Download - CV Agriculture 2010
-
8/8/2019 CV Agriculture 2010
1/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
1
PROSPECTS OF COMPUTER VISION AUTOMATED
GRADING AND SORTING SYSTEMS IN
AGRICULTURAL AND FOOD PRODUCTS FOR
QUALITY EVALUATION
Narendra V G1 and Hareesh K S2
1Sr. Lecturer Dept. Of CSE, Manipal Institute Of Technology, Manipal, Karnataka, India-576 1042Reader Dept. Of CSE, Manipal Institute Of Technology, Manipal, Karnataka, India-576 104
AbstractThe paper presents the recent development and application of image analysis and computer vision systems in sorting and
grading of products in the field of agricultural and food. Basic concepts and technologies associated with computer vision,a tool used in image analysis and automated sorting and grading is highlighted.
For the ever-increasing population, losses in handling and processing and the increased expectation of food products of
high quality and safety standards, there is need for the growth of accurate, fast and objective quality determination of the
characteristics of food and agricultural food products. Computer vision and image analysis, are non-destructive and cost-effective technique for sorting and grading of agricultural and food products during handling processes and commercial
purposes. Different approaches based on image analysis and processing identified is related to variety of applications in
agricultural and food products.
Keywords: Quality, Sorting and Grading, Automation, Image Processing, Machine Vision;
1. INTRODUCTION
Technological advancement is gradually finding
applications in the agricultural and food products, in
response to one of the greatest challenges i.e. meetingthe need of the growing population. Efforts are being
geared up towards the replacement of human operatorwith automated systems, as human operations are
inconsistent and less efficient. Automation means every
action that is needed to control a process at optimum
efficiency as controlled by a system that operates using
instructions that have been programmed into it or
response to some activities. Automated systems in most
cases are faster and more precise. However, there are
some basic infrastructures that must necessarily be in
place in automation [1].
The technology of image analysis is relatively young and
its origin can be traced back to the 1960s. It hasexperienced tremendous growth both in theory and in
application. It has found application in areas such as
medical diagnostics, automated manufacturing, aerial
surveillance, remote sensing and very recently in thefield of automated sorting and grading of agricultural
and food products. Computer vision is a novel
technology for acquiring and analyzing an image of a
real scene by computers and other devices in order to
obtain information or, to control machines or
processes [47].
In Timmermans [51] opinion, computer vision systemincludes the capturing, processing and analyzing
images to facilitate the objective and non-destructiveassessment of visual quality characteristics in
agricultural and food products. 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.
Harvesting and packing account for the major portion
of the effort and cost incurred by farmers producing
fresh fruits and vegetables to market. However the
processing and manufacturing sectors require product
sorting and grading for commercial and productionpurposes. There is continuous growth in the
development of mechanical harvesting system, and the
need for automated inspection, as well as grading
systems so that the losses incurred during harvesting,production and marketing can be minimized. With
these, the need arises to not only grow and harvest a
quality crop, but also to pack in a consistent and
acceptable manner to gain or to maintain market share
-
8/8/2019 CV Agriculture 2010
2/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
2
as well as prepare materials, which suits processingoperations. These cannot be achieved without sorting
and grading.
Sorting and grading of an agricultural and food products
is accomplished based on appearance, texture, colour,shape and sizes. Manual sorting and grading are based
on traditional visual quality inspection performed by
human operators, which is tedious, time-consuming,
slow and non-consistent. A cost effective, consistent,
superior speed and accurate sorting can be achieved with
machine vision assisted sorting and grading. Automated
sorting and grading had undergone substantial growth in
the field of agricultural and food, in the developed and
developing nations because of availability of
infrastructures. Computer application in agriculture andfood industries have been applied in the areas of sorting,
grading of fresh products, detection of defects such ascracks, dark spots and bruises on fresh fruits and seeds.
The new technologies of image analysis and machine
vision have not been fully explored in the developmentof automated machine in agricultural and food
industries. There is increasing evidence that machine
vision is being adopted at commercial level [22].
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. Sun et al.[48] observed that the basis of quality assessment is
often subjective with attributes such as appearance,
smell, colour, texture and flavor frequently examined by
human inspectors. Francis [13] found that human
perception could easily be fooled. It is pertinent toexplore the possibilities of adopting faster systems,which will save time and more 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.
Advances in computer technology have produced a surgeof interest in image analysis during the last decade and
the potential of this technique for the guidance or control
of agricultural and food processes have been recognized
[37]. Series of studies have been conducted in recent
years to investigate the application of computer vision
technology to sorting and grading of fresh agriculturaland food products. Yam and Spyridon [60] used a simple
digital imaging method for measuring and analyzing
colour of food surfaces and found that the method allowsmeasurements and analysis of the colour of food
surfaces that are adequate for food engineering research.Payne and Shearer [33] developed a machine vision
algorithm for grading of fresh market produce according
to colour and damage while Alchanatis et al [3], used a
neural network based classifier and colour machine
vision instead of the conventional use of black andwhite cameras and geometric features for automatic
classification of tissue culture segments of potato
plantlets. It was found that instead of using only
geometric features combining it with colour
significantly increased the separability of the classes inthe feature space.
Real time detection of defects in fruits using a general
hardware and image processing techniques was
reported by Delwiche and Crowe [9]. Two fruits
(apples and peaches) were tested at the rate of five
fruits/second to evaluate system performance. They
developed an algorithm to acquire and analyze two
combined near infrared (NIR) images of each fruit in
real time with a pipeline image processing system.Their result showed that the system was capable of
executing the sorting algorithm at a rate of 14fruits/second, which was greater than conventional
fruit conveying rates. In this submission, it was
observed that acquiring more than 2 images per fruitand using more than 6 lines of structured illumination
per fruit would reduce the sorting errors slightly in the
case of the apples and greatly improve the system
performance with peaches. Algorithms were
developed by Panigrahi and Misra [32] to measure
fractal-based feature and dimensions (length, width,
area and perimeter) of ear corn images to discriminate
among various shapes. The analysis showed that thecombination of the feature is able to discriminate the
shape differences while Carrion et al [8] described
sorting based on an unsupervised vision system.
