Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)

IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

A Review on Plant Disease Detection Using Image

Processing

Anil A. Bharate, M. S. Shirdhonkar

Department of Computer Science and Engineering
B.L.D.E.A’s V. P. Dr. P. G. Halakatti College of Engineering and Technology
Vijayapur, India
[email protected], [email protected]

Abstract— India is the agriculture based country, since it

contributes 7.68 percent of total global agricultural output. In A. Apple Diseases
India, agricultural sector contributes about seventeen percentage The apple fruit is one of the most popular fruits worldwide.
of total Indian gross domestic product (GDP). Effective growth It has many health benefits and comes in a variety of colors
and improved yield of plants are necessary for increment of and flavors. This plant suffers from some common diseases
farmer’s profit and economy of India. For this purpose farmers such as
need domain experts for manual monitoring of plants. But
manual monitoring will not give satisfactory result all the time. 1) Apple Scab: It causes most devastating apple infection.
Moreover, domain experts are not available at all regions and are It occurs throughout the apple-growing areas. During the
expensive as farmers have to pay fees including travelling bloom in cool and wet weather, apple scab is more severe but
charges. Hence, it requires developing an efficient smart farming it is not reasonably significant in dry or warm climates. Signs
technique which will help for better yield and growth with less
human efforts. In this paper, we provide a review on methods
of apple scab are visible on leaves, petals, flowers, husk, fruit,
developed by various researchers for detection of diseases in young shoots and bud scales of apple tree. Mostly infection on
plants, in the field of image processing. It includes research in the fruit and leaves are common and obvious. Fig. 1 shows
disease detection of plants such as apple, grapes, pepper, apple scab disease on fruit.
pomegranate, tomato etc.

Keywords— Classification; Feature extraction; Image

processing; Plant disease; Symptom

I. INTRODUCTION
The diseases in plants can be determined by observing
specific patterns in plants, but it is difficult to obtain these
patterns in order to maintain plant’s health. There are various
techniques available such as spectroscopic and imaging Fig. 1. Apple scab disease
technology which can be used to obtain patterns. Farmers can
make use of automation techniques and tools to integrate 2) Apple Rot: It is a fungus caused due to Botryosphaeria
knowledge, product and services to improve productivity, obtusa. It attacks on fruits, leaves and bark of apple plant. The
grade and yield with the help of smart farming. It helps first symptom of apple rot appears on outer surfaces of leaves
farmers to identify the plant disease in primary stage and take one to three weeks after petiole fall as small, purple blotch
time to time decisions which in turn saves time and reduces after which centre turn brown tan and yellowish brown as
loss of plant due to diseases. Farmers will also be able to shown in Fig. 2. Second stage of apple rot occurs after few
identify different grades of fruit before marketing. weeks. In this stage secondary enlargement of leaf spots occur.
The purpose of this paper is to survey on how to monitor Leaf that is highly infected drops from the tree.
diseases on plants and suggest better solution for healthy yield
and productivity. Many existing systems use two databases of
images, one for query images and other for training images.
Diseases of three fruits namely apple, grapes and pomegranate
have been identified by many systems. Few types of fruit
plants such as apple, grapes and pomegranate and their
respective diseases are discussed below.
Fig. 2. Apple rot disease

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 103

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

3) Marssonina Leaf Blotch: This disease is identified by 1) Black Rot: Black rot is most widely occurring and
the formation of dark green circular blotches over the leaf. It severe disease. In this disease, fungus attacks canes, tendrils,
turns into brown leaf spots which eventually become dark leaves and fruit and it is most devastating in hot and moist
brown over a period of time. When leaf is badly affected by areas. First sign of black rot is visible on leaves as black
this disease large dark brown patches are formed and color of border forms around the edge and small yellowish spot is
leaf turns into yellow as shown in Fig. 3. This disease occurs formed, after that spot enlarges. Mostly symptoms appear after
because of high rainfall and moderate temperature during the half growth of grapes. Fig. 6 shows black rot disease in
formation of fruit in apple tree. grapes.

