 93
COST 85
Cost Distance 13
COUNT 88, 91
CRATIO 89
CRLF 78
CROSSTAB 81, 89, 92
CTG 79
Cursor Inquiry Mode 67
CURVATURE 95
Curve Fitting 94, 95
27, 30
Database Management System 6
Database Query 11, 14, 73, 82
Database Query by Attribute 11, 82
Database Query by Location 11, 84
Database Query Submenu 81
Database Tables 44, 52
DATABASE WORKSHOP 105
Database Workshop 73, 82
Datum 124, 135
Datum parameters 135
Datum transformation 125, 131
dBASE Files 44
Decision Support 8, 15
Decison Support Submenu 90
DECOMP 86
Delphi 113
DEMIDRIS 79
Dempster-Shafer Theory 101
Derivative Mapping 15, 118
DESTRIPE 97
Dialog Box 37
Digital Image Processing 27
Digitizing 6, 116, 117
DISAGGREGATE 93
DISPERSE 86
DISPLAY Launcher 80
Display Launcher 57
Display Menu 80
Display System 6, 57
DISTANCE 85
Distance Operators 13
Distance Operators Submenu 85
DLG 79
Documentation File 47, 51
Documentation Files 42
DRAWFILT 99
DXF Files 79
DXFIDRIS 79
D
Data Compression 116
Data Entry Menu 105
Data Formats 116
Data Import 115, 119, 120
Data Input 115
Data Integration 121
Data Paths 37, 77
Data Tables 52
Data Type 45, 50, 54
Database 5, 9
Database Development 115
E
EDIT 105
Edit 81
Electromagnetic Energy 17, 18, 19
Ellipsoid parameters 133
EMF Files 65, 79
Enhancement (of remotely-sensed images)
Environ 37
Environment 37
Erdas Files 79
IDRISI Guide to GIS and Image Processing Volume 1
27, 28
156
ERDIDRIS 79
ERRMAT 32, 91, 103
European Remote Sensing Satellite (ERS-1 and ERS-2) 26
EXPAND 104
Export 78
EXTRACT 76, 81, 88
GRASSIDR 79
Grid north 126
Grid referencing system
GROUP 87
Group File 42
Gstat 94
F
H
False Color Composite 21, 23
False origin 126
Feature Extraction 96
Feature Properties 64, 67
File Menu 77
File Names 41
File Type 46, 54
FILTER 87, 98
FILTERFQ 99
Filtering 29
Fixed Length ASCII Tables 44
Flag Definition 49
Flag Value 49
FLIP 78
Focus 60
FOURIER 99
Fourier Analysis 99
FRACTAL 95
FREQDIST 99
Friction 13
FUZCLASS 31, 101
FUZSIG 31, 99
FUZZY 91
Fuzzy Set Theory 101
Hammer-Aitoff grid for Africa 154
Hard Classifiers 30, 100
Hardening (image classification) 30
Heads-up Digitizing 117
Help Menu 106
Help System 40
HILLSHADE 95
HISTO 80, 82, 88
HNTRLAND 88
HYPERABSORB 102, 103
HYPERAUTOSIG 31, 100, 102
HYPERMIN 31, 101, 102
HYPEROSP 31, 102, 103
HYPERSAM 31, 101, 102
HYPERSIG 31, 100
Hyperspectral Remote Sensing 22, 31
HYPERUNMIX 102, 103
HYPERUSP 31, 101
G
Geodesy 123
Geodetic coordinates 125, 145
Geographic Information System (GIS) 5
Geoid 123
Geometric Restoration 28
GEOREF folder 128
Georeferencing 10, 28, 63, 119, 123–132
Geostatistics 95
GeoTIFF data import 78
Global Positioning System 69, 117
GPS 69
Graphic Insets 64
GRASS Files 79
Index
126
I
Icon Bar 35
ID Type 52
IDRISI 1
Idrisi File Conversion 80
IDRISI File Explorer 58
Idrisi File Explorer 77
ILLUMINATE 80