Raji [36] developed an algorithm for determining thearea of 2-dimensional objects by image analysis. Thiscan be adapted in detection of leave type as a form of
sorting to select the desired ones during harvesting.
The discrimination of leave from weed using feature
identification by analysis was investigated and
reported by Tsheko [53].
Bull [7] observed the image capture techniques, which
could be or are being used in the field of the
agricultural and food, to generate image that can be
analyzed and used as feedbacks for an automated
systems. The appropriate technique to be adopted by
each industry would depend on the property of thesample to be monitored, the nature of the sample, its
environments and other practical restrictions such as
imaging time. With increased awareness andsophistication on the part of the consumers and the
expectation for improved quality in consumeragricultural and food products which has increased the
need for enhanced quality monitoring in field of
agricultural and food, there is a need to develop
various techniques of image analysis to meet the
-
8/8/2019 CV Agriculture 2010
3/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
3
demand of the growing population. Quality is the sum ofthese attributes that can lead to the production of
products acceptable to the consumer when they are
combined.
In the food industries in developing countries, there isgrowing need for automation due to the fact that the
labour intensive manual processes are not efficient,
accurate and effective. In the developed countries on the
other hand Gunasekaran [14] observed that the food
industry is now ranked among the top 10 industries using
this technology.
This study therefore sets out to review the state and level
of application of image processing and machine vision in
automated sorting and grading with its application andsteps to be taken in improving and developing the
emerging technology beyond its present state toagricultural and food products sorting and grading.
2. TECHNIQUES IN IMAGE PROCESSING
2.1. Equipment and TechniquesA computer vision system consists of two basic
components which are image acquisition: illumination
and image capture device (camera) and image analysis:
an image capture board (frame grabber or digitizer) and
analysis software. A typical laboratory set for image
processing is as shown in Figure 1.
Figure 1: Components of a computer vision system [56] .
2.2. Computer Vision System
The hardware configuration of computer-based
Computer vision systems is relatively standard.
Typically, a computer vision system consists of: An illumination device, which illuminates the sample
under test A solid-state CCD 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 aNrow byMcolumn image
A personal computer or microprocessor system, to
provide disk storage of images and computational
capability with vendor-supplied software and specific
application programs
A high-resolution colour monitor, which aids invisualizing images and the effects of various image
analysis routines.
3. IMAGE ACQUISITION
3.1. IlluminationComputer Vision Systems are affected by the level and
quality of illumination as with the human eye. 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. Illumination systems are the light sources. The
light focuses on the materials (especially when used).
Lighting type, location and colour quality play an
important role in bringing out a clear image of the
object. Lighting arrangements are grouped into front-or back-lighting [15]. Front lighting serve as
illumination focusing on the object for better detectionof external surface features of the product while back-
lighting is used for enhancing the background of the
object. Light sources used include incandescent lamps,fluorescent lamps, lasers, X-ray tubes and infra-red
lamps.
3.2. Image Acquisition
Image capturing devices or sensors are used to view
and generate images of the samples. Some of the
devices or sensors used in generating images include
scanners, ultrasound, X-ray and near infraredspectroscopy. However, in machine vision, image
sensors used are the solid state charged coupled device
(CCD) (i.e. camera) technology with some
applications using thermionic tube devices. Recent
technology has seen the adoption of digital camera,which eliminates the additional component required toconvert images taken by photographic and CCD
cameras or other sensors to readable format by
computer processors. Images captured or taken by
digital camera maintain the features of the images with
little noise due to its variable resolution.
4. IMAGE PRE-PROCESSINGThis 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 areconverted to tiny dots or picture elements representing
the real objects. In some cases preprocessing is done to
improve the image quality by suppressing undesireddistortions referred to as noise or by the
enhancement of important features of interest.
The images or pictures are transformed into computer
digital readable format (i.e. digitized) if a digital
camera did not take them by the image board digitizer.
-
8/8/2019 CV Agriculture 2010
4/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
4
The digitized format is then transferred and used as theinput data by the image processing software to carry out
the necessary processes. Each or a combination of the
digits represent the feature a small portion of the image
called picture element (pixel). Objects are described as
black and white pictures which are represented by digitsranging from 0 to 255 where 0 is black and 255 is white.
Each pixel in coloured pictures is represented by 3 digits
representing RGB [Red, Green, Blue] components with
each being (0 to 255) darkest to lightest RGB. An
arrangement of these digits in row-column format gives
a representation of the image. With this arrangement
using the matrix theory does the analysis in image
processing. Further details can be obtained in Tsheko
[53], Raji [36], Raji et al [37] and any text or reports onimage processing and computer machine vision.
Figure 2: Different levels in the image processing[43].
Image acquisition and image pre-processing arecategorized as low-level processing while the
intermediate and the high level processing stages asclassified by Sun [43] (Figure 2) are further processing
stages required when integrating the preprocessing
stages into an application device such as a sorter and
grader.
The intermediate-level processing involves imagesegmentation, image representation and image
description. 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. Segmentation can be achieved by the following
techniques: thresholding, edge based segmentation andregion based segmentation. Thresholding is used in
characterizing image regions based on constant
reflectivity or light absorption of their surface. This
shows that regions with same features are characterizedand extracted together. Figure 3 shows a thresholding
process where only the dark region is of interest, the
other regions are converted to the background colour in
the threshed image before further processing such as
sending signals to a device to take a feature baseddecision. This process is useful in colour (maturity)
and feature based (defect and damages detection)
sorting.