Fig. 6. Black rot disease in grapes

Fig. 3. Marssonina leaf blotch disease
2) Powdery Mildew: It is originated by fungus uncinula
4) Black rot canker: This disease appears first on leaves in necator, it is also called as Oidium. Only grapes and other few
the beginning of spring season. Purple color dots appear on related species are affected by this fungus. It is most
leaves which will form circular damaged region on leaves. frequently occurring disease on grapes. Primary symptom of
The boundary of damaged region remains purple, but centre powdery mildew is whitish or greenish powdery patches
turns into brown as shown in Fig. 4. After some days these appearing on the underneath of basal leaves. It also causes leaf
spots will enlarge in size and form circular bands with same curling, withering along with blotched or deformation of badly
centre on fruit. Color of these bands varies from brown to infected leaves. Fig. 7 shows powdery mildew disease.
black.

Fig. 7. Powdery mildew disease

Fig. 4. Black rot canker disease 3) Downy mildew: This is the fungal disease of grapes. It
occurs on leaves and can result in crop loss. Yellow color
5) Apple mosaic: Apple mosaic disease appears in the patches will be formed across the veins of leaves. Within few
form of cream color patches on leaves as shown in Fig. 5. days, back side of the leaves will be covered by white cotton
During hot summer these patches might become destructive, like mildew as shown in Fig. 8. This disease usually occurs in
and leads to distortion of leaves and reduction in size of spring season because of cool, moist and cloudy weather.
leaves. Apple stem grooving virus forms symptoms such as
swelling of the stem, chlorotic leaf spots, stern grooving and
pitting.

Fig. 8. Downy mildew disease

Fig. 5. Apple mosaic disease 4) Anthracnose: This disease is also called as bird’s eye
rot since spots on fruit look like eyes of the bird. Initially, red
spots will be formed on the berry. Within few days these spots
B. Grape Diseases
will become disk shaped, depressed and turn into gray color.
Grapes are a type of fruit which grow in clusters and Further, spots will be surrounded by a dark boundary, because
comes in various colors such as green, black, dark blue, of which it looks like eyes of the bird as shown in Fig. 9.
yellow and crimson. During growth, this plant suffers from
Continuous or irregular rainfall and high humidity cause this
bacterial and fungal diseases which are listed below.
disease.

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 104

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

Fig. 9. Anthracnose disease in grapes

Fig. 12. Bacterial blight disease
5) Bacterial Leaf Spot: This disease affects shoots,
berries and leaves. The small water dipped spots are formed 2) Aspergillus Fruit Rot: This disease appears when
on bottom of the leaves across veins as shown in Fig. 10. In flower begins to open after the rainfall and it infect the internal
few days these spots form large blotches. Brown black color portion of pomegranate. Tiny off color in the skin and less
damaged regions are formed on the berry, which will become weight due to internal decay are some of the exterior signs of
tiny and shrivelled. disease. But this problem usually is not apparent until
harvesting or during fruit sorting. Without any external
symptoms fungus may grow within the fruit. Mostly infected
fruit show some yellowish to brownish red discoloration and
are slightly off-color such as a pale red as shown in Fig. 13.

Fig. 10. Bacterial leaf spot disease

6) Rust: The symptoms of this disease appear as

innumerable spots with orange color on the back side of the
leaves as shown in Fig.11. In serious cases these spots which
are rounded swelling are formed on the whole surface of Fig. 13. Aspergillus fruit rot disease
leaves resulting into loss of leaves.
3) Anthracnose: This disease first appears in the form of
black spots on leaves and then color of leaves changes to
yellow and defoliation occurs. It also appears on tender or
mature fruits as brown spots which are circular in the
beginning and later become irregular shaped spots. These
spots spread on part of fruit or on whole fruit as shown in Fig.
14. This disease occurs usually in the month of August or
Fig. 11. Rust disease on leaves of grapes September due to high humidity.

