Accepted Manuscript

Detection of Plant Leaf Diseases Using Image Segmentation and Soft Comput-
ing Techniques

Vijai Singh, A.K. Misra

PII: S2214-3173(16)30015-4
DOI: http://dx.doi.org/10.1016/j.inpa.2016.10.005
Reference: INPA 67

To appear in: Information Processing in Agriculture

Received Date: 2 March 2016

Revised Date: 1 October 2016
Accepted Date: 31 October 2016

Please cite this article as: V. Singh, A.K. Misra, Detection of Plant Leaf Diseases Using Image Segmentation and
Soft Computing Techniques, Information Processing in Agriculture (2016), doi: http://dx.doi.org/10.1016/j.inpa.
2016.10.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Detection of Plant Leaf Diseases Using Image Segmentation and Soft

Computing Techniques

Vijai Singh1, Prof A. K. Misra2

1
Computer Science Department,

IMS Engineering College, Ghaziabad, U.P., India

E- mail : [email protected]
2
Professor Computer Science & Engg. Department

MNNIT Allahabad UP, India

Abstract

Agricultural productivity is something on which Economy highly depends. This is the one of

the reasons that disease detection in plants plays an important role in agriculture field, as having

disease in plants are quite natural. If proper care is not taken in this area then it causes serious

effects on plants and due to which respective product quality, quantity or productivity is affected.

For instance a disease named little leaf disease is a hazardous disease found in pine trees in

United States. Detection of plant disease through some automatic technique is beneficial as it

reduces a large work of monitoring in big farms of crops, and at very early stage itself it detects

the symptoms of diseases i.e. when they appear on plant leaves. This paper presents an algorithm

for image segmentation technique which is used for automatic detection and classification of

plant leaf diseases .It also covers survey on different diseases classification techniques that can be

used for plant leaf disease detection. Image segmentation, which is an important aspect for

disease detection in plant leaf disease, is done by using genetic algorithm.

Keywords: Image processing; Genetic algorithm; plant disease detection; classification

 1. Introduction

The agricultural land mass is more than just being a feeding sourcing in today's world. Indian

economy is highly dependent of agricultural productivity. Therefore in field of

agriculture, detection of disease in plants plays an important role. To detect a plant

disease in very initial stage, use of automatic disease detection technique is beneficial. For

instance a disease named little leaf disease is a hazardous disease found in pine trees in United

States. The affected tree has a stunted growth and dies within 6 years. Its impact is found in

Alabama, Georgia parts of Southern US. In such scenarios early detection could have been

fruitful.

The existing method for plant disease detection is simply naked eye observation by experts

through which identification and detection of plant diseases is done. For doing so, a large team of

experts as well as continuous monitoring of plant is required, which costs very high when we do

with large farms. At the same time, in some countries, farmers do not have proper facilities or

even idea that they can contact to experts. Due to which consulting experts even cost high as well

as time consuming too. In such conditions, the suggested technique proves to be beneficial in

monitoring large fields of crops. Automatic detection of the diseases by just seeing the symptoms

on the plant leaves makes it easier as well as cheaper. This also supports machine vision to

provide image based automatic process control, inspection, and robot guidance [2][4][5].

Plant disease identification by visual way is more laborious task and at the same time, less

accurate and can be done only in limited areas. Whereas if automatic detection technique is used

it will take less efforts, less time and become more accurate. In plants, some general diseases seen

are brown and yellow spots, early and late scorch, and others are fungal, viral and bacterial

diseases. Image processing is used for measuring affected area of disease and to determine the

difference in the color of the affected area [1][2][6].

 Image segmentation is the process of separating or grouping an image into different parts.

There are currently many different ways of performing image segmentation, ranging from the

simple thresholding method to advanced color image segmentation methods. These parts

normally correspond to something that humans can easily separate and view as individual objects.

Computers have no means of intelligently recognizing objects, and so many different methods

have been developed in order to segment images. The segmentation process is based on various

features found in the image. This might be color information, boundaries or segment of an image

[11][13].We use Genetic algorithm for color image segmentation.