Original Image Threshold image
Figure 3. Thresholding
Edge based segmentation relies on detection by edge-
to-edge operators, which detect discontinuities in grey
level, the pixel, colour, texture etc. Edge detection is
useful in shape and size sorting. An example of edge
detection result is shown in Figure 4. The applicationof this process was reported by Raji et al [37] who
demonstrated the detection of defects in the shape of
bread and biscuit samples on a processing line.
Figure 4: Edge based segmentation
Region based segmentation involves the groupingtogether and extraction of similar pixels to form
regions representing single objects within the image.
(Figures 5). In this process the other regions are
deleted leaving only the feature of interest.
High level processing deals with recognition and
interpretation, typically using statistical classifiers or
multiplayer neural networks of region of interest.
These steps provide information necessary for the
process or machine control for quality sorting andgrading.
-
8/8/2019 CV Agriculture 2010
5/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
5
Figure 5. Region based segmentationAn interaction of all these levels and knowledge
database are very important and essential for more
precise decision-making and is seen as an integral part of
the image processing process.
These theories when applied to images of products taken
can be used to extract features, which are needed for the
necessary processes. Generally, edge detection to
determine shape and feature extraction to determine
differences in colour is useful in sorting and harvesting.
5. APPLICATIONS IN AGRICULTURAL AND
FOOD PROCESSING OPERATIONS
Computer vision systems are being used increasingly in
the field of agricultural and food products for qualityassurance purpose. The system offers the generation of
precise descriptive data and reduction of tedious humaninvolvement. Computer vision system has proven
successful for the objective, online measurement of
several agricultural and food products with applicationranging from routine inspection to the computer vision
system guided robotic control [48]. Some of the areas
where the techniques have been applied in agricultural
and food processing which need to be developed further
for commercial purposes include.
5.1 Bakery Products
Appearance of baked products is an important qualityattribute which influences the visual perceptions of
customers and hence potential demands of the products.
The appearance of the internal and external features
contributes to the overall impression of the products
quality. Computer vision system has been used tomeasure characteristics such as colour, size and shapewith a view to sorting them to products with same
characteristics before packing.
Raji et al [37] developed a programme in FORTRAN
using the principle of edge detection in image analysis to
determine the edge of sliced breads and biscuits (roundand rectangular) with a view to detecting defects
(breakage). Scott [41] described a system, which
measures the defects in baked loaves of bread, by
analyzing its weight and slope of the top. The internal
structure (crumb grain) of bread and cake was also
examined by machine vision [40]. Dos Mohammed et al[10] also developed a system for the automated visual
inspection of muffins.
Visual features such as colour and size indicate the
quality of many prepared consumer foods. Sun [43]investigated this in research on pizza in which pizza
topping percentage and distribution were extracted from
pizza images. Combining three algorithms used to
segment many different types of pizzas as the traditional
segmentation techniques were found to be inadequatefor this application developed a new segmentation
algorithm. It was found that the new region-based
segmentation technique could effectively group pixels
of the same topping together. As the result, topping
exposure percentage can be easily determined. Thestudy reported that the accuracy of measuring the
topping percentage by the new algorithm reached
90%.
The evaluation of the functional properties of cheese is
assessed to ensure the necessary quality is achieved,
especially for specialized applications such as
consumer food toppings or ingredients. Wang and Sun
[55] developed a computer vision method to evaluate
the melting and browning of cheese. This novel non-contact method was employed to analyze the
characteristics of cheddar and mozzarella cheesesduring cooking and the results showed that the method
provided an objective and easy approach for analyzing
cheese functional properties [56,57]. Ni andGunasekaran [29] developed an image-processing
algorithm to recognize individual cheese shred and
automatically measure the shred length. It was found
that the algorithm recognized shreds well, even when
they were overlapping. It was also reported that the
shred length measurement errors were as low as 0.2%
with a high of 10% in the worst case.
The application of this method is therefore a promising
approach to solving quality control inspection in the
bakery products. This will improve the quality of
baked products, where there are no standard sizes,
shape and texture. The required system though mayadd to the cost of production but the added advantagewill far outweigh the initial cost involved.
5.2. FruitsExternal qualities i.e. sizes, shapes and colour are
considered of paramount importance in the marketing
and sale of fruit. Presence of blemishes influencesconsumer perceptions and therefore determines the
level of acceptability prior to purchase. Computer
vision system has been used for the automated
inspection and grading of fruit to increase product
throughput.
The number of fruits (ripe and unripe) on a tree has
been counted by image analysis prior to harvesting.
This process involved the development of fruitlocation algorithms [24]. Images were taken from a
distance of about 150cm and all the fruit on the treeplaced within the vision field of a camera were
counted. The visible fruits were considered to be those
that the human eye can distinguish on the monitor.
Algorithms based on the red/green relation and on
-
8/8/2019 CV Agriculture 2010
6/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
6
threshold achieved the highest detection percentages oncitrus fruit.
The adoption can also be extended to the peasants
through tractor hiring bodies or by forming cooperatives
to acquire such devices, which only need to be attachedto a tractor. The problem of unplanned fruit trees, which
may make machine movement very difficult, will also
need to be addressed. The high losses incurred presently
on fruits in the market comes as a result of the damages
and bruises as well as potential damage region inflicted
on the fruits falling from the trees due to the method of
harvesting of shaking the trees or the branches. The
adoption can also be extended to the peasants through
tractor hiring bodies or by forming cooperatives to
acquire such devices, which only need to be attached to atractor. The problem of unplanned fruit trees, which may
make machine movement very difficult, will also need tobe addressed. The high losses incurred presently on
fruits in the market comes as a result of the damages and
bruises as well as potential damage region inflicted onthe fruits falling from the trees due to the method of
harvesting of shaking the trees or the branches.
Strawberry appearance and fruit quality are dependent
on a number of pre- and post- harvest factors, hence
variation occurs, necessitating the need for sorting.
