DISEASE CLASSIFICATION AND GRADING OF

ORANGE FRUIT USING MACHINE LEARNING AND

FUZZY LOGIC

Lipsarani jena

Computer science & Engineering

Veer Surendra Sai University Of Technology, Burla
2018

1
Disease Classification and Grading Of Orange Fruit
Using Machine Learning and Fuzzy Logic

A major project Submitted in partial fulfillment of the

Requirements for the Degree of

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE & ENGINEERING

by
LIPSARANI JENA
Regd.no.-15021236

Under The Guidance of

Mrs. Santi kumari Behera
Asst. professor

DEPARTMENT OF COMPUTER SCIENCE AND ENGG.

VEER SURENDRA SAI UNIVERSITY OF TECHNOLOGY,
BURLA

2
 VEER SURENDRA SAI UNIVERSITY OF TECHNOLGY , BURLA, ODISHA

Declaration

I declare that this written submission represents my ideas in my own words and where others
ideas or words have been included, I have adequately cited and referenced the original
sources. I also declare that i have adhered to all principles of academic honesty and integrity
and have not misrepresented or fabricated or falsified any idea/data/fact/source in my
submission. I understand that any violation of the above will be cause for disciplinary action
by the university and can also evoke penal action from the sources which have thus not been
properly cited or from whom proper permission has not been taken when needed.

DATE: Lipsarani jena

Regd.no.-15021236

3
 Department of computer science & engineering,
Veer surendra sai university of technology, Burla, Odisha

Certificate

This is to certify that the dissertation entitled "Disease detection and grading of orange
fruit using machine learning and fuzzy logic" submitted by LIPSARANI JENA is
approved for the degree of bachelor of technology in computer science and engineering , is
a record of an original research work carried out by her under my supervision and guidance.

Dr. Rakesh Kumar mohanty Mrs. Santi Kumari Behera

Head Of Department Supervisor

4
 Acknowledgement

I would like to express my sincere gratitude to my supervisor, Mrs. Santi Kumari Behera, for
her invaluable help during the course work towards this dissertation. She was a source of
constant ideas and encouragement and provided a friendly atmosphere to work in. I am really
very thankful to her for everything.

I am also thankful to Dr. Rakesh Kumar Mohanty, Head of the Department and to all the
faculties of Department of Computer Science and Engineering for having supported me to
carry out this dissertation and for their constant advice. I would like to thank all my friends
for their encouragement and understanding. I would like to express my heart-felt gratitude to
them.

Lipsarani jena
Regd.no.-15021236

5
 Approval Sheet

This dissertation entitled "Disease detection and grading of orange fruit using machine
learning and fuzzy logic" by Lipsarani jena is approved for the degree of bachelor of
technology in "Computer science and engineering", department of computer science and
engineering.

Examiner

Date: Supervisor
Place: Burla

6
 Abstract

In this project, a system for quality control in orange fruit is presented. In current orange
manufacturing industries, caliper & color are successfully used for the automatic
classification of the fruits using machine vision system. However, the detection of flaws in
the orange surface is carried out by means of human inspection. In this work, a computer
vision system capable of detecting defects in the orange and also classing the type of flaw is
presented. First, a review of citrus illness has been carried out in order to build a database of
digitalized oranges classified by kind of fault, which is used as a training set. The symptoms
of orange diseases imply the seriousness of the disease and suggest choosing the best
approach to dealing with the disease. It is also needful to diagnose the disease properly and in
time to avoid the great harm of the orange the treatment of orange diseases by applying
disproportionate pesticides increase the cost and environmental pollution. So the use of
pesticide must be minimized. This can be actualized by targeting the diseased area, with the
appropriate quantity and concentration of pesticide by estimating disease severity. This paper
introduces multi class SVM with k-means clustering for classification disease with 90% and
Fuzzy Logic compute the degree of disease severity successfully of orange.

7
 Contents

1. Introduction 10
1.1 Motivation 12
1.2 Objective 13
1.3 Organization of Thesis 14
2. Background 15
2.1 Background concepts 16
3. Literature Review 17
4. Materials and Methodology 21
4.1 Disease Classification 21
4.2 Disease Severity 24
5. Result and Discussion 25
6. Conclusion 28
7.Bibliography 29

8
 List Of Figures

Figure-1. Orange Disease Classification Approach.