Evolutionary computing was first introduced in the 1960s by I. Rechenberg. His idea was then

taken forward by other researchers. Sometimes evolutionary changes are small and appear

insignificant at first glance, but they play a part in natural selection and the survival of the

species. Examples of natural selections are

1. The warrior ants in Africa are probably one of the most impressive examples of

adaptation. Within any single colony, ants emit a chemical signal that lets the

others know they all belong to the same compound. Or, put more simply, a

signal that says, "Don't attack me, we're all family." However, warrior ants have

learned how to imitate the signal from a different colony. So if a group of

warrior ants attacks a colony, they will be able to imitate that colony's signal. As

a result, the workers in the colony will continue on, now under the direction of

new masters, without ever realizing an invasion has taken place.

2. All rat snakes have similar diets, are excellent climbers and kill by constriction.

They all have the same reaction when startled (they remain motionless), and will
 avoid confrontation whenever possible. Some will bite if threatened, although

they are non-venomous. However, rat snakes come in a wide variety of colours,

from yellow striped to black to orange to greenish. This is because rat snakes are

found all over the Eastern and Midwestern states, and are subjected to all types

of weather and terrain. Rat snakes are common in urban areas, but they can also

be found in wooded areas, mountains or coastal regions. As a result, rat snakes

have had to adapt to their local environments in an effort to avoid detection and

hunt more effectively.

Genetic algorithms belong to the evolutionary algorithms which generate solutions for

optimization problems. Algorithm begins with a set of solutions called population. Solutions from

one population are chosen and then used to form a new population. This is done with the

anticipation, that the new population will be enhanced than the old one. Solutions which are

selected to form new solutions (offspring) are chosen according to their fitness - the more

appropriate they are, the more probability they have to reproduce [12][14].

Some advantages of Genetic Algorithm are

• Genetic Algorithm optimizes both variables efficiently, continuous or discrete.

• It searches from a large sampling of the cost surface.

• Large number of variables can be processed at the same time.

• It can optimize variables with highly complex cost surfaces.

• Gives a number of optimum solutions, not a single solution. So different image segmentation

results can be obtained at the same time

The basic steps of Genetic Algorithm are as follows:

1) [Start] Generate random population of n chromosomes (suitable solutions for the

problem)
 2) [Fitness] Evaluate the fitness f(x) of each chromosome x in the population

3) [New population] Create a new population by repeating following steps until the new

population is complete

(a) [Selection] Select two parent chromosomes from a population

according to their fitness (the better fitness, the bigger chance to be

selected)

(b) [Crossover] With a crossover probability cross over the parents to

form a new offspring (children). If no crossover was performed,

offspring is an exact copy of parents.

(c) [Mutation] With a mutation probability mutate new offspring at each

locus (position in chromosome).

(d) [Accepting] Place new offspring in a new population

4) [Replace] Use new generated population for a further run of algorithm

5) [Test] If the end condition is satisfied, stop, and return the best solution in current

population

6) [Loop] Go to step 2

2. Literature Review

Savita N. Ghaiwat et al. presents survey on different classification techniques that can be

used for plant leaf disease classification. For given test example, k-nearest-neighbor method is

seems to be suitable as well as simplest of all algorithms for class prediction. If training data is

not linearly separable then it is difficult to determine optimal parameters in SVM, which appears

as one of its drawbacks [1].

 Authors in paper [2] describe that there are mainly four steps in developed processing

scheme, out of which, first one is, for the input RGB image, a color transformation structure is

created, because this RGB is used for color generation and transformed or converted image of

RGB, that is, HSI is used for color descriptor. In second step, by using threshold value, green

pixels are masked and removed. In third, by using pre-computed threshold level, removing of

green pixels and masking is done for the useful segments that are extracted first in this step, while

image is segmented. And in last or fourth main step the segmentation is done.

