one

one
Food
Abd El-Salam et al., J Food Process Technol 2011, S5
http://dx.doi.org/10.4172/2157-7110.S1-001
Processing & Technology
Research Article
Open Access
Application of Computer Vision Technique on Sorting and Grading of
Fruits and Vegetables
Mahendran R*, Jayashree GC, Alagusundaram K
Assistant professor, Indian Institute of Crop Processing Technology, Ministry of Food Processing Industries, GOI Pudukkottai Road, Thanjavur 613 005, Tamil Nadu, India
Abstract
The paper presents the recent development and application of image analysis and computer vision system
in quality evaluation of products in the field of agriculture. Computer vision is a rapid, consistent and objective
inspection technique, which has expanded into many diverse industries. Its speed and accuracy satisfy everincreasing production and quality requirements, hence aiding in the development of totally automated processes.
The requirements and recent developments of hardware and software for machine vision systems are discussed,
with emphases on monochrome imaging, colour imaging and multispectral imaging for modern grading and sorting
systems. Examples of applications for detection of disease, defects, and contamination on fruits and vegetables are
also given. Future trends of machine vision technology applications are discussed.
Keywords: Computer vision system; Image processing; Sorting;
Grading
Introduction
Coding of pictures for transmission through cables was attempted
way back in 1920 (even before the advent of digital computer), but
serious attempts for digital image processing commenced in late 60s
and early 70s. Importance of computer vision increases due to the
fact that vision is one of the most dominant human senses. The use
of computer vision for the inspection of fruits and vegetables has
increased during recent years. Computer vision technique has been a
subject of research and application for more than four decades. This
technique is used in many engineering fields such as robotics, industrial
image processing, food processing and other fields. Quickness, nondestructive evaluation possibilities, easy procedures for application,
quantum of output per unit time are some advantages favouring
application of computer vision to engineering problems. Application
of computer vision to food processing fields evolved first in 1989 for
grain quality inspection [1].
Huge post harvest losses in handling and processing and the
increased demand for food products of high quality and safety
necessitates the growth of accurate, fast and objective quality
determination of food and agricultural products [2]. Major areas of
application of computer vision technology in food industry include
quality evaluation of food grains, fruits, vegetables and processed foods
such as chips, cheese and pizza. The technique had also found useful
for determination insect infestation in grains and blemishes in fruits
and vegetables.
The method used by the farmers and distributors to sort and
grade agricultural and food products are through traditional quality
inspection and handpicking which is time-consuming, laborious and
less efficient. Manual sorting and grading are based on traditional visual
quality inspection performed by human operators, which is tedious,
time-consuming, slow and non-consistent. Harvesting traditionally
is done by manual sensory observations. The quality attributes often
used for deciding on the harvest maturity are color, appearance,
texture and odor [3]. A first and an important step in the post harvest
chain is sorting and grading of harvested produce. Commercially
human senses are employed to sort or grade. Francis [4] found that
human perception could easily be fooled. It is pertinent to explore the
possibilities of adopting faster systems, which will save time and more
J Food Process Technol
accurate in sorting and grading of agricultural and food products.
One of such reliable method is the automated computer vision system
for sorting and grading. This paper reviews the progress of computer
vision in the agricultural and food processing. A cost effective,
consistent, superior speed and accurate sorting can be achieved using
computer vision technique. The processing and manufacturing sectors
requires automated inspection, as well as grading systems so that the
losses incurred during harvesting, production and marketing can
be minimized. Assisted sorting and grading of agricultural and food
products is accomplished based on appearance, texture, colour, shape
and sizes.
Computer Vision System
Computer Vision (CV) is the process of applying a range of
technologies and methods to provide imaging-based automatic
inspection, process control and robot guidance in industrial
applications. While the scope of CV is broad and a comprehensive
definition is difficult to distil, a generally accepted definition of
computer vision is ‘the analysis of images to extract data for controlling
a process or activity’ [5]. Computer vision is a novel technology for
acquiring and analyzing an image of a real scene by computers to
control machines or to process it. It includes capturing, processing
and analyzing images to facilitate the objective and non-destructive
assessment of visual quality characteristics in agricultural and food
products [6]. The techniques used in image analysis include image
acquisition, image pre-processing and image interpretation, leading
to quantification and classification of images and objects of interest
within images. Images are acquired with a physical image sensor and
*Corresponding author: Mahendran R, Indian Assistant professor, Indian
Institute of Crop Processing Technology, Ministry of Food Processing Industries,
GOI Pudukkottai Road, Thanjavur 613 005, Tamil Nadu, India, Tel: 04362 228155/226676; Fax: 04362-227971; E-mail: [email protected]
Received June 23, 2012; Accepted August 02, 2012; Published August 07, 2012
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of
Computer Vision Technique on Sorting and Grading of Fruits and Vegetables. J
Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Copyright: © 2012 Mahendran R, et al. This is an open-access article distributed
under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 2 of 7
dedicated computing hardware and software are used to analyze the
images with the objective of performing a predefined visual task.