Figure-2. Different Type of Orange Fruit Diseases.

Figure-3. Fuzzy logic Overview System.
Figure-4. Severity of Citrus Canker Orange.
Figure-5. Severity of Stubborn Disease Orange.
Figure-6. Severity of Brown Rot Orange.
Figure-7. Severity of Melanoses Orange

List Of Tables

Table-1. Classification of Diseased Orange

9
 Chapter 1

Introduction

India is an agricultural country. International comparisons reveal the average yield of

orange in India is generally 30%-50% of the highest average yield in the world. Agriculture
has comprised of 16.5% GDP by sector (2016 est.) with approximately 50% labor force (2104
est.) & 10% of total export. The budget 2107-18 pitched for more reforms in agriculture sector
& increased funds for almost all areas of Agriculture [1].

India exported $39 billion worth of agriculture product in 2013, marketing it 7th largest
agriculture exporter worldwide. According to 2010 FAO, India is world’s largest producer of
many fresh fruit & vegetables, milk, major spices, jute miller & castor oil seeds. India is
world’s second the largest producer of wheat & rice. India is world’s second or third largest
producer of several dry fruits,agriculture-based textile raw materials, roots& tuber crops,
pulses, farmed fish, eggs, coconut, sugarcane & numbers of vegetables [2].

There are many diseases that hamper the growth & productivity of plants which leads to
great ecological &economic losses. For this reason, it is better to diagnose diseases accurately
& major the severity of the disease to avoid such losses timely.There are four main different
types of diseases that affect orange fruit. These are (a) citrus canker (b) stubborn disease (c)
brown rot (d) melanoses. The detection of orange diseases can be done in many ways
including manual and computer-based systems. Manual measurements are visually made by
plant pathologists. So, they must be used their experience to carefully manner. Such kind of
process can be costly, lengthy and tiresome and can also lead to errors due to fatigue. So, new
techniques are like image processing, computer visions are used to get more accurate with
consuming less time and at an affordable cost [3].

10
 In this work, we introduce an approach to classify mostly appear four type of orange
disease and measure the severity of thedisease. This paper consists of two section: 1)
Classification of diseased orange, 2) Severity measurement of disease.

First, the prepared sample image is segmented using K-means clustering to separate the
background, foreground and defected part. The defected part is our region of interest (ROI).
Then from the ROI part 13 number of texture features extracted and trained to SVM. So that
the SVM match the feature of trained feature and the feature of thetest sample and classify
accordingly.

The second section is meant for measurement of disease severity.For automatic

computation of disease, a fuzzy logic based model is proposed. Here the fuzzy model
computes the severity of disease by estimating the rate of flaw mark appears on the skin of an
orange.

11
1.1 Motivation

Agriculture is considered as backbone of a country. For a developing country like India, its
significance is even more because it is said that “ India lives in its villages” and agriculture is
the main occupation upon which the village mass depend. That is why, agriculture also
affects the Indian economy to a higher extent. But, the traditional techniques, still adopted by
farmers of India, make them unable to cope with the advancing technologies, increasing
demand and producing qualitative crops .This is a vital reason for which farmers of India
constitute a major part of ‘Below Poverty Line’ population of India.

Our Honourable Prime Minister Shri Narendra Modi, has emphasized the advancement of
Indian agriculture in many of his speeches, which shows importance of this research domain
in current trends, especially in India. By the help of image processing, automation of farming
techniques can be accomplished successfully, giving rise to such innovations which will not
only provide relief to the human labor needed in the field of agriculture but also increase the
proficiency of farming techniques, thus helping in improving both the quality of Indian
farming and quantity of production. A new dimension may be given to Indian agriculture by
adopting automated farming approach, in which image processing has got a mandatory
application. Thus, development of agricultural techniques is also an essential task with
advancement in various fields. Estimation of crop yield and quality forms the major part of
agriculture. So, Computer vision and image processing techniques are commonly used for
yield mapping systems. Our work basically focuses on Fruit cultivation techniques.