C. Pomegranate Diseases
The pomegranate is a small fruit tree that grows between 5
meter and 8 meter tall. This is used in cooking, baking, juice
blends and alcoholic beverages such as cocktails and wine.
This plant suffers from few diseases during its growth such as
1) Bacterial Blight: This disease was first recorded in
Delhi (India) in the year 1952. Until 1998 bacterial blight was Fig. 14. Anthracnose disease on pomegranate
considered as a lower economic threat. However, now a day
this disease occurs widely and has been recorded in all states. 4) Cercospora fruit spot: This is a disease which occurs
This disease occurs in all pomegranate growing states like on pomegranate fruit in the form of pale brown spots as shown
Maharashtra, Karnataka, and Andhra Pradesh. Preliminary in Fig. 15. It also appears on twigs as black color spots. The
symptoms for the disease can be black color spots surrounded affected regions on the twig will be depressed and flattened,
by bacterial slime. 90% yield of pomegranate depletes due to surrounded by raised boundary. The twigs which are affected
bacterial blight. Fruits crack due to this disease as shown in by this disease will dry up and in some serious cases, entire
Fig. 12. plant of pomegranate dies.

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 105

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

In 2014, Deshpande et al. [5] have proposed the automatic

system which is used for grading pomegranate leaf diseases.
They have worked on most common disease in pomegranate
that is bacterial blight. For segmentation they have used K-
Means clustering algorithm. This system is not useful for
farmers directly but plant pathologists can make use of this
system. Rupanagudi et al. [6] proposed a method to estimate
maturity in tomato. This method is cost effective and is able to
Fig. 15. Cercospora fruit spot disease classify different stages in ripening of tomato using features
such as color and texture. This method is used only to examine
5) Cercospora leaf spot: This disease occurs on leaves the ripeness of tomato but not to identify the disease of
in the form of small brown color spots surrounded by yellow tomato. Ghaiwat and Arora [7] have done survey on
color circle. These circular spots spread over the leaf area and identification and classification of diseases in leaf. This work
turn into brown color with time as shown in fig. 16. On lower presents a survey on different classification methods. Existing
surface of the leaves grey color structure is formed due to systems have used various classification techniques like k-
group of sunken spots. nearest neighbor classifier, artificial neural network,
probabilistic neural network, support vector machine, genetic
algorithm, principal component analysis and Fuzzy logic.
Classification of leaf diseases has several applications in the
field of agricultural research. In 2013, Jhuria et al. [8]
indicated a method using image processing to identify fruit
diseases and also for fruit grading. The main aim of their work
is to examine disease spread on leaves or fruits and give
various solutions. They have worked on apple and grape
Fig. 16. Cercospora leaf spot disease diseases. Features such as texture, color and morphology are
used as input for artificial neural network for classification of
This paper gives the overview of the technical plant diseases. Tiger and Verma [9] presented an apple
implementations in the field of plant disease detection using recognition technique. Two classes such as normal apple and
image processing. First section gives the introduction about infected apple have been used. Proposed method uses two