Size, shape and color analysis
Size, which is the first parameter identified with quality, has been
estimated using machine vision by measuring either projected area
[7,8], perimeter [9] or diameter [10]. Size measurement is important
for determining produce surface area. The shape is one of the important
visual quality parameters of fruits, vegetables, etc. Currently human
sorters are employed to sort fruits based on shape. Shape is a feature,
easily comprehended by human but difficult to quantify or define by
computer. Most of the machine vision shape detection work has been
done on industrial objects, which have definite structure. Agricultural
and biological products are unique in nature and the growing
environment causes various boundary irregularities which influences
their shapes. Image processing offers solution for sorting of fruits
based on their shape. Colour is also an important quality factor that
has been widely studied [11-14]. The colour of an object is determined
by wavelength of light reflected from its surface. In biological materials
the light varies widely as a function of wavelength. These spectral
variations provide a unique key to machine vision and image analysis.
Hardware
The hardware configuration of computer-based Computer vision
systems is relatively standard [2]. Typically, a computer vision system
consists of:
• An illumination device, which illuminates the sample under
test
• A solid-state Charged Coupled Device array camera, to acquire
an image
• A frame-grabber, to perform the A/D (analog-to-digital)
conversion of scan lines into picture elements or pixels digitized
in a N row by M column image
• A personal computer or microprocessor system, to provide
disk storage of images and computational capability with
software and specific application programs
• A high-resolution colour monitor, which aids in visualizing the
images and the effects of various image analysis routines, on
the images.
Software
Some fruits have one colour homogeneously distributed on the skin
surface, which is called primary colour. The averaged surface colour is a
good quality indicator for these fruits. However, other fruits (e.g. some
varieties of peaches, apples, tomatoes) have a secondary colour that can
be used as a good indicator of maturity. In this case, it is not possible to
rely only on the global colour as a quality parameter.
The image analysis is performed by a specific software application
developed at IVIA (Instituto Valenciano de Investigaciones Agrarias)
using the programming language C, C++, MATLAB. The software was
divided into two modules: an application for training the system; and
another to command the acquisition, process the images and provide
the estimated quality parameters of each fruit [16].
Majumdar and Jayas [15], developed a digital image analysis (DIA)
algorithm based on color features to classify individual kernels of Canada
Western Red Spring (CWRS) wheat, Canada Western Amber Durum
(CWAD) wheat, barley, oats, and rye. Eighteen color features (mean,
variance, range of red, green, blue, hue, saturation, and intensity) were
used for the discriminant analysis. Grains from 15 growing regions
(300 kernels per growing region) were used as the training data set and
another five growing regions were used as the test data set. When the
first 10 most significant color features were used in the color model and
tested on an independent data set (the test data set where total number
of kernels used was 10,500; for CWRS wheat, 300 kernels each were
selected for three grades), the classification accuracies of CWRS wheat,
CWAD wheat, barley, oats, and rye were 94.1, 92.3, 93.1, 95.2, and
92.5% respectively. When the model was tested on the training data set
(total number of kernels used was 31,500), the classification accuracies
were 95.7, 94.4, 94.2, 97.6, and 92.5%, respectively, for CWRS wheat,
CWAD wheat, barley, oats, and rye.