Mrunalini R et al. [3] presents the technique to classify and identify the different disease

through which plants are affected. In Indian Economy a Machine learning based recognition

system will prove to be very useful as it saves efforts, money and time too. The approach given in

this for feature set extraction is the Color Co-occurrence Method. For automatic detection of

diseases in leaves, neural networks are used. The approach proposed can significantly support an

accurate detection of leaf, and seems to be important approach, in case of steam, and root

diseases, putting fewer efforts in computation.

According to Paper [4] disease identification process include some steps out of which four

main steps are as follows: first, for the input RGB image, a color transformation structure is

taken, and then using a specific threshold value, the green pixels are masked and removed, which

is further followed by segmentation process, and for getting useful segments the texture statistics

are computed. At last, classifier is used for the features that are extracted to classify the disease.

The robustness of the proposed algorithm is proved by using experimental results of about 500

plant leaves in a database.

 Anand H. Kulkarni et al. presents a methodology for early and accurately plant diseases

detection, using artificial neural network (ANN) and diverse image processing techniques. As the

proposed approach is based on ANN classifier for classification and Gabor filter for feature

extraction, it gives better results with a recognition rate of up to 91%. An ANN based classifier

classifies different plant diseases and uses the combination of textures, color and features to

recognize those diseases [5].

Table 1.1 Analysis of Various Algorithms

Authors & Year Goals Future perspective

[1] Savita N. Ghaiwat et al , Review of ANN, SVM, In neural network it’s difficult

Detection and Classification of PNN,SELF ORG MAPS to understand structure of

Plant Leaf Diseases Using Image AND FUZZY LOGIC algorithm and to determine

processing Techniques: A Review optimal parameters when

(2014) training data is not linearly

separable

[2] Prof. Sanjay B et al , vision-based detection NN’s can be used to increase

Agricultural plant Leaf Disease algorithm with Masking the recognition rate of

Detection Using Image the green-pixels and Color classification process

Processing(2013) co-occurrence Method

[3]Mrunalini R et al. , An K-means clustering Artificial Neural Network and

Application of K-Means algorithm with Neural Fuzzy Logic with other soft

Clustering and Artificial networks for automatic computing technique can

Intelligence in Pattern detection of leaves be used to classify the crop

Recognition for Crop Diseases diseases diseases

(2011)

[4] S. Arivazhagan et al , Color co-occurrence The training samples can be

Detection of unhealthy region of method with SVM increased and shape feature

plant leaves and classification of classifier and color feature along with

plant leaf diseases using texture the optimal features can be

features given as input

(2013) condition of disease

identification

[5] Anand H. Kulkarni et al Gabor filter for feature Recognition rate can be

,Applying image processing extraction and ANN increased

technique to detect plant diseases classifier for classification

(2012)

[6] Sabah Bashir et al ,Remote Texture segmentation by Bayes classifier, K-means

Area Plant Disease Detection co-occurrence matrix clustering and principal

Using Image Processing (2012) method and K-means component classifier can be

Clustering Technique used to classify various plant

diseases.

[7]Smita Naikwadi et al , The color co-occurrence Better result of detection can

ADVANCES IN IMAGE texture analysis method be obtained with the large

PROCESSING FOR was developed through the database and advance feature

DETECTION OF PLANT use of Spatial Gray-level of color extraction

DISEASES (2013) Dependence

Matrices
[8] Sanjay B. Patil et al , LEAF Simple threshold and Nil

DISEASE SEVERITY Triangle thresholding

MEASUREMENT USING segmentation methods

IMAGE PROCESSING (2011)

[9] Piyush Chaudhary et al , Color Median filter is used for Disease spot area can be

Transform Based Approach for image smoothing and computed for assessment of

Disease Spot Detection on Plant threshold can be loss in agriculture crop.

Leaf (2012) calculated by applying Disease can be classified by

Otsu method calculating dimensions of

disease spot.

[10]Arti N. Rathod et al, Image survey of different development of hybrid

Processing Techniques for techniques for leaf disease algorithms & neural networks

Detection of Leaf Disease (2013) detection in order to increase the

recognition rate of final

classification process

Authors present disease detection in Malus domestica through an effective method like K-mean

clustering, texture and color analysis [6]. To classify and recognize different agriculture, it uses

the texture and color features those generally appear in normal and affected areas. In coming

days, for the purpose of classification K-means clustering, Bayes classifier and principal

component classifier can also be used.