diseases of fruit plants such as apple, grapes and pomegranate. layer feed forward network for classification. Dubey et al. [10]
Second section describes related work carried out in past ten proposed a method for fruit segmentation using color features.
years for detection of diseases in plants using image Unsupervised K means clustering algorithm is used. Apple
processing techniques by various researchers. In third section, fruit is used for case study. However, it fails to determine
we discuss challenges in plant disease detection. Fourth number of clusters required for segmentation of the defects.
section explains system architecture which is commonly used
to detect plant diseases and last section gives the conclusion. In 2012, Naganur and sannakki [11] presented the sorting
and grading of fruits using image processing techniques.
II. RELATED WORK Fuzzy logic approach is used for classification and grading.
Dubey and Jalal [12] have proposed disease identification
Number of attempts has been made to bring smart farming method for fruits. The method identifies apple diseases such as
together to work on plant disease detection. In 2017, Zang et apple scab, apple rot and apple blotch. Image segmentation is
al. [1] have proposed a correlation based feature selection done using K-means clustering technique. Features such as
method to identify apple leaf disease. Color, texture and shape Color Coherence Vector (CCV), color histogram, local binary
features are used as input for support vector machine (SVM) patterns and complete local binary patterns are extracted.
to classify apple leaf diseases such as mosaic, powdery Identification and classification of disease is done using
mildew and rust. However, it fails to identify apple leaf multiclass support vector machine. However, this method,
disease under natural illumination. In 2016, Francis et al. [2] fails to make fusion of more than one type of features. Pertot
have proposed a method to identify leaf diseases in pepper et al. [13] have provided multilingual web based tool for
plants. It identifies two types of diseases, quick wilt and berry visual plant disease detection. This system has two users; one
spot disease with the help of neural networks. Matlab R2012A who is able to use it for detecting plant disease and another is
is used for image analysis. In 2015, Bhange and Hingoliwala admin. This system is used for strawberry disease detection. In
[3] suggested a solution for the recognition of pomegranate 2011, Patel and Joshi [14] gave improved solution for locating
fruit disease. In this process, web based technique is applied to the fruits on the plant based on multiple features. Multiple
help farmers in recognizing fruit diseases. Farmers could feature extraction technique can include steps like extraction
capture an image of fruit affected by disease and upload it to of color and intensity feature, extraction of orientation feature,
the system. After this farmers would be able to see if the fruit extraction of edge feature, extraction of area from feature
is affected by bacterial blight or not. Awate et al. [4] proposed maps. The process is entirely automatic and it can work
a system to supervise the diseases on fruits like pomegranate, without user involvement. To improve performance it
apple and grapes and suggest alternate solution for healthy considers numerous features. Al-Hiary et al. [15] suggested an
yield and good productivity. However it fails to estimate the improved solution for automated diagnosis and grading of
severity of the disease automatically. plant leaves. Disorder can be diagnosed with help of K means