Image capturing method
Blasco et al. [16] developed computer vision techniques for
online estimation of the quality of oranges, peaches and apples based
on quality attributes: size, colour, stem location and detection of
external blemishes. The segmentation procedure, based on a Bayesian
discriminant analysis, allowed fruits to be precisely distinguished from
the background. To evaluate the accuracy of colour estimation by the
developed sensor, colour measurements were taken from 22 surface
sectors of several tomatoes. These sectors varied from a red-greenish
colour to red. The ability of the machine vision system to determine fruit
colour was evaluated by comparing several standard colour indices that
are commonly used for different fruits. These indices were calculated
from Hunter Lab co-ordinate values provided by a colorimeter in three
random circular areas (8mm diameter) in each of the selected sectors.
Image acquisition
J Food Process Technol
Pixel: A pixel is a single point picture element in a graphic image.
A digital image, however, is discrete in nature and there is some
point, which cannot be magnified further. This point is called a pixel
of that digital image. Graphics monitors display pictures by dividing
the display screen into thousands (or millions) of pixels, arranged in
rows and columns. The pixels are so close together that they appear
connected. The number of bits used to represent each pixel determines
how many colors or shades of gray can be displayed. For example, in
8-bit color mode, the color monitor uses 8 bits for each pixel, making it
possible to display 2 to the 8th power (256) different colors or shades of
gray. On color monitors, each pixel is actually composed of three dots a red, a blue, and a green dot. Ideally, the three dots should all converge
at the same point, but all monitors have some convergence error that
can make color pixels appear fuzzy. The quality of a display system
largely depends on its resolution, how many pixels it can display, and
how many bits are used to represent each pixel.
A digital image is a numeric representation (normally binary) of
a two dimensional image. Depending on whether or not the image
resolution is fixed, it may be of vector or raster type. Digital images
are electronic snapshots taken of a scene or scanned from documents
such as photographs, manuscripts, printed texts and art work. Image
acquisition is the transfer of the electronic signal from the sensing
device into a numeric form. Image pre-processing refers to the initial
processing of the raw image data for correction of geometric distortions,
removal of noise, grey level correction and correction for blurring [17].
Singh and Delwiche employed a single CCD video camera mounted
above the sample to capture image details for detecting and identifying
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 3 of 7
major peach defects. Three images were taken of each peach to cover the
entire surface, thus showing a need to acquire detailed image features.
Shape was quantified in two dimensional spaces by the eccentricity
of the peach, as the maximum dimension divided by the minimum
dimension of the peach. The technology aims to duplicate the effect of
human vision by electronically perceiving and understanding an image
[18].
CCDs are sensors used in digital cameras and video cameras to
record still and moving images. The CCD captures light and converts
it to digital data that is recorded by the camera. For this reason,
a CCD is often considered the digital version of film and is a major
technology for digital imaging. In a CCD image sensor, reverse-biased
p–n junctions (essentially photodiodes) are used to absorb photons
and produce charges representing sensed pixels; the CCD is used to
read out these charges. Although CCDs are not the only technology
to allow for light detection, CCD image sensors are widely used in
professional, medical, and scientific applications where high-quality
image data is required. Typically the image sensors used in machine
vision are usually based on solid state charged coupled device (CCD)
camera technology with some applications using thermionic tube
(vacuum tube) devices. CCD cameras are either of the array type or
line scan type. Array or area type cameras consist of a matrix of minute
photosensitive elements (photosites) from which the complete image
of the object is obtained based on output proportional to the amount
of incident light. Alternatively, line scan cameras use a single line of
photosites which are repeatedly scanned up to 2000 times per minute
to provide an accurate image of the object as it moves under the sensor
[19].
Images were captured using an image acquisition system for color
digital camera similar to that developed by Papadakis et al. [20] (Figure
1). Samples were illuminated using four fluorescent lamps (length of
60 cm) with a color temperature of 650°C (Philips, Natural Daylight,
18W) and a color rendering index (Ra) close to 95%. The four lamps
were arranged as a square 35 cm above the sample and at an angle of
45 with the sample plane to give a uniform light intensity over the food
sample. A color digital camera (CDC) Power Shot G3 (Canon, Japan)
was located vertically at a distance of 22.5 cm from the sample. The
angle between the camera lens axis and the lighting sources was around
45. Sample illuminators and the CDC were inside a wood box whose
internal walls were painted black to avoid the light and reflection from
the room. The white balance of the camera was set using a standardized
gray color chart from Kodak. Color standards were photographed and
analyzed periodically to ensure that the lighting system and the CDC
were working properly. Images were captured with the mentioned
CDC at its maximum resolution (2272·1704 pixels) and connected to
the USB port of a Pentium IV, 1200 MHz computer. Canon Remote
Capture Software (version 2.7.0) was used for acquiring the images
directly in the computer in TIFF format without compression.