According to [7] histogram matching is used to identify plant disease. In plants, disease appears

on leaf therefore the histogram matching is done on the basis of edge detection technique and

color feature. Layers separation technique is used for the training process which includes the

training of these samples which separate the layers of RGB image into red, green, and blue layers

and edge detection technique which detecting edges of the layered images. Spatial Gray-level

Dependence Matrices are used for developing the color co-occurrence texture analysis method.

Analysis of various algorithms is given in table 1.1.

Paper [8] presents the Triangle threshold and simple threshold methods. These methods are

used to lesion region area and segment the leaf area respectively. In final step, categorization of

disease is done by calculating the quotient of leaf area and lesion area. According to the research

done, the given method is fast and accurate for calculating leaf disease severity and leaf area

calculation is done by using threshold segmentation.

Authors describe an algorithm for disease spot segmentation in plant leaf using image processing

techniques [9]. In this paper, process of disease spot detection is done by comparing the effect of

HSI, CIELAB, and YCbCr color space. For Image soothing Median filter is used. In final step, by

applying Otsu method on color component, calculation of threshold can be done to find the

disease spot. There is some noise because of background which is shown in the experimental

result, camera flash and vein. CIELAB color model is used to remove that noise.

The state of art review of different methods for leaf disease detection using image processing

techniques is presented in paper [10]. The existing methods studies are for increasing throughput

and reduction subjectiveness which comes due to naked eye observation through which

identification and detection of plant diseases is done.

According to [15] soft computing methods such as artificial neural networks (ANN), genetic

programming, and fuzzy logic can be used as an alternative method for modeling complex

behavior of materials such as graphene. These algorithms require input training data for solving

problems. These computing methods generate meaningful solutions for complicated optimization

problems based on the input. In many models feed-forward network of three layers can be used.

Root-mean-square error method can be used to determine the number of neurons in hidden layer.

Tabu search is the meta heuristic search method which uses local search techniques used for

mathematical optimization. Local searches pick up a solution to a problem which is potential and

checks its immediate neighbours i.e. those solutions which are similar except for some minor

details, with an intention of finding a better solution. Local search methods tend to get stuck in

suboptimal regions or on plateaus where many solutions are equally fit. If any solution which is

potential has been previously visited within a short duration or if it doesn’t satisfies a rule, then it

is marked as “tabu”. By doing so, the algorithm doesn’t consider that possibility repeated [24].

Genetic algorithms were used to evolve programs to perform certain tasks by John Koza

in 1992. His method was known as "genetic programming" (GP). Genetic programming is

considered to be the most famous for solving symbolic regression problems and is widely used

for solving optimization problem. The working principle behind GP and GA are same but there

lies a major difference between the two that GP gives solutions in terms of weighted sum of

coefficients, whereas GA gives solutions represented by a number in binary or real form. Thus,

we can say that GP is a structure optimization method whereas GA is a parameter optimization

method. MGGP is the Genetic programming in which evolutionary stage is a combined set of

several trees which are regressed using least squares method. We can use trial-and-error method

for the effective implementation of MGGP. [16] [17][19]

V. Vijayaraghavan et al in their work [18] stated that a support vector machine is a very potential

AI method and can apply extensively to solve classification problems. The SVM which is used to

solve regression problems is known as support vector regression (SVR). SVR is very popular

among researchers for providing generalization ability to the solution model.

The manifestation of pathogens in plantations is the one of the most important cause of losses in

many crops. Bernardes, Alexandre A. (et al.) give the method of the automatic classification of

cotton diseases based on the feature extraction of foliar symptoms from digital images. For the

feature extraction this method uses the energy of the wavelet transform and a SVM for the actual

classification [20].