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 106

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

clustering procedure. Barrios et al. [16] provides a solution In 2008, Meunkaewjinda et al. [29] proposed a system
through which we can achieve fast and accurate testing in using multiple artificial intelligent techniques for automatic
laboratories for tomato grading instead of manual testing. diagnosis of plant disease. It can diagnose disease spread on
Grade assessment of tomato is prime task which can be plant leaf without need of any expert once the system is
acquired using image processing. Reis et al. [17] have trained. Work is done on grapes. SVM classifier is used to
proposed system for identifying white color grapes for robotic classify the grape leaves into scab or rust disease or no
harvesting system. The system was targeted to detect white disease. In 2007, Zhang et al. [30] proposed a method to
grapes, during night. However, it is unable to bring the identify disease in cotton leaf. Fuzzy curves and surfaces are
information about the grape maturation. They also have used to select features. Fuzzy curves are used to separate a
proposed a system [18] to identify bunches of grapes. It small set of notable features from the set of actual features. No
identifies and locates bunches of grapes and also classify as proper treatment advisory can be given by this approach
white and red grapes. Limitation of this system is that automatically to prevent further loss.
detection of bunches of grapes is done only during night
conditions. Sannakki et al. [19] proposed a system to grade the The system architecture for fruit disease detection using
disease on plant leaf automatically. It inculcates information image processing includes various phases as shown in Fig. 17.
The various phases for fruit disease detection are discussed in
and communication technology in farming. It uses fuzzy logic
for grading the disease spread on plant leaf. However, it is the following section.
unable to provide disease prevention information to the
farmers. III. SYSTEM ARCHITECTURE
In 2010, Sankaran et al. [20] suggested imaging, The common system architecture for plant disease
spectroscopic and molecular techniques. The imaging and detection using image processing is as shown in Fig. 17. The
spectroscopic technologies are used in agricultural vehicle for various phases for plant disease detection are
detection of plant diseases. Arivazhagan et al. [21] proposed
an efficient method for fusion of features such as color and A. Image Acquisition
texture to identify fruit. Classification method such as It is the process of pictorial image creation of a physical
minimum distance classifier is used for identifying fruit. view or the internal structure of an object. Image acquisition
Sandoval et al. [22] have proposed a classification system can be widely described as the activity of restoring a picture
using machine vision to classify coffee fruits (cherries) from some origin, usually a hardware-based source which can
depending on level of ripeness. Several classes are considered be processed along with processes that need to appear
for different stages of ripening process of coffee fruit. Bashish afterwards. Image acquisition is consistently the initial
et al. [23] proposed a method for detection of plant or stem condition for the work flow series of image processing
diseases. Segmentation is done using K-means algorithm. because as processing is possible only with the help of an
Segmented images are used as input to the neural network. image. The image obtained is entirely natural and is the
However, this method is unable to give prevention consequence of any hardware which was handled to produce
information. In 2009, Effendi et al. [24] presented that the it.
quality of fruit is judged by the color, size and type of defects.
In this work image identification method is developed to find B. Pre processing
the maturity level of Jatropha curcas fruit. A back propagation
diagnosis model (BPDM) is used for recognizing the image of Pre processing involves enhancement of the visual
the matured fruit. Hannan et al. [25] proposed a method for appearance of images, increasing or decreasing the number of
fruit detection. Orange fruit is considered for case study. The pixels of the dataset, resizing images and removing noise from
enhancement is done using the red chromaticity coefficient. the images.
To improve the efficiency of fruit identification, they have
used fuzzy logic along with laplacian pyramid transform Training Database Query Image
(LPT). Fernandez et al. [26] proposed a computer vision
system to grade the grapes in wine cellars. It provides
automatic classification of grapes in cooperative wine cellars. Pre processing Pre processing
It also sorts grapes for the production of different quality
wines. However, it is unable to test different variety of grapes.
Syahrir et al. [27] proposed image processing method for Feature Extraction Feature Extraction
estimating tomato maturity. It judges the maturity and expiry
date of tomato based on their color. Red and green values are
extracted during feature extraction. These values are used for
identifying the maturity percentage of tomato and expiry date. Classification Disease Detection
It fails to separate tomato from other fruits or vegetables. Yin
et al. [28] proposed a method of ripe tomato extraction using Fig. 17. System architecture
features such as color and morphology. There are some
drawbacks of this method such as it results in the false cluster C. Image Segmentation
segmentation for ripe tomato. It is the method of dividing a digital image into several
parts. The primary aim of segmentation is to recognize objects