Monochrome imaging
Monochrome imaging requires a single-chip CCD. The resolution
of a CCD image depends on how many pixels are in the CCD arrays.
Depending on the nature of applications, the camera resolution can
range from 480 to 1024 lines or even higher. It is used for the detection of
blemishes and bruises on apples [21-23]. Monochrome machine vision
technology was also used for detecting scars, cracks, and spreading tips
for asparagus [24]. Grading apples with on-line machine vision has
been attempted [22,25]. The major challenges for on-line inspection
are to produce quality images that provide clearly identifiable features
J Food Process Technol
Figure 1: Potato chip image acquisition system.
and to have both efficient hardware and software to process the images
fast enough for on-line implementation.
Color imaging
Color features of fruits and vegetables included mean, variance,
ranges of the red (R), green (G), and blue (B) color primaries and the
derived hue (H), saturation (S), and intensity (I) values. The 256 grey
levels of the R, G, and B values were grouped in bands of 16 and were
called the histogram features. A single chip CCD camera can also be
used for color imaging. This is done by alternating the pixels in the CCD
camera for red, green and blue (RGB) color acquisition in the area array
CCD to simulate the colours seen by the human eye. Color imaging can
also be achieved using three-chip CCD camera systems. Each CCD in
a three-chip camera receives RGB colours to produce nearly true color
images of the objects. This is accomplished using a prism assembly
with band pass filters and a dichroic coating on selected surfaces of the
prisms that separate broad band light into RGB channels. Throop et al.
[26] used a color difference between bruised and non-bruised regions
on ‘Golden Delicious’ apples. Daley et al. [27,28] applied color imaging
techniques to on-line poultry quality grading.
Illumination
Illumination is an important prerequisite of image acquisition for
food quality evaluation. The quality of captured image can be greatly
affected by the lighting condition. A high quality image can help to
reduce the time and complexity of the subsequent image processing
steps, which can decrease the cost of an image processing system.
Different applications may require different illumination strategies.
The performance of the illumination system greatly influences the
quality of image and plays an important role in the overall efficiency
and accuracy of the system. Lighting type, location and colour quality
play important roles in bringing out a clear image of the object.
Lighting arrangements are grouped into front- or back-lighting. Front
lighting serve as illumination focusing on the object for detection of
external surface features of the product while back-lighting is used for
enhancing the background of the object.
Light sources also differ but may include incandescent, fluorescent,
lasers, X-ray tubes and infrared lamps. The choice of lamp affects
quality and image analysis performance [29]. Sarkar [30] found that
by adjustment of the lighting, the appearance of an object can be
radically changed with the feature of interest clarified or blurred
(Figure 2, Arivazhagan et al.) [31]. When radiation from the lighting
system illuminates an object, it is transmitted through, reflected, or
absorbed. These phenomena are referred to as optical properties. The
absorbed light can also be re-emitted (fluorescence), usually at longer
wavelengths. A number of compounds emit fluorescence in the visible
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 4 of 7
region of the spectrum when excited with UV radiation. The optical
properties and fluorescence emission from the object are integrated
functions of the angle and wavelength of the incident light and chemical
and physical composition of the object.
Image processing
Image processing in agricultural applications consist of three steps:
(1) image enhancement, (2) image feature extraction and (3) image
feature classification. Image enhancement is commonly applied to a
digital image to correct problems such as poor contrast or noise. Image
enhancement procedures such as morphological operations, filters, and
pixel-to-pixel operations are generally used to correct inconsistencies
in the acquired images caused by inadequate and/or non-uniform
illumination. Statistical procedures from basic image statistics such as
mean, standard deviation, and variance to more complex measurement
such as principle component analysis can be used to extract features
from digital images.
For the test fruit image, color and texture features are derived as
that of the training phase and compared with corresponding feature
values, stored in the feature library. The classification is done using
the Minimum Distance Criterion. The image from the training set
which has the minimum distance when compared with the test image
says that the test image belongs to the category of that training image.
(Figure 3, Arivazhagan et al.) [31].