Nagata et al. [26] investigated the use of computer vision
to sort fresh strawberries, based on size and shape. Theexperimental results show that the developed system was
able to sort the 600 strawberries tested with an accuracy
of 94-98% into three grades based on shape and five
grades on size. Another automatic strawberry sorting
system was developed by Bato et al. [6]. Average shapeand size accuracies of 98 and 100%, respectively, wereobtained regardless of the fruit orientation angle with
judgement time within 1.18 s.
Prior to export, papayas are subjected to inspection for
the purpose of quality control and grading. For size
grading, the fruit is weighted manually hence thepractice is tedious, time consuming and labour intensive.
Therefore, a computer vision system for papaya size
grading using shape characteristic analysis. The shape
characteristics consisting of area, mean diameter and
perimeter were extracted from the papaya images.
Slamet Riyadi et al. [43] classified according tocombinations of the three features to study the
uniqueness of the extracted features. The proposed
technique showed the ability to perform papaya sizeclassification with more than 94% accuracy.
5.3. VegetablesThe need to be responsive to market demands places a
greater emphasis on quality assessment resulting in the
greater need for improved and more accurate grading
and sorting practices. Computer vision system has
shown to be a viable means of meeting these increasedrequirements for the vegetables field.
Plantlet segments of potato were subcultures for
classification by colour machine vision system by
Alchanatis et al., [3]. In related development, twoalgorithms were developed for analyzing digital binary
images and estimating the location of stem root joints
in processing carrots [5]. Both algorithms were
capable of estimating the stem/root location with a
standard deviation of 5mm. Also Haworth and Searcy
[17] classified carrots on surface defects, curvature
and brokenness. A line scans images with discrete
Fourier transform was developed for the classification
of broccoli heads for assessing its maturity. For the
160 observations from each of three broccoliscultivates, an accuracy of 85% was achieved for
multiple cultivars.
Mushrooms discolouration is undesirable in
mushroom houses and it reduces market value. Thecolour and shape of the cap is the most important
consideration of fresh mushrooms. Felfodi and
Vizhanyo [11] used mushroom images recorded by a
machine vision system to recognize and identify
discolouration caused by bacterial disease. The
method identified all the diseased spot as diseased
and none of the healthy mushrooms parts weredetected as diseased. Reed et al [38] used camera-based technology to select mushroom by size for
picking by a mushroom harvester.
Potatoes have many possible shapes, which need to be
graded for sale into uniform classes for differentmarkets. This created difficulties for shape separation.A Fourier analysis based shape separation method for
grading of potatoes using machine vision for
automated inspection was developed by Tao et al.
[50]. A shape separator based on harmonics of the
transform was defined. Its accuracy of separation was
89% for 120 potato samples, in agreement withmanual grading. Earlier, Lefebvre et al. [19] studied
the use of computer vision for locating the position of
pulp extraction automatically for the purpose of
further analysis on the extracted sample. An image
acquisition system was also constructed for mounting
on a sweet potato harvester for the purpose of yieldand grade monitoring [58]. It was found that culls
were differentiated from saleable sweet potatoes with
classification rates as high as 84%.
Chilli is a variety grown extensively consumed byalmost all the population. It has a high processing
demand and proper sorting is required before filling or
canning. A sorter that, Federico Hahn [12] classifies
chilli by three different width sizes was built. The
-
8/8/2019 CV Agriculture 2010
7/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
7
conveyor used baby suckers to align each chilli duringsensing. Chilli width was determined by means of a
photodiode scanner, which detected the incoming
radiation sent by a laser line generator. Chilies
presenting necrosis were detected with a radiometer and
removed to increase product quality. Horizontal andvertical widths were measured for 200 chilies. The
accuracy on the necrosis detection and width
classification was of 96.3 and 87%, respectively. On-line
necrosis measurements were 85% accurate when only
the relative reflectance at 550nm was used.
Vegetable quality is frequently referred to size, shape,
mass, firmness, colour and bruises from which fruits can
be classified and sorted. However, technological by
small and middle producers implementation to assessthis quality is unfeasible, due to high costs of software,
equipment as well as operational costs. Based on theseconsiderations, the Antonio Carlos Loureiro Lino et al.
[4] research is to evaluate a new open software that
enables the classification system by recognizing fruitshape, volume, colour and possibly bruises at a unique
glance. The software named ImageJ, compatible with
Windows, Linux and MAC/OS, is quite popular in
medical research and practices, and offers algorithms to
obtain the above-mentioned parameters. The software
allows calculation of volume, area, averages, border
detection, image improvement and morphological
operations in a variety of image archive formats.
Some other earlier studies of computer vision associated
with vegetable grading and inspection include colour and
defect sorting of bell peppers [42]. Morrow et al. [25]
presented the techniques of vision inspection ofmushrooms, apples and potatoes for size, shape andcolour. The use of computer vision for the location of
stem/root joint in carrot has also been assessed [5].
Feature extraction and pattern recognition techniques
were developed by Howarth and Searcy [17] to
characterize and classify carrots for forking, surface
defects, curvature and brokenness. The rate ofmisclassification was reported to be below 15% for the
250 samples examined. More recently sweet onions were
line scanned for internal defects using X-ray imaging
[52]. An overall accuracy of 90% was achieved when
spatial and transform features were evaluated for product
classification.
Image analysis and machine vision in general from the
foregoing can be said to offer a fast and reliable processin product sorting and grading, separation and detection
of some other facilities.
5.4. GrainsGrain quality attributes are very important for all users
and especially the milling and baking industries.
Computer vision has been used in grain qualityinspection for many years. An early study by Zayas et
al. [61] used machine vision to identify different
varieties of wheat and to discriminate wheat from non-
wheat components.