Zhen Ma, J M R. S. Tavares (et al,) [21] makes a review on the current segmentation algorithms

used for medical images. Algorithms mainly categories in three categories according to their main

concepts: the first based on threshold, the second based on pattern recognition techniques and

third one based on deformable models. In recent years the third category of algorithms are

focused on deformable models as a result of intensive investigation. Some of the main

applications of these algorithms are segmenting organs and tissues in pelvic cavity area. These

are discussed through several preliminary experiments.

Authors in paper [22], classified an algorithm on the basis of the principal methodologies.

Algorithms of each category are discussed and the important ideas, application fields, advantages

and disadvantages of each category are summarized. Experiments that use these algorithms to

segment the tissues and organs of the female pelvic cavity are to show their unique

characteristics. In the last, the important guidelines for designing the segmentation algorithms of

the pelvic cavity are proposed.

J. M. R. S. Tavares describes that the computational analysis of images is challenging due to tasks

such as segmentation, extraction of representative features, matching, alignment, tracking, motion

analysis, deformation estimation, and 3D reconstruction. In paper [23], the methods for

processing and analyzing objects in images and their use in applications like medicine,

biomechanics, engineering and materials sciences are proposed.

Limitation of existing work:

• The implementation still lacks in accuracy of result in some cases. More optimization is

needed.

• Priori information is needed for segmentation

• Database extension is needed in order to reach the more accuracy.

• Very few diseases have been covered. So, work needs to be extended to cover more

diseases.

• The possible reasons that can lead to misclassifications can be as follows: disease

symptoms varies from one plant to another, features optimization is needed , more

training samples are needed in order to cover more cases and to predict the disease more

accurately.

To remove these research gaps a new methodology for automatic detection as well as

classification of plant leaf diseases using image segmentation has been proposed. The advantages

of proposed algorithm are as follows:

1. Use of estimators for automatic Initialization of cluster centers so there is no need of user input

at the time of segmentation

2. The detection accuracy is enhanced with proposed algorithm

3. Proposed method is fully automatic while existing methods require user input to select the best

segmentation of input image.

4. It also provides environment friendly recovery measures of the identified disease.

3. Proposed Methodology

Digital camera or similar devices are use to take images of leafs of different types, and then

those are used to identify the affected area in leafs. Then different types of image-processing

techniques are applied on them, to process those images, to get different and useful features

needed for the purpose of analyzing later.

Algorithm written below illustrated the step by step approach for the proposed image recognition

and segmentation processes:

1) Image acquisition is the very first step that requires capturing an image with the help of a

digital camera.

2) Preprocessing of input image to improve the quality of image and to remove the undesired

distortion from the image. Clipping of the leaf image is performed to get the interested image

region and then image smoothing is done using the smoothing filter. To increase the contrast

Image enhancement is also done.

3) Mostly green colored pixels, in this step, are masked. In this, we computed a threshold value

that is used for these pixels. Then in the following way mostly green pixels are masked: if pixel

intensity of the green component is less than the pre-computed threshold value, then zero value is

assigned to the red, green and blue components of the this pixel.
4) In the infected clusters, inside the boundaries, remove the masked cells.

5) Obtain the useful segments to classify the leaf diseases. Segment the components using genetic

algorithm

For doing clustering appropriately, the search capability of GAs can be used, to set of

unlabeled points in N-dimension into K clusters. On image data, we have applied the same idea in

our proposed scheme. We have taken a color image of size m x n and every pixel has Red, Green

and Blue components. Every chromosome shows a solution, which is a sequence of K cluster

centers. Population is initialized in various rounds randomly and from existing chromosome best

chromosome survives in each round for the next round processing.

In the first step of fitness computation the dataset of pixel is clustered according to nearest

respective cluster centers such that each pixel xi of color image is put into the respective cluster

with cluster center zj for j = 1, 2, ..., K by the following equations

if || xi - zj|| < || xi – zl||,

i = 1, 2, ..., m x n ,l = 1,2,...,K, and p ≠ j.