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 107

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

or extract the related information from images, so that variety of plant diseases may not be available at some farms
analyzing an image becomes easier. Objects and bounding line and diseases occur only during some seasons.
of images are located by using image segmentation. Pixels
with similar label portion share distinguishing features for B. Image background
allocating a label to each pixel in an image. Image segmentation is important phase of image
processing, where we separate most required part of image.
D. Feature Extraction Leaf image segmentation may be a challenge if background
Feature extraction is the process of deriving a set of values contains plants, leaves and some other green elements.
called features from an image, which provides information
about the image for further processing. In the process of C. Image capture condition
identifying the diseases in plants, features such as color, Automatic plant disease detection systems give steady and
texture, morphological and color coherence vector are efficient results, only if all the images are captured under same
commonly used. condition. Capturing images under same condition is possible
1) Color features: A basic feature to observe change in only inside laboratories. It’s a challenge to capture images
plants is color. Color feature is commonly used by many under same condition in the field because of uncontrollable
researchers for classification of plant diseases. There are many environment.
methods for color feature extraction such as color histogram,
histogram intersection, color correlogram, color co occurrence D. Symptom segmentation
matrix, color coherence vector etc. Most plant disease symptoms have no well defined edges
2) Texture features: Texture is nothing but a duplicated and they fade on plants slowly because of which there will not
structure of information or the arrangement of the structure be a proper segmentation, which will affect final result.
that occurs at uniform intervals. Since every plant disease has
E. Symptom variations
unique structure, texture features are used to detect and
classify plant diseases. Symptom depends on environment, disease and plant. Any
3) Morphological features: Morphological features refer change in these elements may result in symptom variations.
It’s a challenge to identify the plant disease with symptom
to a shape or a particular form of an object. As plants and
variations.
plant diseases come with various shapes, morphological
features are mostly used for plant disease detection. Using F. Multiple simultaneous disorders
these features we can extract various components of an image.
Many times automatic plant disease detection systems may
These components are used for extracting the boundary which
wrongly assume that there is only one disorder present in an
separates different regions in an image and hence, plant image. Pests and nutritional deficiencies may occur
disease can be identified easily. simultaneously because there is maximum possibility of plant
E. Pattern Matching and Classification being attacked by other disorders after having infected by
some disease.
It is the procedure of examining stated successions of
tokens for the existence of the elements of some pattern. The
task of matching algorithm is to compare features with index G. Different disorders with similar symptoms
features of the image present in the database. Classification is Many plant disorders have similar symptoms such as
a process of identifying the category of the observed pattern. diseases, nutritional deficiencies, pests, phytotoxicity,
Two major categories of classification are supervised and excessive cold or heat. It’s a challenge to differentiate and
unsupervised. In supervised classification, training is required identify the disorders by automatic plant disease detection
where user can choose sample pixels to form a class. techniques.
Unsupervised classification needs no training and results are
based on the software analysis without sample classes. V. CONCLUSION
Classification techniques such as support vector machine,
In this paper, we present a review of the technical
neural networks, k- nearest neighbour, fuzzy logic etc. are
implementations in the research area of plant disease detection
used for plant disease detection.
using image processing techniques. From literature it is
evident that color, texture and morphological features are most
IV. CHALLENGES suitable to identify and classify the diseases in plants. Most
There are few challenges [31] in plant disease detection commonly used classification techniques for classifying plant
using image processing techniques which are listed below diseases are artificial neural networks (ANN) and support
vector machine (SVM). Automatic detection of plant diseases
A. Collection of data set would solve the problem of expensive domain expert.
Detection of plant diseases in early stage would help farmers
Basic need of image processing is creating a database of to improve the crop yield, which in turn improves Indian gross
images. To acquire images of plant diseases, one has to travel domestic product (GDP).
to different places. Data collection will be a challenging since

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 108

 Proceedings of the International Conference on Intelligent Sustainable Systems (ICISS 2017)
IEEE Xplore Compliant - Part Number:CFP17M19-ART, ISBN:978-1-5386-1959-9