In later research Zayas et al. [62] found that wheatclassification methods could be improved by
combining morphometry (computer vision analysis)
and hardness analysis. Hard and soft recognition rates
of 94% were achieved for the seventeen varieties
examined. Twenty-three morphological features were
used for the discriminant analysis of different cereal
grains using machine vision [23]. Classification
accuracies of 98, 91, 97,100 and 91% were recorded
for CWRS (Canada Western Red Spring) wheat,
CWAD (Canada Western Amber Durum) wheat,barley, oats and rye, respectively. 25 kernels per image
were captured from a total of 6000 for each grain typeexamined.
The relationship between colour and texture features
of wheat samples to scab infection rate was studiedusing a neural network method [39]. It was found that
the infection rates estimated by the system followed
the actual ones with a correlation coefficient of 0.97
with human panel assessment and maximum and mean
absolute errors of 5 and 2%, respectively. In this study
machine vision-neural network based technique
proved superior to the human panel. Image analysis
has also been used to classify dockage components forCWRS (Canada Western Red Spring) wheat and other
cereals [27]. Morphology, colour and morphology-
colour models were evaluated for classifying the
dockage components. Mean accuracies of 89 and 96%
for the morphology model and 71 and 75% for thecolour model were achieved when tested on the testand training data sets, respectively. Overall 6000
kernels for each grain type were analyzed. Machine
vision was used to identify weeds commonly found in
wheat fields in experimentation by Zhang and
Chaisattapagon [63]. Five shape parameters were used
in leaf shape studies and were found effective indistinguishing broadleaf weed species such as
pigweed, thistle and kochia from wheat.
In order to preserve corn quality it is important to
obtain physical properties and assess mechanical
damage so as to design optimum handling and storageequipment. Measurements of kernel length, width and
projected area independent of kernel orientation have
been performed using machine vision [30]. Thealgorithm accuracy was between 0.86 and 0.89
measured by the correlation coefficient betweenpredicted results and actual sieving for a 500 g sample.
The processing time of the size-grading program was
reported as being between 0.66 and 0.74 s per kernel.
Steenhoek and Precetti [46] performed a study to
-
8/8/2019 CV Agriculture 2010
8/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
8
evaluate the concept of two-dimensional image analysisfor classification of maize kernels according to size
category. A total of 320 maize kernels were categorized
into one of 16 size categories based on degree of
roundness and flatness. Classification accuracy of both
machine vision and screen systems was above 96% forround-hole analysis. However, sizing accuracy for
flatness was less than 80%.
Ng et al. [28] developed a machine vision algorithm for
corn kernel mechanical and mould damage
measurement, which demonstrated a standard deviation
less than 5% of the mean value of the 250 grains
examined. They found that this method was more
consistent than other methods available. The automatic
inspection of 600 corn kernels was also performed by Niet al. [31] using machine vision. For whole and broken
kernel identification on-line tests had successfulclassification rates of 91 and 94% for whole and broken
kernels, respectively.
The whiteness of corn has been measured by an on-line
computer vision approach by Liu and Paulsen [20]. For
the 63 samples (50-80 kernels per sample) tested the
technique was found to be easy to perform with a speed
of 3 kernels per s. In other studies Xie and Paulsen [59]
used machine vision to detect and quantify tetrazolium
staining in corn kernels. The tetrazolium-machine vision
algorithm was used to predict heat damage in corn due todrying air temperature and initial moisture content.
As rice is one of the leading food crops of the world its
quality evaluation is of importance to ensure it remains
appealing to consumers. Liu et al. [20] developed adigital image analysis method for measuring the degreeof milling of rice. They compared the method with
conventional chemical analysis and obtained a
coefficient of determination of R2
=/0.9819 for the 680
samples tested. Wan et al. [54] employed three online
classification methods for rice quality inspection:-
namely range selection, neural network and hybridalgorithms. The highest recorded online classification
accuracy was around 91% at a rate of over 1200
kernels/min. The range selection method achieved this
accuracy but required time-consuming and complicated
adjustment. In another study, milled rice from a
laboratory mill and a commercial-scale mill wasevaluated for head rice yield and percentage whole
kernels, using a shaker table and a machine-vision
system called the GrainCheck [21].
6. 3-D techniqueIn general, only 2-dimensional (2D) data are needed for
grading, classification, and analysis of most agricultural
images. However, in many applications 3-dimensional
image analysis maybe needed as information on
structure or added detail is required. A 3-D visiontechnique has been developed to derive a geometric
description from a series of 2-D images [44]. In
practice this technique might be useful for food
inspection. For example, when studying the shape
features of a piece of bakery, it is necessary to take 2-D images vertically and horizontally to obtain its
roundness and thickness, respectively.
Kanali et al. [18] investigated the feasibility of using a
charge simulation method (CSM) algorithm to process
primary image features for three dimensional shape
recognition. The required features were transferred to
a retinamodel identical to the prototype artificial retina
and were compressed using the CSM by computing
output signals at work cells located in the retina. Anoverall classification rate of 94% was obtained when
the prototype artificial retina discriminated betweendistinct shapes of oranges for the 100 data sets tested.
Gunasekaran and Ding [16] obtained 3-D images of fat
globules in cheddar cheese from 2-D images. Thisenabled the in situ 3-D evaluation of fat globule
characteristics so as the process parameters and fat
levels may be changed to achieve the required textural
qualities.
7. PROBLEMS OF AUTOMATED SORTING
AND GRADING
The major problem of automated sorting and gradingis one of socio-economic effects, which tend to reduce
employment when the number of operators required in
the processing line is reduced. It is not suitable in
processes where manual skill is necessary or
economically more attractive. It requires higher initialand maintenance cost and there may be the need for aprecise understanding of the process for programming
to achieve the required product quality Equipment
associated problem involves radiance on accurate
sensors to precisely measure process condition and the
increased risk, delays and cost if the automatic system
fails. The farm layout and very low production levelwhich may make it uneconomically viable has also
been highlighted.