Once image features are identified, the next step is feature
classification. Numerical techniques such as neural networks and fuzzy
inference systems have been successfully applied to perform image
feature classification. Pre-processing refers to the initial processing of
the raw image. The images captured or taken are transferred onto a
computer and are converted to digital images. Digital images though
displayed on the screen as pictures, are digits, which are readable
by the computer and are converted to tiny dots or picture elements
representing the real objects. In some cases pre-processing is done to
improve the image quality by suppressing undesired distortions referred
to as “noise” or by the enhancement of important features of interest.
The intermediate-level processing involves image segmentation, image
representation and image description.
Image segmentation
Image segmentation is a process of cutting, adding and feature
analysis of images aimed at dividing an image into regions that have a
strong correlation with objects or areas of interest using the principle of
matrix analysis. Image segmentation is one of the most important steps
in the entire image processing technique, as subsequent extracted data
are highly dependent on the accuracy of this operation. Its main aim
is to divide an image into regions that have a strong correlation with
objects or areas of interest. If objects in image cannot be segmented
correctly, it is difficult for object measurement; classification and
recognition, hence impact interpreting and understanding that
image. Earlier studies proposed the use of a ‘flooding’ algorithm to
segment patch-like defects (russet patch, bruise, and also stalk or calyx
area) [32]. It was found that this method of feature identification is
applicable to other types of produce with uniform skin colour. This
technique was improved by Yang and Marchant [33], who applied a
‘snake’ algorithm to closely surround the defects. Segmentation can
be achieved by three different techniques: thresholding, edge-based
segmentation and region-based segmentation as shown in Figure 4:
(Sonka et al.; Sun) [18,34]. Thresholding is a simple and fast technique
for characterising image regions based on constant reflectivity or light
absorption of their surfaces. Edge-based segmentation relies on edge
detection by edge operators. Edge operators detect discontinuities in
grey level, colour, texture. Region segmentation involves the grouping
together of similar pixels to form regions representing single objects
within the image. Raji et al. [35] developed a programme in FORTRAN
using the principle of edge detection in image analysis to determine the
edge of sliced breads and biscuits (round and rectangular) with a view
to detect defects (breakage).
Computer vision has been used for such tasks as shape classification,
defect detection, quality grading and variety classification. A colour
model developed was used as a standard for comparison with sample
images. The developed algorithm gave satisfactory results with wellcontrasted defects; however two further enhancements following
segmentation were required to improve accuracy. A novel adaptive
spherical transform was developed and applied in a machine vision
defect sorting system [36]. The transform converts a spherical object
image to a planar object image allowing fast feature extraction, giving
the system an inspection capacity of 3000 apples min-1 from the three
cameras, each covering 24 apples in the field of view. A 94% success
rate was achieved for sorting defective apples from good ones for the
600 samples tested [37].
Image analysis
Figure 2: Illumination Difference.
H
S
Statistical
Features
Feature
Database
RCB to
HSV
Conversion
Fruit
Image
(RCB)
V
DWT
Co-occurence
Matrix
Figure 3: Feature Extraction.
J Food Process Technol
Texture
Features
Image analysis is the process of distinguishing the objects
(regions of interest) from the background and producing quantitative
information, which is used in the subsequent control systems for
decision making. Luis Gracia et al. [38] present the flowchart of the
developed application for image analysis as shown in the Figure 5. The
first step is image acquisition, which includes the software filtering
described in subsection “Erode and Dilate process”. Next, the image
is binarized. After that, the opening/closing operations and the erode/
dilate algorithms are applied, hence only relevant blobs (i.e., the flowers/
fruits/vegetables) remain in the image. The “feature extraction” step
obtains the characterization of each detected blobs, i.e. the position of
the centre of the flower/fruit/vegetable and the cutting point that allows
separating the element from its peduncle. Finally, the “movement
action” step sends the characterization information of each flower/
fruit/vegetable to the robotic mechanism responsible for the desired
task: recollection, cutting, packaging, classification, fumigation, etc.
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 5 of 7
Color analysis based on histogram
In image processing and photography, a color histogram is a
representation of the distribution of colors in an image. For digital
images, a color histogram represents the number of pixels that have
colors in each of a fixed list of color ranges that span the image’s color
space, the set of all possible colors. The color histogram can be built for
any kind of color space, although the term is more often used for threedimensional spaces like RGB or HSV. For monochromatic images,
the term intensity histogram may be used instead. For multi-spectral
images, where each pixel is represented by an arbitrary number of
measurements (for example, beyond the three measurements in RGB),
the color histogram is N-dimensional, with N being the number of
measurements taken. Each measurement has its own wavelength range
of the light spectrum, some of which may be outside the visible spectrum.