[16] Abraham Gastelum-Barrios, Rafael A. Borquez-Lopez, Enrique Rico-

Acknowledgment Garcia, Manuel Toledano-Ayala and Genaro M. Soto-Zarazua, “Tomato
Authors wish to thank ICAR-National Research Centre on Quality Evaluation with Image Processing: A Review”, African Journal
of Agricultural Research, Vol. 6(14), pp. 3333-3339, 2011.
Pomegranate, Solapur (Maharashtra) for providing
[17] Manuel Cabral Reis, Raul Morais, Carlos Pereira, Salviano Soares, A.
information about pomegranate diseases. Valente, J. Baptista, Paulo J. S. G. Ferreira and J. B. Cruz, “Automatic
Detection of White Grapes in Natural Environment Using Image
References Processing”, In proc. of Sixth International Conference on Advances in
Intelligent and Soft Computing, pp. 19-26, 2011.
[1] Zang Chuanlei, Zang Shanwen, Yang Jucheng, Shi Yancui and Chen Jia, [18] Manuel J. C. S. Reis, Raul Morais, Carlos Pereira, Olga Contente,
“Apple Leaf Identification Using Genetic Algorithm and Correlation Miguel Bacelar, Salviano Soares, Antonio Valente, Jose Baptista, Paulo
Based Feature Selection Method”, International Journal of Agricultural J. S. G. Ferreira and Jose Bulas-Cruz, “A Low-Cost System to Detect
and Biological Engineering, pp. 74-83, 2017. Bunches of Grapes in Natural Environment from Color Images”, In
[2] Jobin Francis, Anto Sahaya Dhas and Annop B. K., “Identification of proc. of International Conference on Advanced Concepts for Intelligent
Leaf Diseases in Pepper Plants Using Soft Computing Techniques”, In Vision Systems, pp. 92-102, 2011.
proc. of IEEE Conference on Emerging Devices and Smart Systems, pp. [19] Sanjeev S. Sannakki, Vijay S. Rajpurohit, V. B. Nargund, ArunKumar
168-173, 2016. R. and Prema S. Yallur , “Leaf Disease Grading by Machine Vision and
[3] Manisha Bhange and H. A. Hingoliwala, “Smart Farming: Pomegranate Fuzzy Logic”, International Journal of Computer Technology and
Disease Detection Using Image Processing”, In proc. of the Second Applications, Vol. 2 , pp. 1709-1716, 2011.
International Symposium on Computer Vision and the Internet [20] Sindhuja Sankaran, Ashish Mishhraa, Reza Ehsania and Cristina Davis,
(VisionNet’15), pp. 280-286, 2015. “A Review of Advanced Techniques for Detecting Plant Disease”,
[4] Ashwini Awate, Damini Deshmankar, Gayatri Amrutkar, Utkarsha Computers and Electronics in Agriculture, Vol. 72, pp. 1-13, 2010.
Bagul and Samadhan Sonavane, “Fruit Disease Detection using Color, [21] S. Arivazhagan, R. Newlin Shebiah, S. Selva Nidhyanandhan and L.
Texture Analysis and ANN”, In proc. of International Conference on Ganesan, “Fruit Recognition using Color and Texture Features”, Journal
Green Computing and Internet of Things(ICGCIoT), pp. 970-975, 2015. of Emerging Trends in Computing and Information Sciences, Vol. 1,
[5] Tejal Deshpande, Sharmila Sengupta, and K. S. Raghuvanshi, “Grading October 2010.
& Identification of Disease in Pomegranate Leaf and Fruit”, [22] Sandoval, Z. Prieto F. and Betancur J., “Digital Image Processing for
International Journal of Computer Science and Information Classification of Coffee Cherries”, Electronics, Robotics and
Technologies, Vol. 5, pp. 4638-4645, 2014. Automotive Mechanics Conference (CERMA), 2010.
[6] Sudhir Rao Rupanagudi, Ranjani B.S., Prathik Nagaraj and Varsha G. [23] Dheeb Al Bashish, Malik Braik and Sulieman Bani-Ahmad, “A
Bhat, “A Cost Effective Tomato Maturity Grading System using Image Framework for Detection and Classification of Plant Leaf and Stem
Processing for Farmers”, In proc. of IEEE International Conference on Diseases”, In proc. of International Conference on Signal and Image
Contemporary Computing and Information(IC3I), pp. 7-12, 2014. Processing(ICSIP), pp. 113-118, 2010.
[7] Savita N. Ghaiwat and Parul Arora, “Detection and Classification of [24] Effendi Z., Ramli R., Ghani J. A. and Rahman, “Pattern Recognition