IMAGE CALCULATOR 82, 85
Image Calculator 107, 110
Image Classification 27, 100, 101
Image Enhancement 97
Image Processing Submenu 96
Image Restoration 97
Image Transformation 98
IMAGEDIFF 92
IMAGERATIO 92
Import 78, 115, 119, 120
Indian Space Research Organization (IRS)
INITIAL 104, 105
25
157
Input File 38
Instantaneous Field of View (IFOV) 24
Integer Data Type 45
Interaction Mechanisms 19
Interactive Display Features 67
INTERCON 94
INTERPOL 94
Interpolation 94
ISOCLUST 31, 101
J
Japanese Earth Resource Satellite (JERS) 26
K
Kriging 94
Kriging and Conditional Simulation
95
L
Lambert azimuthal equal area projection 154
LANDSAT 120
Landsat data import 78
LANDSAT satellites and data 21, 25
Latitude and longitude 125, 145
LATLONG REF file 145
Launcher 80
Layer 41
Layer Frame 58
Layer Properties 61
Least-Cost Pathway 13
Legend Categories 49
Legends 63
Lineage Field 49
Linear Spectral Unmixing 31
LINEGEN 104
LINERAS 104
Link File 42
Linked-Table Collections 73
LINTOPNT 94, 104
Logical Operations 11
LOGITREG 89
M
Macro
77
Macro Modeler 107
MAKESIG 99
Map Algebra 12
Map Analysis Package Files 79
Map Composition 6
Map Composition Files 55
Map Grid 63
Map Layer 41
Map Layer Names 41
Map Properties 62
Map Window 58
MAPIDRIS 79
MapInfo Files 79
MARKOV 93
Mathematical Operators Submenu 84
MAXBAY 31, 102
MAXBEL 31, 102
MAXFUZ 31, 102
Maximum Value 48
MAXLIKE 30, 31, 100
MAXSET 31, 101
MCE 90
MDCHOICE 91
Media Viewer 66, 93
Menu 77
Menu System 35
Metadata 42, 77
MIFIDRIS 79
MINDIST 30, 100
Minimum Value 48
MIXCALC 31, 102
Mixed pixels 30
MODIS 22, 27
Modules 77
MOLA 90
Molodensky transform procedure 131
Molodensky transformation constants 131,
Motorolla Processor 119
Move 67
MULTIREG 88
135
N
NADCON procedure 131
NADUSLAT 131
NADUSLON 131
NDVICOMP 97
NMEA Format 69
NORMALIZE 93
Normalized Difference 32
IDRISI Guide to GIS and Image Processing Volume 1
158
North Arrow
63
Project 127, 131
Project File 37
Projection 125
O
On-Screen Digitizing 70
On-screen Digitizing 117
ORTHO 80
Orthographic Display 80
Output File 38
OVERLAY 81, 84
Overlay 7
Overwrite Protection 39
Q
QUADRAT 89
QUERY 82
Query by Attribute 82
Query by Location 84
R
P
Packed Binary File Type 46
Palette Files 55, 57, 80
Palette Files Import 80
PALIDRIS 80
Pan 65, 68
PARE 78
Passive Sensors 17
PATHWAY 86
PATTERN 87, 88
PCA 33, 98
PCLASS 82, 91
Pearson Product Moment Coefficient of Correlation 93
PERIM 82
Philosophy of GIS 16
Photogrammetry 24, 28
Pick List 38
PIPED 30, 100
Placemarks 64, 69
Plane reference system 127
Plate Carrée projection 127
PNTGEN 94, 104
POINTRAS 104
POINTVEC 105
POLYRAS 104
POLYVEC 105
Position Error 48
Preferences 77
Principal Components Analysis 21, 33
Print Composition 65
Priority 60
Process Modelling 15
PROFILE 82, 92
Progress Report 36
PROJECT 103
Index
Radar 18, 26
RADARSAT 79