In the further step new cluster centres are obtained by calculating the mean of each pixel of

the assigned clusters. The new centre of cluster Zi is given by for the cluster Ci as:

 , ,  =  , ,   = , , … .  (1)

  ∈
 Now the fitness function is computed by calculating Euclidean distance between the

pixels and their respective cluster by using following equations

 = ∑  (2)

 = |  , ,  −  , ,  | (3)

 ∈

6) Computing the features using color co-occurrence methodology

For feature extraction the method used is color co-occurrence method. It is the

methodology in which both the texture and color of an image are considered, to come to the

unique features, which shows that image.

Over the traditional gray-scale representation, in the visible light spectrum, the use of color

image features provides an additional feature for image characteristic. There are three major

mathematical processes in the color co-occurrence method. First, conversion of the RGB images

of leaves is done into HIS color space representation. After completion of this process, to

generate a color co-occurrence matrix, each pixel map is used, which results into three color co-

occurrence matrices, one for each of H, S, I.

Features called as texture features, which include Local homogeneity, contrast, cluster shade,

Energy, and cluster prominence are computed for the H image as given in Equations (4) to

(7).

CONTRAST = ∑!"#
$,%&' i, j Ci, j

(4)
 Energy = ∑!"#
$,%&' Ci. j

(5)

Local Homogeneity = ∑!"#

(,%&' Ci, j / 1 + i − j 

(6)

Entropy = − ∑!"#
$,%&' Ci. jlog Ci. j (7)

7) Classification of disease

In this phase of classification, extraction and comparison of the co-occurrence features for the

leaves with the corresponding feature values are stored in the feature dataset. First, the Minimum

Distance Criterion and then SVM classifier are used to done the classification. The measurement

of success of classification is done by using the classification gain and following Equation (8) is

used for calculation:

Gain (%) = number of correct classification /Total no of test images * 100 (8)

4. Results

All the experiments are performed in MATLAB. For input data disease, samples of plant

leaves like rose with bacterial disease, beans leaf with bacterial disease, lemon leaf with Sun burn

disease, banana leaf with early scorch disease and fungal disease in beans leaf are considered. Fig

1 shows the original images which are followed by output segmented images. Segmented image

can be classified into different plant diseases. Fig 2 shows the input and output image where input

image is a banana leaf with early scorch disease and output image shows the classification of

disease using feature extraction method.

Fig.1 Input and Output Images
Fig. 2 Input and Output Image of banana leaf and output Diseases is early scorch disease

In the same manner classification of diseases of other input plant leafs are shown in Fig
3, Fig 4, Fig 5 and Fig 6.

Fig. 3 Input and Output Image of beans leaf and output Diseases is bacterial leaf spot

Fig. 4 Input and Output Image of rose leaf and output Diseases is bacterial leaf spot
Fig. 5 Input and Output Image of lemon leaf and output Diseases is Sun burn disease

Fig. 6 Input and Output Image of beans leaf and output Diseases is fungal disease

The co-occurrence features are calculated after mapping the R, G, B components of the input

image to the thresholded images. The co-occurrence features for the leaves are extracted and

compared with the corresponding feature values that are stored in the feature library. The

classification is first done using the Minimum Distance Criterion with K-Mean Clustering and

shows its efficiency with accuracy of 86.54%. The detection accuracy is improved to 93.63% by

proposed algorithm. In the second phase classification is done using SVM classifier and shows its

efficiency with accuracy of 95.71%. Now the detection accuracy is improved to 95.71% by SVM

with proposed algorithm. The training and the testing sets for each type of leaf along with

their detection accuracy is shown in Table 1.2 and figure 7. From the results it can be
seen that the detection accuracy is enhanced by SVM with proposed algorithm compared

to other approaches reported in [4],[5] and [7].