Plant Leaf Diseases Using Image Processing Techniques: A Review”, System of Jatropha Curcas Fruits using Back Propagation”, In proc. of
International Journal of Recent Advances in Engineering & Technology IEEE International Conference on Signal and Image Processing
(IJRAET), Vol. 2, pp. 2347 - 2812, 2014. Applications (ICSIPA), 2009.
[8] Monica Jhuria, Ashwani Kumar and Rushikesh Borse, “Image [25] Hannan M. W., Burks T. F. and Bulanon D. M., “A Machine Vision
Processing for Smart Farming: Detection of Disease and Fruit Grading”, Algorithm for Orange Fruit Detection”, Agricultural Engineering
In Proc. of the IEEE Second International Conference on Image International: the CIGR E Journal, Vol. 11, pp. 1-7, 2009.
Information Processing(ICIIP), pp. 521-526, 2013.
[26] Esteban Vazquez-Fernandez, Angel Dacal-Nieto, Fernando Martin,
[9] Bindu Tiger and Toran Verma, “Identification and Classification of Arno Formella, Soledad Torres-Guijarro and Higinio Gonzalez-Jorge,
Normal and Infected Apples using Neural Network”, International “A Computer Vision System for Visual Grape Grading in Wine Cellars”,
Journal of Science and Research (IJSR), Vol. 2, Issue 6, pp. 160-163, In proc. of seventh International Conference on Computer Vision
2013. Systems, pp. 335-344, 2009.
[10] Shiv Ram Dubey, Pushkar Dixit, Nishant Singh and Jay Prakash Gupta, [27] W. Md. Syahrir, A. Suryanti and C. Connsynn, “Color Grading in
“Infected Fruit Part Detection using K-Means Clustering Segmentation Tomato Maturity Estimator using Image Processing Technique”, In
Technique”, International Journal of Artificial Intelligence and proc. of IEEE Second International Conference on Computer Science
Interactive Multimedia, Vol. 2, pp. 65-72, 2013. and Information Technology(ICCSIT), pp.276-280, 2009.
[11] Naganur and sannakki, “Fruits Sorting and Grading using Fuzzy Logic”, [28] Hongpeng Yin, Yi Chai, Simon X. Yang and Gauri S. Mittal, “Ripe
International Journal of Advanced Research in Computer Engineering & Tomato Extraction For A Harvesting Robotic System”, In proc. of IEEE
Technology (IJARCET), Vol. 1, Issue 6, pp. 117-122, 2012. International Conference on Systems, Man, and Cybernetics, pp. 2984-
[12] Shiv Ram Dubey and Anand Singh Jalal, “Detection and Classification 2989, 2009.
of Apple Fruit Diseases using Complete Local Binary Patterns”, In proc. [29] Meunkaewjinda, Kumsawat, At-takitmongcol K. and Srikaew, A.,
of IEEE, Third international conference on Computer and “Grape Leaf Disease Detection from Color Imagery using Hybrid
Communication Technology, pp. 247-251, 2012. Intelligent System”. In proc. of fifth International Conference on
[13] Illaria Pertot, Tsvi Kuflik, Igor Gordon, Stanley Freeman and Yigal Electrical Engineering/Electronics, Computer, Telecommunications and
Elad, “Identificator: A Web-Based Tool for Visual Plant Disease Information Technology, Vol. 1, pp.513–516, 2008.
Identification, a Proof of Concept with a Case Study on Strawberry”, [30] Yan Cheng Zhang, Han Ping Mao, Bo Hu and Ming Xili, “Features
Computers and Electronics in Agriculture, Elsevier, Vol.88, pp.144-154, Selection of Cotton Disease Leaves Image Based on Fuzzy Feature
2012. Selection Techniques”, In proc. of the IEEE International Conference on
[14] Hetal N. Patel, Dr. M. V. Joshi, “Fruit Detection using Improved Wavelet Analysis and Pattern Recognition, Beijing, China, Vol. 1, pp.
Multiple Features Based Algorithm”, International Journal of Computer 124-129, 2007.
Applications, Vol. 13, pp. 0975– 8887, 2011. [31] Jayme Garcia Arnal Barbedo, “A review on the main challenges in
[15] H. Al-Hiary, S. Bani-Ahmad, M. Reyalat, M. Braik and Z. automatic plant disease identification based on visible range images”,
ALRahamneh, “Fast and Accurate Detection and Classification of Plant Science Direct, Biosystems Engineering, pp. 52-60, 2016.
Diseases”, International Journal of Computer Applications, Vol. 17, pp.
0975 – 8887, 2011.

978-1-5386-1959-9/17/$31.00 ©2017 IEEE 109