In case of colour image, in the RGB colour space, every individual
colour component, namely red, green and blue has its histogram. Then,
the percentage composition of every individual colour component,
which a fruit possesses are to be evaluated. Using this percentage
composition the level for a component can be set as a standard in
classifying the apples based on a particular colour orientation. For an
Apple the higher percentage composition of the red component was
assigned the superior grade, the next lower composition the second
grade and likewise the descending grades were assigned. This enabled
the sorting of apples based on the colour as a parameter.
Image interpretation
The methodology was developed previously and is carefully detailed
by Leon et al. [39]. Five models for the conversion from RGB images
to L*a*b* units were developed and tested: linear, quadratic, gamma,
direct, and neural network. In the evaluation of the performance of
those models, the neural network model stands out with an error of
only 0.96%. So it was possible to find in each pixel of the image a L*a*b*
T
(a)
(b)
(c)
Figure 4: Typical segmentation techniques (a) Thresholding, (b) edge-based
segmentation and (c) region based segmentation (Sun, 2000).
start
image (first step)
acquisition
periodic loop (T=66ms)
filtered image
binarize
image
open
close
(erode, dilate, ...)
only relevant blobs in the image
feature
extraction
blob
treatment
movement (e.g. pan/tilt unit or
actions
cartesian command)
Figure 5: Flowchart of the of the developed application.
J Food Process Technol
color measuring system that is appropriate for an accurate, exact and
detailed characterization of a food item, thus improving quality control
and providing a highly useful tool for the food industry based on a
color digital camera.
Sorting
Automatic fruit sorting can improve the quality of the product,
abolish inconsistent manual evaluation, and reduce dependence
on available manpower. Quality sorting is based on a multitude of
measures, flavor (sweetness, acidity), appearance (color, size, shape,
blemishes, glossiness), and texture (firmness, mouthfeel). Edan [40]
indicated that multi-sensor quality classification can be applied in real
sorting and improves the overall classification. Quality classification of
tomatoes was successfully applied using vision and an impact sensor
[41]. To sort according to the surface defects requires an analysis
as complete as possible of the entire fruit surface. The company
working with the CEMAGREF Institute in France, in this research has
produced for its automatic colour sorting system (already marketed) a
satisfactory conveyor belt. This conveyor belt, built with rotary rubber
rollers, carries the fruit along while they rotate. This rotation enables
the colour detection system or a human observer to look at most of the
surface of the fruit. But, with this system, it is not possible to inspect
the poles. The shape of the rubber rollers (bicones) enables the fruit to
be aligned in three to eight lines across the conveyor belt as shown in
Figure 6 and 7. The company had asked CEMAGREF to integrate the
new processing system for blemish detection into it. The estimation of
the surface defects, unlike the colour sorting, requires a very precise
analysis of the fruit. Even a small damaged area can downgrade the
fruit. The system must be able to analyse elementary surfaces of a few
square millimetres. To do this, cameras were used as sensors. The
light colour of the Golden Delicious apples allows examination by a
monochromatic camera with a suitable filter.
Delwiche et al. [11] developed a multi-camera, multi-processor
system to sort prunes for surface defects in real time. Three line-scan
cameras were used to view the prune, thus showing again the need
for acquiring three-dimensional images. With a combined sample of
prunes containing 28% defective fruit, 6% for good prunes and 8% for
defective prunes that utilized an image acquisition system with two
windows in the sides of the vision chamber allowing charge coupled
device (CCD) video camera mounting. The relationship between
object shape and its boundary spectrum values in Fourier domain was
explored for shape extraction. A shape separator based on harmonics
of the Fourier transform was designed in a two-dimensional space for
potato shape separation.