8. PROSPECTS OF AUTOMATED SORTING
AND GRADING
The adoption of this emerging technology by firstputting more effort into researches on the appropriate
methods and ways of application will be of immense
benefit to this country. Some of the other associatedbenefits include increased production rates (e.g.
through optimization of equipment utilization), moreefficient operation, production of more consistent
product quality, greater product stability and safety.
-
8/8/2019 CV Agriculture 2010
9/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
9
With the above in mind, the fruit and vegetableprocessing and marketing industries stand to gain from
this emerging technology. This is because the losses
incurred during the harvesting season on fruit are
enormous. This results from large number of products to
be handled and sold at the same time. These productsconsist of ripe and unripe fruits. The introduction of
automated sorting and grading will encourage the sorting
of the unripe (which can be kept for a relatively longer
period) from the ripe, which are to be sold immediately.
Presently the practice at the fruit market in to sell
baskets of fruits containing both ripe and unripe. These
products are found to get spoilt before they get to the
final destination.
One of the major hindrances to the introduction of thistechnique is the level of production, which remains at
the peasant level. Most of the products found in themarkets are owned by a number of individuals with each
controlling not more than 4 to 5 baskets, which will be
too small for the adoption of an automated technique.However, encouraging wholesalers and retailers for
groups and cooperatives in marketing as practiced in the
village level for maize shelling will be a way out.
The adaptation of computer vision system for quality
evaluation of processed foods is the area for the greatest
potential uptake of this technology, as analysis can be
based on a standard requirement in already automated
controlled conditions. More complex systems are neededfor the automated grading and sorting of fresh produce
because of the greater range in variability of quality and
also as produce orientation may influence results. With
the idea of precision and more environmental friendly
agriculture becoming more realistic the potential forcomputer vision system in this area is immense with theneed in field crop monitoring, assessment and guidance
systems. However, techniques such as 3D and colour
vision will ensure computer vision development
continues to meet the accuracy and quality requirements
needed in this highly competitive and changing industry.
9.CONCLUSIONSThe paper presents the recent developments in computer
vision system in the field of agricultural and food
products. 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 allsectors. Even though adequately efficient and accurate
algorithms have been produced, processing speeds stillfail to meet modern manufacturing requirements. With
few exceptions, research in this field has dealt with trials
on a laboratory scales thus the area of mechatronics has
been neglected, and hence it needs more focused anddetailed study.
REFERENCES[1].A. Raji and A. Alamutu, Prospects of Computer
Vision Automated Sorting Systems in AgriculturalProcess Operations in Nigeria, Agricultural
Engineering International: the CIGR Journal of
Scientific Research and Development, Vol. VII,
Invited Overview, February 2005.
[2].Ahmada, U., Kondo, N., Montaa, M., and H.
Muraseh, Machine Vision Based Quality Evaluation
of Iyokan Orange Fruit using Neural Networks,
Computers and Electronics in Agriculture, 29(1-2),
pp.135-147, 2000.
[3].Alchanatis V., Peleg, K., and M. Ziv,Classification of Tissue Culture Segments by Colour
Machine Vision,Journal of Agricultural EngineeringResearch, 55, pp. 299-311,1993.
[4].Antonio Carlos Loureiro Lino; Juliana Sanches;
Inacio Maria Dal Fabbro, Image processingtechniques for lemons and tomatoes classification,
Journal of Food and Engineering, 83(4), pp. 433
440, 2008.
[5].Batchelor, M.M., and S.W. Searcy, Computer
Vision Determination of Stem/Root Joint on
Processing Carrots, Journal of Agricultural
Engineering Research, 43, pp. 259-269, 1989.
[6].Bato, P.M., Nagata, M., Cao, Q.X., Hiyoshi, K.,Kitahara, T., Study on sorting system for strawberry
using machine vision (part 2): development of sorting
system with direction and judgement functions forstrawberry (Akihime variety), Journal of the
Japanese Society of Agricultural Machinery, 62(2),pp. 101-110, 2000.[7].Bull, C.R., A Review of Sensing Techniques
which could be used to Generate Images of
Agricultural and Food Materials, Computers andElectronics in Agriculture, 8, pp. 1-29, 1992.
[8].Carrion, J.A., Torregrosa, E.O., and E. Molto,First Result of an Automatic Citrus Sorting MachineBased on an Unsupervised Vision System,
Proceedings of the InternationalConference on
Agricultural Engineering, Oslo, August 1998.[9].Delwiche, M.J., and T.G. Crowe, Real-Time
Defect in Food-Part II: An Algorithm and
Performance of a Prototype System, Transactions ofthe ASAE, 39(6), pp. 2309-2317, 1996.
[10].Dos Mohammed, A.M., Abdullah, Z.M., and A.S.Aziz, Quality Inspection of Bakery Products using a
Colour-Based Machine Vision System, Journal of
Food Quality, 23(1), pp. 39-50, 2000.[11].Felfodi, J., and T. Vizhanyo, Enhancing Colour
Differences in Images of Diseased Mushrooms,
Computers and Electronics in Agriculture, 26, pp.187-
198, 2000.
-
8/8/2019 CV Agriculture 2010
10/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
10
[12].Federico Hahn,Automatic Jalapeno Chilli Gradingby Width,Biosystems Engineering, 83(4), pp. 433440,
2002.
[13].Francis, F.J., Colour Quality Evaluation of
Horticultural crops, HortScience, 15(1), pp. 14-15,
1980.[14].Gunasekaran, S., Computer Vision Technology for
Food Quality Assurance, Trends in Food Science and
Technolog, 7(8), pp. 245-256, 1996.
[15].Gunasekaran, S., Non-Destructive Food
Evaluation Techniques to Analyze Properties andQuality, Food Science and Technology Vol. 105,
Marcel Decker, New York, 2001.