Thus suitable approaches has been implemented for developing an algorithm that can be
used in automated systems for fruit disease detection and fruit gradation.

12
1.2 Objective

Objective of our work is to develop an algorithm, that would be used in building an

automated approach, which can be effectively used in the field of horticulture, in the farming
of fruits, for the following purposes -

• To classify the diseased orange

• To calculate the disease severity of that diseased orange.

13
1.3. Organization of the thesis

The whole thesis has been organized in the following manner.

Chapter 2 Includes some basic concept of image processing, that has been used in the
work.

Chapter 3 presents an existing methodology that has been implemented and analyzed.

Chapter 4 presents the materials and methodology and its algorithms and flowcharts.

Chapter 5 is about results and discussion upon various sample data set.

Chapter 6 concludes the thesis with some prospects of future work.

14
 Chapter 2

Background Study

Image refers to a still picture of a time, captured via photo-sensitive sensor or devices with
these sensors. An image is always two-dimensional, but its perception and processing may be
done in two ways i.e. two-dimensional imaging and three-dimensional imaging, though the
imaging techniques for both are different. Before going through the techniques of processing
an image, understanding of few basic concepts is needed. Thus, some of the basic concepts of
image processing are described in the following sections.

2.1. Basic Concepts

Pixel

The basic logical unit of an image, which have a specific co-ordinate value i.e. X-axis value
and Y-axis value on the plane, on which the image is displayed. Every pixel stores some
values of intensity basing upon the color model in which the image is displayed.

Preprocessing

Image preprocessing is a prophase relative to feature extraction& image reorganization. The

image which has input is always not satisfactory regardless of what image acquisition devices
are adopted. Here preprocessing phase consists of two steps: Image Enhancement and Lab
Color Transformation. Image enhancement is one of the most interesting and important issues

15
in image processing field. The main purpose of image enhancement is to bring out details that
are hidden in an image or to increase the contrast in a low contrast image. Image
enhancement produces an output image that subjectively looks better than the original image
by changing the pixel’s intensity of the input image. We use L*a*b* color space because the
color information in the L*a*b* color space is stored in only two channels (i.e.; a* & b*
component), and it causes reduced processing time for the defect segmentation.

Segmentation

Image segmentation is the classification of an image into different groups. K-means

clustering algorithm is an unsupervised algorithm and it is used to segment the interest area
from the background. Here we have taken the number of clusters, K=3 to separate
background, foreground& the affected area.

Gray scale Image

Gray scale is a range of intensity values varying between black and white, thus generating
different lighter and brighter shades of black and white. An image is said to be gray scale , if
its pixel are occupied by the values of gray scale. A gray scale image is two dimensional.

Binary Image

Pixels of a binary image can have only 2 values i.e 0 or 1. A binary image is two-
dimensional, and its every pixel hold only one value, either 0 or 1.

16
 Chapter 3

Literature Review

Previously many researches have done to detect, classify and quantify the diseases of plant
diseases. The paper [4] introduce K-Means clustering technique to detect the defected part of
rice leaf. Also they calculate affected area. The support vector machine is an efficient tool for
classification. The SVM with K-means and Fuzzy C means clustering able to identify the five
different type of mineral deficiency of rice crop from leaf images with accuracy of 85% and
90% respectively [5] .

For automatic grading purpose of rice leaf, fuzzy logic is implemented [6]. In the field of
disease management, grade of the disease is determined to provide a precision and accurate
treatment advisory. The result is getting from the automatically grading system using fuzzy
logic which proved to be accurate and satisfactory in contrast with manual grading. We also
focus on the previous work by several researchers in the area of image fruit diseases
identification & risk of that disease.The paper [7] [8] study various grading method of leaf
diseases and fruit diseases of many species.

In paper [9] present the framework for Carica papaya grading using the Artificial Bee
Colony algorithm (ABC). They consider image features indicates that affected area, shape
and textures could be used as the parameters for the ABC algorithm for classifying the
papaya fruits into its respective grades. Xiaoling Deng et al.[10] have described citrus
greening detection using visible spectrum imaging and C-SVC. This work has been given an
accuracy of 91.93%. XubeingBai et al. [11] have described a fuzzy clustering segmentation

17
method based on neighborhood grayscale informing for defining cucumber leaf spot disease
images. This analysis has given an accuracy of 98.78%.