Conclusion
The paper presents the recent developments in computer vision
system in the field of agricultural and food products. The adoption of
this emerging technology in sorting and grading of fruits and vegetables
will be of immense benefit to this country. Some of the other associated
benefits include more efficient operation, production of more consistent
product quality, greater product stability and safety. Computer vision
systems have been used increasingly in industry for inspection and
evaluation purposes as they can provide rapid, economic, hygienic,
consistent and objective assessment. However, difficulties still exist,
evident from the relatively slow commercial uptake of computer vision
technology in all sectors. Even though adequately efficient and accurate
algorithms have been produced, processing speeds still fail to meet
modern manufacturing requirements. With few exceptions, research
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 6 of 7
CCD comera
Looding
Analysed truit
Figure 6: Lateral view of the conveyor belt.
Bicones
3 to 8
groding
lines
Running
direction
Bicones
Bicones
33 cm
Figure 7: Plan view of the conveyor.
in this field has dealt with trials on a laboratory scales and hence it
needs more focused and detailed study.
References
1. Zayas I, Pomeranz Y, Lai FS (1989) Discrimination of wheat and non wheat
components in grain samples by image analysis. Cereal Chemistry 66: 233237.
2. Narendra VG, Hareesh KS (2010) Prospects of computer vision automated
grading and sorting systems in agricultural and food products for quality
evaluation. Int J Comp App 1: 0975 – 8887.
3. Sun DW, Brosnan T (2003) Improving quality inspection of food products by
computer vision: a review. J Food Engineering 61: 3-16.
4. Francis FJ (1980) Colour quality evaluation of horticultural crops. HortScience
15: 14-15.
5. Relf Christopher G (2004) Image Acquisition and Processing with LabVIEW,
Taylor & Francis, pp. 1-264.
6. Timmermans AJM (1998) Computer vision system for online sorting of pot
plants using an artificial neural network classifier. Acta Horticulturae 421: 9198.
7. Tao Y, Morrow CT, Heinemann PH, Sommer JH (1990) Automated machine
vision inspection of potatoes. ASAE 90-3531.
8. Varghese Z, Morrow CT, Heinemann PH, Sommer HJ, Tao Y, et al. (1991)
Automated inspection of Golden Delicious apples using colour computer vision.
ASAE 91-7002.
12.Singh N, Delwiche MJ, Johnson RS (1993) Image analysis methods for realtime color grading of stonefruit. Computers and Electronics in Agriculture 9:
71-84.
13.Hahn F (2002) Multispectral prediction of unripe tomatoes. Biosystems
Engineering 81: 147-155.
14.Dobrzanski B, Rybczynski R (2002) Colour change of apple as a result of
storage, shelf-life, and bruising. International Agrophysics 16: 261-268.
15.Majumdar S, Jayas DS (2000) Classification of cereal grains using machine
vision: II Color models. Transactions of the ASAE 43: 1677-1680.
16.Blasco J, Aleixos N, Molto E (2003) Machine vision system for automatic quality
grading of fruit. Biosystems Engineering 85: 415-423.
17.Shirai Y (1987) Three dimensional computer vision. Berlin: Springer-Verlag,
pp. 1-297.
18.Sonka M, Hlavac V, Boyle R (1999) Image processing, analysis, and machine
vision. California, USA: PWS Publishing.
19.Wallin P, Haycock P (1998) Foreign body prevention, detection and control: A
Practical Approach. 4: 359.
20.Papadakis SE, Abdul-Malek S, Kamdem RE, Yam KL (2000) A versatile and
inexpensive technique for measuring color of foods. Food Technology 5: 48-51.
21.Davenel A, Guizard C, Labarre T, Sevila F (1988) Automatic detection of
surface defects on fruit using a vision system. J Agric Eng Res 41: 1-9.
22.Rehkugler GE, Throop JA (1989) Image processing algorithm for apple defect
detection. Trans ASABE 32: 267-272.
9. Sarkar N, Wolfe RR (1985) Feature extraction techniques for sorting tomatoes
by computer vision. Transactions of the ASABE 28: 970-974.
23.Singh N, Delwiche MJ (1994) Machine vision methods for defect sorting stone
fruit. Transactions ASAE 37: 1989-1997.
10.Brodie JR, Hansen AC, Reid JF (1994) Size assessment of stacked logs via the
hough transform. Transactions of the ASAE 37: 303-310.
24.Rigney MP, Brusewitz GH, Stone ML (1996) Peach physical characteristics for
orientation. Trans ASAE 39: 1493-1497.
11.Singh N, Delwiche MJ, Johnson RS, Thompson J (1992) Peach maturity
grading with color computer vision. ASAE 92-3029.