[16].Gunasekaran, S., Ding, K., Three-dimensionalcharacteristics of fat globules in cheddar cheese,
Journal of Dairy Science 82, pp.1890-1896, 1999.[17].Howarth, M.S., and S.W. Searcy, Inspection of
Fresh Carrots by Machine Vision, In Food Processing Automation II Proceedings of the 1992 ASAE
Conference, St. Joseph, Michigan, USA, 1992.
[18].Kanali, C., Murase, H., Honami, N., Three-dimensional shape recognition using a chargesimulation
method to process image features, Journal of
Agricultural Engineering Research, 70, pp. 195-208,
1998.
[19].Lefebvre, M., Gil, S., Brunet, D., Natonek, E., Baur,
C., Gugeril, P., Pun, T., Computer vision and
agricultural robotics for disease control: the potatooperation, Computers and Electronics in Agriculture 9,pp. 85-102, 1993.[20].Liu, J., Paulsen, M.R., Corn whiteness
measurement and classification using machine vision,
In:1997 ASAE Annual International Meeting, Technical
Papers, Paper No. 973045, ASAE, 2950 Niles Road, St.Joseph, MI 49085-9659, USA, 1997.[21].Lloyd, B.J., Cnossen, A.G., Siebenmorgen, T.J.,
Evaluation of two methods for separating head rice
from brokens for head rice yield determination, In:2000 ASAE Annual International Meeting, Paper No.
006074 ASAE, 2950 Niles Road, St. Joseph, MI 49085-
9659, USA, 2000.[22].Locht, P., Thomas, K., and P. Mikkelsen, Full
Colour Image Analysis as a Tool for Quality Control and
Process Development in the Food Industry, In 1997
ASAE AnnualInternational Meeting, Paper No 973006,
St. Joseph, Michigan, USA,1997.
[23].Majumdar, S., Jayas, D.S., Bulley, N.R.,Classification of cereal grains using machine vision,
part 1: Morphological features,In: 1997 ASAE Annual
International Meeting Technical Papers, Paper No.973101, ASAE, 2950 Niles Road, St. Joseph, MI 49085-
9659, USA, 1997.[24].Molto, E., Pla, F., and F. Juste, Vision Systems for
the Location of Citrus Fruit in a Tree Canopy,Journal
of Agricultural Engineering Research, 52, pp. 101-110,
1992.
[25].Morrow, C.T., Heinemann, P.H., Sommer, H.J.,Tao, Y., Varghese, Z., Automate inspection of
potatoes, apples, and mushrooms, In Proceedings of
the International Advanced Robotics Programme,
Avignon, France, pp. 179-188, 1990.
[26].Nagata, M., Cao, Q., Bato, P.M., Shrestha, B.P.,Kinoshita, O., Basic study on strawberry sorting
system in Japan,In: 1997 ASAE Annual International
Meeting Technical Papers, Paper No. 973095, ASAE,
2950 Niles Road, St. Joseph, Michigan 49085-9659,
USA,1997.[27].Nair, M., Jayas, D.S., Bulley, N.R., Dockage
identification in wheat using machine vision, 1997
ASAE Annual International Meeting Technical Papers,
Paper No. 973043, ASAE, 2950 Niles Road, St.
Joseph, Michigan 49085-9659, USA,1997.[28].Ng, H.F., Wilcke, W.F., Morey, R.V., Lang, J.P.,
Machine vision evaluation of corn kernel mechanicaland mould damage, In: 1997 ASAE Annual
International Meeting Technical Papers, Paper No.
973047, ASAE, 2950 Niles Road, St. Joseph,Michigan 4 9085-9659, USA, 1997.
[29].Ni, H., Gunasekaran, S., A computer vision
system for determining quality of cheese shreds, In:Food Processing Automation IV Proceedings of the
FPAC Conference. ASAE, 2950 Niles Road, St.
Joseph, Michigan 49085-9659, USA, 1995.[30].Ni, B., Paulsen, M.R., Reid, J.F., Size grading of
corn kernels with machine vision, In: 1997 ASAEAnnual International Meeting Technical Papers, Paper
No. 973046, ASAE, 2950 Niles Road, St.Joseph,
Michigan 49085-9659, USA, 1997a.
[31].Ni, B., Paulsen, M.R., Liao, K., Reid, J.F.,
Design of an automated corn kernel inspectionsystem for machine vision, Transactions of the ASAE,
40(2), pp. 491-497, 1997b.
[32].Panigrahi, S., and M.K. Misra, Feature
Extraction Techniques for Corn Germplasm by Colour
Computer Vision, In 1990 ASAE International
Summer Meeting, paper No 90-7050, St. Joseph,
Michigan, USA ,1990.[33].Payne, F.A., and S.A. Shearer, Colour and
Defect Sorting of Bell Peppers using Machine Vision,Transactions of the ASAE, 33(6), pp.2045-2050, 1990.[34].Pearson, T., and N. Toyofuku, Automated
Sorting of Pistachio Nuts with Closed Shells,Applied
Engineering in Agriculture, 16(1), pp.91-94, 2000.[35].Qui, W., and Shearer, S.A., Maturity Assessment
of Broccoli using the Discourse Fourier Transform,
Transaction of the ASAE, 35(6), pp. 2057-2062, 1992.[36].Raji, A.O., Discrete Element Modeling of the
Deformation of Bulk Agricultural Particulates,Unpublished Ph.D Thesis, University of Newcastle-
upon-Tyne, 1999.
[37].Raji, A.O., Fagboun, A.A., and M.K. Dania, An
Approach to Detecting Defects in Food Products,
-
8/8/2019 CV Agriculture 2010
11/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
11
Proceedings of the First International Conference of the
Nigerian Institution of Agricultural Engineers, pp.36-39,
2000.
[38].Reed, J.N., Crook, S., and W. He., Harvesting
Mushrooms by Robot,. In Science and cultivation of
edible fungi, pp . 385-391, Rotterdam: Elliott, Balkema,1995.