In paper [12] authors present image processing technique for Apple fruit disease detection.
Apple blotch, apple rot and apple scab are diseases that studied and evaluated in author's
research. First read input image and transform it from RGB to L*A*B* color space in image
preprocessing. The features used for fruit disease identification are global color histogram,
local binary pattern, Color Coherence Vector and completed Local Binary Pattern. K
Means clustering in 'a*b*' space is used for defect segmentation. Then from the result of K-
Means, features are extracted to label each pixel in the image. It generates image of fruit that
segment the image by color & shows the disease containing segments. CLBP feature gives
more accurate result for identification of diseases. Multiclass SVM performs training &
classification of fruit diseases.

In paper [13] author present image processing technique for grapes & apple fruit disease
detection also grading of these fruit. Two main diseases of grapes are black Rot &Powdery
Mildew. Two main diseases of apple are Rot & Apple Scab. Image acquisition method is
capturing p of digitals images, after that author explained image preprocessing technique
which includes feature extraction. Three feature vector namely: morphology, texture & color
are used for feature extraction. Image components for boundaries are extracted using
morphology. Various visual patterns are described by texture feature. RGB color space is
converted to HSI color space in color feature extraction also histogram of image is computed.
ANN neural network &backpropagation algorithm is used for classification diseases in
network. Lastly fruit grading is determined by the calculation of spread of disease on fruit&
weight of fruit.

In paper [14] authors present image processing technique for Pomegranate fruit diseases
detection. This fruit is affected by Bacterial Blight, fruit spot, fruit Rot & Leaf Spot. Images
of fresh pomegranate are captured for creation of database& further processing. In pre-
processing to remove noise filtering is done using (3*3), (5*5), (7*7) masks. In segmentation
K-Means methods is used to divide image into objects and regions. Statistical texture features
are extracted using Gary level co-occurrence matrix (GLCM) formula. For training the neural
networks, authors use Multilayer Perceptrons (MLP) to solve difficult problems by training
the images using back-propagation algorithm.

18
In paper [15] authors present image processing technique for fruit disease detection with
improved K-Means clustering algorithm. User captures the disease image using digital media
and loads fruit samples with its name to software after registration process. The image is then
converted from RGB to HSV and grayscale. In threshold process, histogram and multi-level
threshold techniques are used to isolate the relevant image from digital image after that
boundary detection is carried out the get the required area for next processing. Software crops
they require portion of image & generates the blocks after the segmentation for proper
evaluation the proton. Improved K-Means algorithm calculates centroid value for each block
& compares process data with database to obtain the result. The final result shows the defect
detection & the disease of infected fruit.

In paper [16] authors present image processing technique for Apple, Grape and Pomegranate
fruits disease detection. Powdery Mildew, Downy Mildew and Black Rot are diseases of
grape. Apple Scab, Apple Blotch and Apple Rot are diseases of apple fruit and Fruit Rot,
Gray Mold and Bacterial Blight are diseases of pomegranate fruit. In image acquisition
natural images of fruits are obtained by hardware. Apply image segmentation for locating the
objects and bounding line of given images. K-means clustering algorithm is used for labelling
each pixel in the image and also they used SURF (Speed up Robust Feature) algorithm for
extracting features such as color, structure, texture and morphology. SURF algorithm uses the
blob detector and local descriptors. Blob analysis consists of refinement, extraction and
analysis for detecting the scope of interest of given images. Artificial neural network (ANN)
is applied for pattern matching and classification of diseases.

In paper [17] author’s present image processing techniques for mango fruit disease detection.
In image acquisition web camera is used for taking video of mango fruit and further video is
converted to 100 frames for each mango sample. Preprocessing converts the original image
to binary image also display color threshold ranges by computing histograms of image.
Authors used watershed algorithm in image segmentation for automatic detection of video
frames in order to identify defected regions. Features are extracted from segmented image
using blob extraction method in template matching algorithm. Finally diseases are classified
using Normalized correlation methods in template matching algorithm for automatic fruit
disease identification. This algorithm takes less time to recognize objects in an image also it

19
shows region that is defected in the fruit. At last, the fruit grading is performed for detecting
the diseased fruit.

In paper [18] authors present image processing technique for Apple fruit disease detection.
Apple fruit’s most mutual diseases are apple blotch, and apple rot. Euclidean distance is used
for K-means clustering and image is converted from RGB color space to L*a*b* color space.
In segmentation, K-means algorithm is used for region of interest. K-means algorithm is
carried to find infected region in an image. Author described color feature and texture feature
extraction methods to extract color, shape and texture of fruits for disease classification.
Color features are Color coherence vector (CCV) and Global color histogram (GCH). Texture
features are complete local binary pattern (CLBP), Gabor Features, Local binary pattern
(LBP) and Local ternary pattern (LTP). Then authors performed feature level fusion by
fusing more than two texture and color features also applied Random forest classifier on
segmented image. This method tags every pixel in image and after classification it shows
fruit diseases.

20
 Chapter 4

Materials and Methodology

In this work, we present a fast computer vision citrus fruits grading system based on using
image processing and fuzzy logic. Our system observes the external surface of orange fruits
and detects the flaw mark using K-means clustering and from the segmented image, the
featuresare extracted using GLCM and finally Multiclass SVM classify the type of disease in
orange. Another part of our work is to measure the severity of disease using K-means
clustering and Fuzzy logic.

A. DISEASE CLASSIFICATION

Image Test Image

Preparation Sample
Preprocessing Preprocessing
Test Train
Image Segmentation Image Segmentation
Feature
Feature Extraction
Extraction

Trained Multi-Class SVM

Classification of Diseases Present in Orange

Fig.1 Orange Disease Classification Approach.

Theframework of Orange Disease Classification is shown in Fig.1 and proposed Algorithm is

as follow:

Step 1: Read input image

Step 2: Contrast Enhancement.
Step 3: transform image from RGB to L*a*b* color space
Step 4: Extract disease containing segment of Orange using K-means clustering.
Step 5: Extract feature from the segmented image.
Step 6: Classification of disease using multi-class SVM.

21
 1) Collection of Sample

We have collected four type diseased affected orange sample and healthy orange sample from
the database of University of California Agriculture and Natural Resources. In Fig.2 show the
four type of diseased orange i.e. Citrus Canker, Stubborn, Brown spot and Melanoses.

(a) (b) (c) (d)

Fig.2. Example of different types of orange fruit diseases (a) Citrus canker, (b) stubborn,
(c) Brown rot, (d) melanoses

2) Preprocessing

Image preprocessing is a prophase relative to feature extraction& image reorganization. The

image which has input is always not satisfactory regardless of what image acquisition devices
are adopted. Here preprocessing phase consists of two steps: Image Enhancement and Lab
Color Transformation. Image enhancement is one of the most interesting and important issues
in image processing field. The main purpose of image enhancement is to bring out details that
are hidden in an image or to increase the contrast in a low contrast image. Image
enhancement produces an output image that subjectively looks better than the original image
by changing the pixel’s intensity of the input image. We use L*a*b* color space because the
color information in the L*a*b* color space is stored in only two channels (i.e.; a* & b*
component), and it causes reduced processing time for the defect segmentation.

22
 3) Segmentation

Image segmentation is the classification of an image into different groups. K-means

clustering algorithm is an unsupervised algorithm and it is used to segment the interest area
from the background. Here we have taken the number ofclusters, K=3 to separate
background, foreground& the affected area.

4) Feature Extraction

It involves simplifying the number of resources required to describe a large set of data
accurately. When performing analysis of complex data one of the major problem stems from
the number of variables generally requires a large amount of memory & computation power
or a classification algorithm which overfits the training sample and generalizes poorly to
anew sample. Thirteen number of texture are extracted from GLCM i.e.; contrast,
correlation, energy, homogeneity, mean, standard deviation, entropy, RMS, variance,
smoothness, kurtosis, skewness, IDM.

5) Classification

Support vector machines are supervised learning models with an associated learning
algorithm that analyze data used for classification and regression analysis. Given a set of
training example, each marked as belonging to one or the other of two categories, module an
SVM training algorithm builds a model that assigns new example to one category or the
other, making it a non- probabilistic binary linear classifier. Multi-class SVM classifies
different types of Orange disease by considering the features of thetrained image and test
sample image.

23
 B. DISEASE SEVERITY

Fuzzy Logic is considered one of the approaches/techniques for Artificial Intelligence, where
the intelligent behavior is achieved by creating fuzzy classes of some parameters. Fuzzy rule-
based model as Fig. 3 have a simple structure & consist of four major components: 1) a
Fuzzification module, which translates crisp inputs (classical measurements) into fuzzy
values through linguistic variable’s: 2) an if-then fuzzy rule base, which consists of a set of
conditioned fuzzy propositions; 3) an inference method, which applies fuzzy reasoning
mechanism to obtain outputs i.e., carries out the computation using fuzzy rules); and 4)
Defuzzification

FUZZY RULE BASE

I/P O/P
DECISION
FUZZIFICATION MAKING DEFUZZIFICATION
LOGIC

MEMBERSHIP FUNCTION

Fig. 3 Fuzzy Logic System OverView.

24
 Chapter 4

Result and Discussion

In the proposed algorithm, first of all, feature extraction process has occurred. During this
process, 13 number of features have been extracted. By considering those features we can
classify the disease of orange fruits. We have experimented 20 number of diseased orange
and out of 20 number, only 2 of them are misidentified by the proposed algorithm. The
observation is shown in Table.1.

Sample Disease Detected Identification

image type diseased
type using
algorithm
1 CC CC √
2 SD H ×
3 SD SD √
4 BR H ×
5 CC CC √
6 BR BR √
7 BR BR √
8 SD SD √
9 H H √
10 H H √
11 CC CC √
12 M M √
13 M M √
14 CC CC √
15 H H √
16 M M √
17 M M √
18 CC CC √
19 SD SD √
20 BR BR √
H: Healthy, CC: Citrus Canker, M: Melanoses,
BR: Brown Rot, SD: Stubborn disease.

25
 Table.1-Classification of Diseased Orange.

Again this work proposed a fuzzy logic based model to measure the severity of disease in
orange and validated using four type of diseased orange. The model estimated the flaw mark on
the skin of orange and simulated disease severity based on population dynamics, as show Fig.4-7.

(a) (b) (c) (d)

Fig.4 (a) Citrus Canker Orange (b) Clustered image, (c) Binary image (d) Severity of Disease
of 0.9.

(a) (b) (c) (d)

Fig.5 (a) stubborn orange (b) Clustered image, (c) Binary image (d) Severity of disease
0.85.

26
 (a) (b) (c) (d)

Fig.6 (a) Brown rot orange, (b) clustered image, (c) Binary image (d) Severity of disease
1.25

(a) (b) (c) (d)

Fig.7 (a) Melanose orange (b) Clustered image, (c) Binary image (d) Severity of disease 1.25.

27
 Chapter 5

Conclusion and Future Scope

In this work, we successfully classify the different type of disease with 90% accuracy and
also compute the disease severity of four type of diseased orange successfully. This research
can be extended to increase the accuracy using soft computing technique and validate with
more number of sample.