25.Throop JA, Aneshansley DJ, Upchurch BL (1995) An image processing
algorithm to find new and old bruises. Appl Eng Agric 11: 751-757.
J Food Process Technol
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Citation: Mahendran R, Jayashree GC, Alagusundaram K (2012) Application of Computer Vision Technique on Sorting and Grading of Fruits and
Vegetables. J Food Process Technol S1-001. doi:10.4172/2157-7110.S1-001
Page 7 of 7
26.Throop JA, Aneshansley DJ, Upchurch BL (1993) Near-IR and color imaging
for bruise detection on Golden Delicious apples. Proc SPIE 1836: 33-44.
27.Daley W, Carey R, Thompson C (1993) Poultry grading/inspection using color
imaging. Proc SPIE 1907: 124-132.
28.Daley W, Carey R, Thompson C (1995) Real-time color grading and defect
detection of food products. Proc SPIE 2345: 403-411.
29.Bachelor BG (1985) Lighting and viewing techniques in automated visual
inspection. Bedford, UK: IFSPublication Ltd. Computers and Electronics in
Agriculture 20: 117-130.
30.Sarkar NR (1991) Machine vision for quality control in the food industry. In
I Fung & E Matthews (Eds.), Instrumental methods for quality assurance in
foods, pp. 167–187. New York: Marcel Decker.
31.Arivazhagan.S, Newlin Shebiah R, Selva Nidhyanandhan S, Ganesan L (2010)
Fruit recognition using color and texture features. Journal of Emerging Trends
in Computing and Information Sciences 1: 90-94.
32.Yang Q (1994) An approach to apple surface feature detection by machine
vision. Computers and Electronics in Agriculture 11: 249-264.
33.Yang Q, Marchant JA (1995) Accurate blemish detection with active contour
models. Computers and Electronics in Agriculture 14: 77-89.
35.Raji AO, Fagboun AA, Dania MK (2000) An approach to detecting defects in
food products, Proceedings of the First International Conference of the Nigerian
Institution of Agricultural Engineers 36-39.
36.Tao Y, Wen Z (1999) An adaptive spherical image transform for high-speed
fruit defect detection. Transaction of the ASAE 42: 241-246.
37.Tadhg Brosnan, Da-Wen Sun (2002) Inspection and grading of agricultural
and food products by computer vision systems-a review. Computers and
Electronics in Agriculture 36: 193-213.
38.Luis Gracia, Carlos Perez-Vidal, Carlos Gracia (2011) Computer vision
applied to flower, fruit and vegetable processing. World Academy of Science
Engineering and Technology 78: 430-436.
39.Leon K, Mery D, Pedreschi F, Leon J (2006) Color measurements in L*a*b*
units from RGB digtal images. Food Research International 39: 1084-1091.
40.Edan Y, Shmulevich I, Rachmani D, Fallik E, Grinberg S (1994) Neural networks
for quality grading of Tomatoes based on mechanical properties. Proceedings
of the Food Automation III Conference: 346-355.
41.Edan Y, Pasternak H, Shmulevich I, Rachmani D, Guedalia D, et al. (1997)
Color and firmness classification of fresh market tomatoes. Journal of Food
Science 62: 793-796.
34.Sun DW (2000). Inspecting pizza topping percentage and distribution by a
computer vision method. Journal of Food Engineering 44: 245-249.
Submit your next manuscript and get advantages of OMICS
Group submissions
Unique features:
User friendly/feasible website-translation of your paper to 50 world’s leading languages
Audio Version of published paper
Digital articles to share and explore
Special features:
This article was originally published in a special issue, Food Processing
handled by Editor(s). Dr. Tony Jin, USDA’s Eastern Regional Research
Center, USA
J Food Process Technol
200 Open Access Journals
15,000 editorial team
21 days rapid review process
Quality and quick editorial, review and publication processing
Indexing at PubMed (partial), Scopus, DOAJ, EBSCO, Index Copernicus and Google Scholar etc
Sharing Option: Social Networking Enabled
Authors, Reviewers and Editors rewarded with online Scientific Credits
Better discount for your subsequent articles
Submit your manuscript at: http://www.omicsonline.org/submission
Food Processing
ISSN:2157-7110 JFPT, an open access journal
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement