GUIDE- Prof.

KASHI NATH DEY

AMANISHA DAS
MADHURIMA CHANDA
SHALINI YADAV
SAYANTAN BANERJEE
 CONTENTS
 INTRODUCTION
 STEPS OF IMAGE PROCESSING
 WHAT IS PLANT DISEASE?
 DIAGNOSIS OF A PLANT DISEASE
 IMAGE PRE PROCESSING
 K MEANS CLUSTERING
 FEATURE EXTRACTION
 CLASSIFICATION
 RESULTS
 CONCLUSION
 FUTURE SCOPE
 REFERENCES
What is Image Processing?
Image processing is a method to convert an image into digital form and perform some
operations on it, in order to get an enhanced image or to extract some useful information
from it. It is a type of signal dispensation in which input is image, like video frame or
photograph and output may be image or characteristics associated with that image.
Usually Image Processing system includes treating images as two dimensional signals while
applying already set signal processing methods to them.
It is among rapidly growing technologies today, with its applications in various aspects of a
business. Image Processing forms core research area within engineering and computer
science disciplines too.
 STEPS of image processing

Image processing basically includes the following three steps.

 Importing the image with optical scanner or by digital photography.
 Analyzing and manipulating the image which includes data compression and
image enhancement and spotting patterns that are not to human eyes like
satellite photographs.
 Output is the last stage in which result can be altered image or report that is
based on image analysis.
 A plant disease is usually defined as abnormal growth
and/or dysfunction of a plant. Diseases are the result of
some disturbance in the normal life process of the plant.
Diseases may be the result of living and/or non-living
causes. Biotic diseases are caused by living organisms
(e.g., fungi, bacteria, and viruses). Abiotic diseases are
caused by non-living environmental conditions, (e.g., soil
compaction, wind, frost, soil salt damage, and girdling
roots).
Diagnosis consists of 4 phases:-

 The image is segmented using K means clustering

technique.
 Unnecessary part within the leaf area is removed.
 Texture features for the segmented infected object is
calculated.
 The extracted features are passed through a pre trained
Support Vector Machine.
 IMAGE PRE PROCESSING
Various pre processing techniques are considered
like Image clipping i.e. cropping of the leaf image
to get the interested image region. Image
enhancement is carried out for increasing the
contrast.

Original cropped image of leaf with affected region Contrast enhanced image
 K – MEANS CLUSTERING
The K-means clustering is used for classification of object based
on a set of features into K number of classes. The classification of
object is done by minimizing the sum of the squares of the
distance between the object and the corresponding cluster.
The algorithm for K –means Clustering:
1. Pick center of K cluster, either randomly or based on some
heuristic.
2. Assign each pixel in the image to the cluster that minimizes
the distance between the pixel and the cluster center.
3. Again compute the cluster centers by averaging all of the
pixels in the cluster. Repeat steps 2 and 3 until convergence is
attained.
Leaf image segmentation with three clusters formed by K-means clustering
method
 FEATURE EXTRACTION
1. In feature extraction desired feature vectors such as color,
texture, morphology and structure are extracted.

2. Feature extraction is method for involving number of resources

required to describe a large set of data accurately.

3. Statistical texture features are obtained by Gray level co-

occurrence matrix (GLCM) formula for texture analysis and texture
features are calculated from statistical distribution of observed
intensity combinations at the specified position relative to others.

4. Different statistical texture features of GLCM are energy, sum

entropy, covariance, information measure of correlation, entropy,
contrast and inverse difference and difference entropy.
 CLASSIFICATION
1. The standard form of SVM was intended for two-class problems. However, in real life situations,
it is often necessary to separate more than two classes at the same time.

2. In this project, we explore how SVM can be extended from binary problems to multi
classification problems with k classes where k > 2. There are two approaches, namely the one-
against-one approach and the one-against-all approach.

3. In one-to-one approach binary SVM is trained for every two classes of data to construct a
decision function. There are k (k−1)/2 decision functions for the k-class problem. In the
classification stage, a voting strategy is used where the testing point is designated to be in a
class having the maximum number of votes. The voting approach is called the “Max Wins”
strategy.

4. In one-against-all approach, there will be one binary SVM for each of the class to isolate the
members of one class from the other class.
In the figure
the input
image should
be loaded,
then the
contrast of
the image will
be enhanced.
The image
then
undergoes
segmentation.
The input
image is first
preprocessed
and the
contrast is
enhanced so
that the
minute
details are
visible on a
larger scale.
Now select
the Segment
Image
option.
Select one
among the
segmented
images in which
the disease can
be identified
clearly. The
selected image
will then be
used for further
processing.
Among the
segmented
images, one
image is
selected and
classified based
on the ROI,.
The classifier
detects that the
input leaf image
belongs to the
Cercospora Spot
disease type.
The
computed
accuracy will
be displayed.
For accuracy
calculation,
the kernel
function is
changed and
the cross
validation
along with
class
performance
is used.
 CONCLUSION
 This project provides efficient and accurate plant disease
detection and classification technique by using MATLAB image
processing as shown in the figures.

 The proposed methodology in this paper depends on K-means and

Multi SVM techniques which are configured for leaf disease
detection. The MATLAB software is ideal for digital image
processing.

 K-means clustering and SVM algorithm provides high accuracy and

consumes very less time for entire processing. In future work, we
will extend our database for more plant disease identification.
 FUTURE SCOPE
This project deals with the detection of disease in plants
by the means of image processing using MATLAB. But it can
at this point in time only detect one disease per leaf
sample. This can be made into a more advanced and
accurate project by being able to detect multiple diseases
(if any) per leaf sample. The accuracy results in an
available range from mid-90 to top 90%. This can be
improved by increasing the database. However, the results
obtained from real life images are very encouraging.
 REFERENCES
1. P. Chaudhary, A. K. Chaudhari, Dr. A. N. Cheeran and S.Godara, “Color
Transform Based Approach for Disease Spot” International Journal of Computer
Science and Telecommunications Volume 3, Issue 6, pp.65-70, June 2012.

2. H. Al-Hiary, S. Bani-Ahmad, M. Reyalat, M. Braik, and Z. AL Rahamneh, “Fast

and Accurate Detection and Classification of Plant Diseases”, IJCA, Vol-17, No.-
1, pp. 31-38, March 2011.

3. Suman T. and Dhruva Kumar T., “Classification of paddy leaf diseases using
shape and color features”, IJEEE, Volume 07, Issue 01, PP.239250, Jan- June
2015.