28
 REFERENCES

[1] Agriculture in India information about Indian Agriculture & Its Importance. (Online)
https://www.ibef.org/industry/agriculture-india.aspx Date: 18.06.2017.
[2] Agriculture in India (Online) https://en.m.wikipedia.org/wiki/Agriculture_in_India.
Date:18.06.2017
[3] Barbedo, Jayme Garcia Arnal (2013). Digital Image Processing Techniques for Detecting,
Quantifying and Classifying Plant Diseases, SpringerPlus.
[4] Prabira Kumar Sethy, Baishalee Negi, and Nilamani Bhoi. Detection of Healthy and
Defected Diseased Leaf of Rice Crop using K-Means Clustering Technique. International
Journal of Computer Applications 157(1):24-27, January 2017.
[5] Prabira Kumar Sethy,Chhavi Kumari,N.K. Barpanda, Baishalee Negi, Santi Kumari
Behera, & A.K.Rath (2017). Identification of Mineral Deficiency in Rice Crop-based on
SVM in Approach of K-Means & Fuzzy C-Means Clustering. Helix, 7(5), 1970-1983.
[6] Prabira Kumar Sethy, Baishalee Negi, Santi Kumari Behera, N.K. Barpanda, & A.K.Rath
(2018) Measurement of Disease Severity of Rice Crop Using Machine Learning and
Computational Intelligence. In: Cognitive Science and Artificial Intelligence.
SpringerBriefs in Applied Sciences and Technology. Springer, Singapore.
[7] Prabira Kumar Sethy, Shwetapadma Panda, Santi Kumari Behera & A.K.Rath (2017). On
Tree Detection, Counting & Post- Harvest grading of fruits Based on Image Processing
and Machine Learning Approach-A Review. International Journal of Engineering and
Technology (IJET), vol.9(2),pp. 649-663. DOI: 10.21817/ijet/2017/v9i2/170902058.
[8] Prabira Kumar Sethy, Baishalee Negi, Santi Kumari Behera,N.K.Barpanda &
A.K.Rath.(2017).An Image Processing Approach for Detection, Quantification, and
Identification of Plant Leaf Diseases -A Review. International Journal of Engineering and
Technology (IJET),vol.9(2),pp.635-648.DOI: 10.21817/ijet/2017/v9i2/170902059.
[9] Swetapadma Panda, Prabira Kumar Sethy. (2017). Post- Harvesting Grading of carica
papaya fruit using image segmentation and soft computing.International Journal
ofAdvanced Research in Computer Science; Jul/Aug2017, Vol.8 Issue 7, p692-697.

29
[10] Xiaoling Deng et al. “Citrus greening detection using visible spectrum imaging and C-
SVC” Computers and Electronics in Agriculture, journal homepage:
www.elsevier.com/locate/compag.
[11] Xuebing Bai et al., “A fuzzy clustering segmentation method based on neighborhood
grayscale information for defining cucumber leaf spot disease images” Computers and
Electronics in Agriculture, journal homepage:www.elsevier.com/locate/compag.
[12] Anand Singh Jalal, Shiv Ram Dubey “Detection and Classification of Apple Fruit
Diseases Using Complete Local Binary Patterns” IEEE Third International Conference on
Computer and Communication Technology, pp. 978-0-7695-4872, 2012.
[13] Monika Jhuria, Rushikesh borse, Ashwani Kumar “ImageProcessing for Smart Farming:
Detection of Disease and Fruit Grading” Proceeding of the IEEE Second International
Conference on Image Information Processing, pp. 978-1-4673-6101, 2013.
[14] Mrunmayee Dhakate, Ingole A.B. “Diagnosis of Pomegranate Plant Diseases using
Neural Network” IEEE pp.978-1-46738564, 2015.
[15] Ridhuna Rajan Nair, Swapnal Subhash Adsul, Namrata Vitthal Khabale,Vrushali Sanjay
Kawade “Analysis and Detection of Infected Fruit Part Using Improved k-means
Clustering and Segmentation Techniques” IOSR Journal of Computer Engineering
(IOSR-JCE), pp. 3741, 2015.
[16] Ashwini Awate, Damini Deshmankar, Prof. Samadhan Sonavane “Fruit Disease
Detection using Color, Texture Analysis and ANN” IEEE International Conference on
GreencComputing and Internet of Things (ICGCIoT), pp. 978-1-4673-7910, 2015.
[17] Pujitha N, Swathi C, Kanchana V “Detection Of External Defects On Mango”
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11,
Number 7, pp. 4763-4769 2016.
[18] Bhavini J. Samajpati, Sheshang D. Degadwala “Hybrid Approach for Apple Fruit
Diseases Detection and Classification Using Random Forest Classifier” IEEE
International Conference on Communication and Signal Processing, pp. 978-50900396,
2016.

30
31