Table 1.2 Comparison of results

Disease No. of images No. of images Detection accuracy/%

samples used used MDC with MDC with SVM with

for training for testing Kmean Proposed Proposed

algorithm algorithm

Banana 15 10 80.00 90.00 90.00

Beans 15 14 92.85 92.85 92.85

Lemon 15 10 90.00 100.00 100

Rose 15 12 83.33 91.66 100

Overall accuracy 86.54 93.63 95.71

 120

100

80

MDC with Kmean

60

MDC with Proposed

40 algorithm

20

0
Banana Beans Lemon Rose Overall
accuracy

Fig. 7 Comparison of results

The numbers of leaf disease samples that were classified into five classes of leaf disease using

proposed algorithm are shown in Table 1.3 and Figure 8. From the results it can be seen that only

few samples from Frog eye leaf spot and bacterial leaf spot leaves were misclassified. Only two

leafs with bacterial leaf spot disease are classified as frog eye leaf spot and one frog eye leaf spot

is classify as bacterial leaf spot. The average accuracy of classification of proposed algorithm is

97.6 compared to 92.7 reported in [25].

 Frog
bacterial Sun burn
eye fungal early
leaf disease leaf spot disease Accuracy
leaf disease scorch
spot
bacterial leaf
23 2 0 0 0 92
spot
Frog eye leaf
1 24 0 0 0 96
spot
Sun burn
0 0 25 0 0 100
disease
fungal
0 0 0 25 0 100
disease
early scorch 0 0 0 0 25 100
Average 97.6

Table 1.3 Classification results per class for Proposed Method

Accuracy
102
100
98
96
94 Accuracy
92
90
88
bacterial Frog eye Sun burn fungal early Average
leaf spot leaf spot disease disease scorch

Fig. 8 Classification results per class for Proposed Method

5. Conclusion

This paper presents the survey on different diseases classification techniques used for plant leaf

disease detection and an algorithm for image segmentation technique that can be used for

automatic detection as well as classification of plant leaf diseases later. Banana, beans, jackfruit,
lemon, mango, potato, tomato, and sapota are some of those ten species on which proposed

algorithm is tested. Therefore, related diseases for these plants were taken for identification. With

very less computational efforts the optimum results were obtained, which also shows the

efficiency of proposed algorithm in recognition and classification of the leaf diseases. Another

advantage of using this method is that the plant diseases can be identified at early stage or the

initial stage. To improve recognition rate in classification process Artificial Neural Network,

Bayes classifier, Fuzzy Logic and hybrid algorithms can also be used.

References

[1] Savita N. Ghaiwat, Parul Arora. Detection and Classification of Plant Leaf Diseases Using Image processing Techniques: A Review.

International Journal of Recent Advances in Engineering & Technology. ISSN (Online): 2347 - 2812, Volume-2, Issue - 3, 2014

[2] Prof. Sanjay B. Dhaygude, Mr.Nitin P.Kumbhar. Agricultural plant Leaf Disease Detection Using Image Processing. International

Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering. Vol. 2, Issue 1, January 2013

[3] Mrunalini R. Badnakhe and Prashant R. Deshmukh. An Application of K-Means Clustering and Artificial Intelligence in Pattern

Recognition for Crop Diseases. International Conference on Advancements in Information Technology. 2011 IPCSIT vol.20 (2011)

[4] S. Arivazhagan, R. Newlin Shebiah, S. Ananthi, S. Vishnu Varthini. Detection of unhealthy region of plant leaves and classification of

plant leaf diseases using texture features. Agric Eng Int: CIGR Journal, 15(1): 211－217 2013

[5] Anand.H.Kulkarni, Ashwin Patil R. K. Applying image processing technique to detect plant diseases. International Journal of Modern

Engineering Research. Vol.2, Issue.5, Sep-Oct. 2012 pp-3661-3664

[6] Sabah Bashir, Navdeep Sharma. Remote Area Plant Disease Detection Using Image Processing. IOSR Journal of Electronics and

Communication Engineering. ISSN : 2278-2834 Volume 2, Issue 6 2012, PP 31-34

[7] Smita Naikwadi, Niket Amoda. ADVANCES IN IMAGE PROCESSING FOR DETECTION OF PLANT DISEASES. International

Journal of Application or Innovation in Engineering & Management. Volume 2, Issue 11, November 2013

[8] Sanjay B. Patil et al. LEAF DISEASE SEVERITY MEASUREMENT USING IMAGE PROCESSING. International Journal of