RADIANCE 97
Radiometric Restoration 26, 27
RANDOM 90, 91
RANK 90
Raster 8, 9
Raster Documentation File 47
Raster Files 44
Raster/Vector Conversion Submenu 104
Real Data Type 45
RECLASS 81
Reference ellipsoid 123, 124, 133
Reference surface 123
Reference System 121
Reference system 123
Reference system parameter file (REF) 127, 128,
Reference System Parameter Files 55
Reflectance 19, 20
Reformat Menu 103
REGRESS 88
RELOCATE 86
Remote Sensing 7, 17, 22
Remove Layer 60
Representative Fraction 68
Representative Fraction (RF) 36
RESAMPLE 28, 97, 103
Resize 67
Resolution 48, 118, 122
Resource Folder 37
Restoration (of remotely-sensed images) 27
Restore Original Window 68
RESULTANT 86
RF 36
RGB24 Data Type 46
RGB8 Data Type 46
145
159
Run Macro 77
Symbol Workshop 65
System Operation 35
System Pour L'Observation de la Terre (SPOT) 25,
120
S
SAMPLE 32, 90, 91, 103
Sample Variogram Modeling 94, 95
Satellite Imagery 120
Satellite Platforms 24
Saturation Points 61
Save Composition 64
SCALAR 85
Scale Bar 63
Scale of Display 36
Scanner 6, 117
SCATTER 100
Screen Dump 64
SDTS data import 79
SEPARATE 80
SEPSIG 100
SHAPEIDR 79
SIGCOMP 100
Signature (in Remote Sensing) 19, 20,
Signature Development 99
SLOPE 95
Soft Classifiers 30, 101
Spatial Resolution 24
SPDIST 85
Spectral Resolution 24
Spectral Response Pattern 19, 20, 30
SPOT data import 78
Spreadsheet Files 78
SQL 73, 75
SRFIDRIS 79
SSTIDRIS 78
STANDARD 90, 91
State plane coordinate system 145
Statistical Analysis 7
Statistics Submenu 88
Status Bar 36
STCHOICE 93
STRETCH 80, 97
Sub-Pixel Classification 30
Supervised Classification 30, 99
SURFACE 87
Surface Analysis Submenu 94
Surface Interpolation 94
Surfer Files 79
Surrogate Data 118
Surveying 117
Symbol Files 55, 57, 80
T
22, 30
TASSCAP 33, 99
Terminate 36
Text Inset 64
TEXTURE 33, 87, 99
Thematic Mapper 120
THERMAL 33, 99
THIESSEN 86, 95
TIFF Files 79
TIFIDRIS 79
Time Series Analysis 33, 92
TIN 94
TINPREP 94
TINSURF 94
Titles 64
Topographic Variables 95
TOPOSHAPE 96
Training Sites 30
TRANSFORM 85
Transformation (of remotely-sensed images)
TRANSPOSE 104
TREND 89, 95
True origin 126
TSA 33, 92
27, 32
U
Universal Transverse Mercator (UTM) 145
UNMIX 31, 101
Unsupervised Classification 30, 31
UPDATE 105
US State plane coordinate system 145
User Preferences 77
UTMREF 105
V
Value Error 49
Value Units 49
Values File 76
Values Files 44,
VAR2FIX 78
VARCOST 86
IDRISI Guide to GIS and Image Processing Volume 1
52
160
Vector 8, 9
Vector Documentation File
Vector Files 49
Vector Layer Type 60
Vegetation Indices 32
VEGINDEX 33, 99
Video 66
VIEWSHED 87
Visibility of Layers 60
Visual Basic 113
Visual C++ 113
W
WATRSHED 87
WAV Files 80
Index
51
Waypoints 69
WEIGHT 90
WINDOW 104
Window List 106
WMF Files 65, 79
Working Folder 37
X
XYZIDRIS
78
Z
ZEROPAD 99
Zoom 65, 68
161