Fault Area Detection in Leaf Diseases using k-means
Clustering
Subhajit Maity
1
,Sujan Sarkar
2
,Avinaba Tapadar
3
,Ayan Dutta
4
,Sanket Biswas
5
,
Sayon Nayek
6
,Pritam Saha
7
1,2,3,4,6,7
Department of Electronics and Communication Engineering,
5
Department of Computer Science Engineering
Jalpaiguri Government Engineering College
Jalpaiguri, West Bengal, India
smaity.jgec18@gmail.com
1
,sujansa19997@gmail.com
2
, avinaba.bwn@gmail.com
3
, dutta.ayan1998@gmail.com
4
,
sanketbiswas1995@gmail.com
5
,sayon.bwn@gmail.com
6
, pritamsaha125@gmail.com
7
Abstract— With increasing population the crisis of food is
getting bigger day by day. In this time of crisis, the leaf disease of
crops is the biggest problem in the food industry. In this paper,
we have addressed that problem and proposed an efficient
method to detect leaf disease. Leaf diseases can be detected from
simple images of the leaves with the help of image processing and
segmentation. Using k-means clustering and Otsu’s method the
faulty region in a leaf is detected which helps to determine proper
course of action to be taken. Further the ratio of normal and
faulty region if calculated would be able to predict if the leaf can
be cured at all.
Keywords—k-means clustering, image segmentation,
unsupervised learning, leaf disease, fault area detection, Otsu’s
method, background clipping
I. INTRODUCTION
Based on some research, by 2100 earth’s estimated
population is 11.2 billion, and with this day to day growing
issue, there is an argent need to expand the production of food.
As there is there very few number of cultivated lands left, to
feed the whole world we have to produce food beyond our
limit. It is also observed that crops, of worth several billion
dollars are losses annually due to corps diseases. But the main
problem is production of food slows down by the influence of
diseases. At this moment it will be a necessary step to
minimize the loss and to secure the corps by using
technological support. In most cases pests and diseases are
found on the leaves or branches of the plant.
TABLE I . The Total loss of crop according to Food and Agriculture
Organization of the United Nations
Because of the compilation of visual patterns, enough
study is not done on these optically observed diseases, and for
that, the demand for more precise and sophisticated image
pattern discerning is increasing continuously. By using image
processing techniques, an image can be defined over two
dimensions (feasibly more), using that more precise image
pattern can be found, which plays an important role in crops
cultivation. There are a number of popular digital image
processing techniques are available like Hidden Markov
models, Image restoration, Anisotropic diffusion, Image
editing, Linear Filtering, Partial differential equations,
Independent Component analysis, Pixilation, Principal
Components Analysis, Wavelets, Self-organizing.
Classification of crop diseases using image
processing was researched by Ying[15].Ying said, “ Leaves
with marks must be carefully examined in order to carry out
intelligent diagnosis on the basis of image processing”.
Important methods of image processing :
• Image clipping: Based on marks , Classification of
leaves.
• Thresholding: Image segmentation into spot
background.
• Noise reduction: Noises are wiped out by medium
filter.
By experts, two different methods for the diagnosing of plant
diseases were put forward:
1. Graphical representation
2. Step by step descriptive methods
For the grading process of flue-cured tobacco leaves
image feature extraction is useful. In machine vision
techniques [16], automated investigation of flue-cured tobacco
leaves was mentioned. The above mentioned techniques were
used to solve the problems of feature extraction.
II. LITERATURE SURVEY
In 1979 and 1980 Punjab and Haryana, states of India, heavily
infected by a disease Xanthomonas oryzae, a bacterial disease
causes most destructive bacterial blight of rice which causes
almost 50% of worldwide annual yield loss[21,22].the
bacterial blight is most common disease on hybrid rice in
Zhejiang Province in China[23] reported by Cai and Zhong.
Another disease has been found in over 85
countries in the world, the name of the disease is called
Magnaporthe grisea (a fungal disease), It is capable of
destroying food which is enough to feed more than 60 million
people every year.
By image processing techniques and using neural
networks , pest damage in pip fruits can be detected, here
wavelets are used as a means by line detection, which was