Engineering and Technology. Vol.3 (5), 2011, 297-301

[9] Piyush Chaudhary et al. Color Transform Based Approach for Disease Spot Detection on Plant Leaf. International Journal of Computer

Science and Telecommunications. ,Volume 3, Issue 6, June 2012

[10] Arti N. Rathod, Bhavesh Tanawal, Vatsal Shah. Image Processing Techniques for Detection of Leaf Disease. International Journal of

Advanced Research in Computer Science and Software Engineering. Volume 3, Issue 11, November 2013

[11] S.Beucher, F.Meyer. The morphological approach to segmentation: The watershed transforms. Mathematical Morphology Image

Processing, E. R. Dougherty, Ed. New York Marcel Dekker, 1993, vol. 12, pp. 433–481.
[12] B. Bhanu, S. Lee, J. Ming. Adaptive image segmentation using a genetic algorithm. In IEEE Transactions on Systems, Man and

Cybernetics, volume 25, pages 1543–1567, Dec 1995.

[13] B. Bhanu, J. Peng. Adaptive integrated image segmentation and object recognition. In IEEE Transactions on Systems, Man and Cy-

bernetics, Part C. volume 30, pages 427–441, November 2000.

[14] Keri Woods. Genetic Algorithms: Colour Image Segmentation Literature Review. July 24, 2007.

[15] Venkatesh Vijayaraghavan, Akhil Garg, Chee How Wong, Kang Tai1 and Yogesh Bhalerao. Predicting the mechanical characteristics of

hydrogen functionalized graphene sheets using artificial neural network approach. Journal Of Nanostructure in Chemistry 2013, 3:83

[16] Akhil Garg, Ankit Garg and K. Tai .A multi-gene genetic programming model for estimating stress-dependent soil water retention

curves. Computational Geosciences,2014:p.1-12.

[17] Akhil Garg, AnkitGarg , K.Tai and S.Sreedeep. An integrated SRM-multi-gene genetic programming approach for prediction of factor

of safety of 3-D soil nailed slopes. Engineering Applications of Artificial Intelligence, 30(2014)30–40.

[18] V. Vijayaraghavan, A. Garg, C. H. Wong and K. Tai. Estimation of mechanical properties of nanomaterials using artificial intelligence

methods. Applied Physics A, 2013: p.1-9.

[19] A. Garg, V. Vijayaraghavan, C.H. Wonga and K. Tai. Measurement of properties of graphene sheets subjected to drilling operation

using computer simulation. Measurement, 2014.

[20] Bernardes, Alexandre A. (et al.) 2013, “Identification of Foliar Diseases in Cotton Crop”. In: Tavares, Joao Manuel R.S., Natal Jorge,

Renato, Topics in Medical Image Processing and Computational Vision, Book Series: Lecture Notes in Computational Vision and

Biomechanics, Vol. 8, pp. 67-85, April 2013

[21] Zhen Ma, J M R. S. Tavares, R. M. Natal Jorge “A review on the current segmentation algorithms for medical images”, 1st International

Conference on Imaging Theory and Applications (IMAGAPP), ISBN: 978-989-8111-68-5, pp. 135-140, Portugal, 2009

[22] Ma Z, Tavares J.M., Jorge RN, Mascarenhas T. “A Review of Algorithms for Medical Image Segmentation and their Applications to the

Female Pelvic Cavity”, Computer Methods in Biomechanics and Biomedical Engineering, 13(2):235-246, 2010.

[23] J. M. R. S. Tavares “Image Processing and Analysis: Applications and Trends”, AES-ATEMA’2010, Fifth International Conference,

ISBN: 978-0-9780479-7-9, pp. 27-41, Canada, June 27-July 03, 2010.

[24] Fred Glover .Tabu search for nonlinear and parametric optimization (with links to genetic algorithms). Discrete Applied Mathematics 49

(1994) 231-255.

[25] Al-Bashish, D., M. Braik and S. Bani-Ahmad, “Detection and classification of leaf diseases using Kmeans-based segmentation and

neural-networks-based classification”, Inform. Technol. J., 2011, 10: 267-275.