[39].Ruan, R., Ning, S., Ning, A., Jones, R., Chen, P.L.,
Estimation of scabby wheat incident rate using machine
vision and neural network, In: 1997 ASAE Annual
International Meeting Technical Papers, Paper No.
973042, ASAE, 2950 Niles Road, St. Joseph, MI 49085-
9659, USA, 1997.[40].Sapirstein, H.D.,Quality Control in Commercial
Baking: Machine Processing, Automation IV
Proceedings of the ASAE FPAC Conference, St. Joseph,Michigan, USA, 1995.
[41].Scott, A., Automated Continuous OnlineInspection, Detection and Rejection, Food Technology
Europe, 1(4), pp. 86-88, 1994.
[42].Shearer, S.A., Payne, F.A., Colour and defectsorting of bell peppers using machine vision,
Transactions of the ASAE, 33 (6), pp. 2045-2050, 1990.
[43].Slamet Riyadi, Ashrani A. Abd, Rahni Mohd,
Marzuki Mustafa, and Aini Hussain, Member, Shape
Characteristics Analysis for Papaya Size Classification,The 5th Student Conference on Research and
Development-SCOReD 2007.[44].Sonka, M, Hlavac, V, Boyle, R., ImageProcessing, Analysis, and Machine Vision, PWS
publishing, California, USA, 1999.
[45].Steinmertz, V., Rogr, J.M., Molto, E., and J. Blasco,Online Fusion of Colour Camera and
Spectrophotometer for Sugar Content Prediction ofApples, Journal of Agricultural Engineering Research,73, pp. 207-216, 1999.
[46].Steenhoek, L., Precetti, C., Vision sizing of seed
corn, In: 2000 ASAE Annual International Meeting,
Paper No. 003095 ASAE, 2950 Niles Road, St. Joseph,
MI 49085-9659, USA, 2000.[47].Sun, D.W., Computer Vision: An Objective,Rapid and Non-Contact Quality Evaluation Tool for the
Food Industry, Journal of Food Engineering, 61, pp.1-
2, 2003.[48].Sun, D.W., and T. Brosnan, Improving Quality
Inspection of Food Products by Computer Vision: A
Review, Journal of Food Engineering, 61, pp. 3-16,2003.[49].Tao, Y., and Z. Wen, An Adaptive Spherical
Image Transform for Characterizing Corn Growth andDevelopment, Transactions of the ASAE, 34(5), pp.
2245-2249, 1999.[50].Tao, Y., Morrow, C.T., Heinemann, P.H., Sommer,
H.J., Fourier based separation techniques for shape
grading of potatoes using machine vision, Transactions
of the ASAE, 38(3), pp. 949-957, 1995b.
[51].Timmermans, A.J.M., Computer Vision System
for Online Sorting of Pot Plants Based on Learning
Techniques, Acta Horticulturae, 421, pp. 91-98,
1998.
[52].Tollner, E.W., Shahin, M.A., Maw, B.W.,
Gilaitis, R.D., Summer, D.R., Classification of onionsbased on internal defects using imaging processing
and neural network techniques, In: 1999 ASEA
International Meeting, Toronto, Onteroi, Paper no.
993165, ASAF, 2950 Niles Road, St. Joseph, MI
49085-9659, USA, 1999.[53].Tsheko, R., Image Analysis by Machine
Vision, Ph.D Thesis, University of Newcastleupon-
Tyne, 1998.
[54].Wan, Y.N., Lin, C.M., Chiou, J.F., Adaptive
classification method for an automatic grain qualityinspection system using machine vision and neural
network, In: 2000 ASAE Annual InternationalMeeting, Paper No. 003094 ASAE, 2950 Niles Road,
St. Joseph, MI 49085-9659, USA, 2000.
[55].Wang, H.-H., Sun, D.-W., Evaluation of thefunctional properties of cheddar cheese using a
computer vision method, Journal of Food
Engineering, 49(1), pp. 47-51, 2001.
[56].Wang, H.-H., Sun, D.-W., Melting
characteristics of cheese: analysis of effects of cooking
conditions using computer vision techniques,Journal
of Food Engineering, 52(3), pp. 279-284, 2002a.
[57].Wang, H.-H., Sun, D.-W., Correlation betweencheese meltability determined with a computer vision
method and with Arnott and Schreiber, Journal of
Food Science, 67(2), pp. 745-749, 2002b.
[58].Wooten, J.R., White, J.G., Thomasson, J.A.,
Thompson, P.G., Yield and quality monitor for sweetpotato with machine vision, In: 2000 ASAE Annual
International Meeting, Paper No. 001123, ASAE,
2950 Niles Road, St. Joseph, MI 49085-9659, USA,
2000.
[59].Xie, W., Paulsen, M.R., Machine vision
detection of tetrazolium in corn, In: 1997 ASAE
Annual International Meeting Technical Papers, PaperNo. 973044, ASAE, 2950 Niles Road, St. Joseph, MI
49085-9659, USA, 1997.
[60].Yam, K.L., and E.P. Spyridon , A Simple Digital
Imaging Method for Measuring and Analyzing Colourof Food Surfaces, Journal of Food Engineering, 61,
pp. 137-142, 2003.[61].Zayas, I., Pomeranz, Y., Lai, F.S.,Discrimination of wheat and nonwheat componentsin grain samples by image analysis, Cereal
Chemistry, 66(3), pp. 233-237, 1989.
[62].Zayas, I.Y., Martin, C.R., Steele, J.L., Katsevich,A., Wheat classification using image analysis and
crush force parameters, Transactions of the ASAE39
(6), pp. 2199-2204, 1996.
-
8/8/2019 CV Agriculture 2010
12/12
2010 International Journal of Computer Applications (0975 8887)
Volume 1No. 4
12
[63].Zhang, N., Chaisattapagon, C., Effective criteria
for weed identification in wheat fields using machine
vision, Transactions of the ASAE, 38(3), 965-974,
1995.