Cotton Paper

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Image Processing Based Detection and Classification of

Various Foliar Maladies In Cotton leaf


Deepti Mahale, Kirthi Mathad, Vani Tadakod
B.E Student(Electronics and Communication Engg.)
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal,Karnataka
email: deeptikmahale28@gmail.com

Dr. Mahendra M.Dixit


Head of Department(Electronics and Communication Engg.)
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal,Karnataka
email: mmdixit@yahoo.co.in

Abstract—Agriculture is the primary source of livelihood the plant instead to the soil also. There are two important
for about 58% of India’s population. Almost 70% people aspects of plant disease detection methods that is to achieve
depend on it & shares major part of the GDP. The major speed and accuracy. There is need for developing technique
part of agricultural commodity is the harvest, quality and
such as automatic plant disease detection and classification
quantity of the crop. Cotton is one of the most important
fiber and cash crop of India and plays a dominant role in the using leaf image processing techniques. This will prove
industrial and agricultural economy of the country. The useful technique for farmers and will alert them at the right
major constraint in production of cotton crop is that it is time before spreading of the disease over large area. The
highly prone to pests and diseases. Therefore to overcome ideology behind comprises five stages and they are: in the
this problem we have found a solution to control the quality first stage we create a color transformation structure for the
and quantity of growth of cotton crop. We are providing RGB leaf image and then, we apply color space
smart solution to automatically detect and classify cotton
transformation for the color transformation structure. Then
plant leaf diseases. In this we are using image processing
techniques to classify diseases & quickly detect the type of image is segmented using the K-means clustering technique.
disease. This approach will enhance productivity of crops. It In the second phase, unnecessary part (green area) within
includes several steps viz. image acquisition, image pre- leaf area is removed. In stage calculate the texture features
processing, segmentation, features extraction and for the segmented infected object. In the fourth stage the
classification. extracted features are passed through a pre-trained neural
network plants. Finally we classify and detect the type of
Keywords-ImageProcessing, Feature extraction, Segmentation, disease in the cotton plant.
Support Vector Machine.

II. ORIGIN OF RESEARCH


I. INTRODUCTION
Agriculture has played a philanthropic role in the A. Image Analysis techniques:
development mankind. Cotton is a heavily traded Image analysis is really helpful to protect the plant from
agricultural commodity .Over 150 countries involved in disease attached.
Exports / Imports. Cotton provides direct employment to 60  Detect diseased leaf part.
Lakh Farmers of the Country. Cotton is used in Clothing,  Segregate affected area by disease.
Seed Oil, in Food, Cosmetics, Coffee Filters Its seeds are  Find the boundaries of the affected area.
crushed to make Oil, Rubber and Plastics. If there is  Determine the shape and size of the affected area.
decrease in agro products, total economy will get affected.  Identify the Object correctly.
Disease detection plays a major role in crop cultivation
because the diseases in crops are inevitable. The traditional B. Types of cotton leaf diseases:
method adopted by farmer was to observe the plant through  Alternaria leaf spot
there naked eye and followed by chemical testing was the  Cercospora leaf spot
way of detection and classification of agricultural plant  Fusarium wilt
diseases.A nd this process becomes tedious as the farm land  Grey mildew
is larger. If farmers are unaware of non-native diseases.  Bacterial Blight
Consultation of experts for this might be time consuming &  Grey mildew
costly. Also the abrupt use of pesticide is not only damaging
III. METHODOLOGY Segmentation can be carried out by various techniques:
 Thersholding technique: Thersholding creates
The proposed methodology for leaf diseases detection
binary image from grey scale level image by
and classification using image processing consists of
setting all pixels below some threshold to zero
following steps:
and all pixels above that threshold to one.
 Edge detection method: Edge detection is an
A. Image Acquisition: image processing technique for finding the
In this step the plant is selected which is affected by disease boundaries of objects within images.
and then collect the leaf of plant and take a snapshot of leaf  K-clustering method: It is used for classification
and load the leaf into system. The RGB color images of of objects based on a set of features into K-
cotton leaf are captured using high resolution digital camera. number of classes. The classification of object is
Images are stored in jpg format. The HSI color space is very done by minimizing the sum of square of
important and attractive color model for image processing distance between the object and corresponding
applications because it represents colors similarly how the clusters.
human eye senses colors. The HSI color model represents
 ANN method: Based on the simulation of
every color with three components: hue(H), saturation(S), learning process for decision making the ANN
intensity(I). works. It used to model the non-linear problems.
cos−1(((𝑅−𝐺)−(𝑅−𝐵))  CNN method: A convolution neural network is
H= used to detect and classify objects in an image.
2√(𝑅−𝐺)2 +(𝑅−𝐵)(3𝐺−𝐵)
It is a feed-forward neural network that is
generally used to analyze visual images by
S=1-3min(R G B)/I processing data with grid-like topology.
I=(R+G+B)/3
D. Feature Extraction
B. Image Pre-processing: The technique that represents the most informative part
The pre-processing task involves some procedures to of an image by dimensionality reduction is called feature
prepare the images enhancement. The collected database extraction. There are three important types of features, such
image is then processed through the computer system for as color, shape, and texture. It plays an important role for
preprocessing captured images stored in RGB format are identification of an object. After performing the image
then cropped and resized to some standard size. segmentation the disease portion from the image is
extracted. In many application of image processing feature
 Noise Removal: Here smooth pixel intensity values extraction is used.
are reduced to see the impact of variation on
binarization. Image smoothening is done using the  Color based Feature Extraction: Color is an important
smoothening filter. sign in recognizing different classes. The pixel in a
color image is commonly represented in the RGB
 Background isolation and subtraction: here the space. The red, green, and blue Value from a color
isolation and removing of background of an image is image capturing device.
done before binarizing.Image clipping is done to get
the only interested image region.  Shape based Feature Extraction: Sharpness, Eigen
values, Area, Perimeter and Aspect-ratio are
 Binary Morphology: In a binary image, it computed as shape based features to detect the
emphasized for a particular patterns or shapes . this diseases. Area of an object is defined as the
enhances the contraction of image number of pixels comprising the object. So area
C. Segmentation of Plant Image: can be enumerated by counting the object pixels.
In segmentation the bright background is separated from E. Classification and detection of leaf disease
a dark text. Separating an image into distinct parts is called After segmentation, processing and feature extraction we go
Segmenting an image. Segmentation is the process of for learning and storing of database image using neural
partitioning an image into parts or regions. This division networking. This stage is called the decision stage of image
into parts is often based on the characteristics of the pixels processing the efficiency of system is highly dependent on
in the image. the type of classifier we use. Classification is dependent of
features extraction input. A large set of database are used to
Segmentation can also be used to: train data and this is done with the help of Support Vector
 Identify individual objects in the image. Machine (SVM).Here we have used genetic algorithm using
 Determine connected pixel regions with similar SVM to build our classifier. The principle used in SVM is
properties. Structural Risk Minimization(SRM). Support Vector
 Determine the boundaries between regions. Machines are based on the concept of decision planes
 Remove unwanted areas. that define decision boundaries. A decision plane is one
that separates between a set of objects having different
class memberships. The SVM is based on statistical learning
theory.

IV. RESULTS
 Image is captured through the camera.
 Image is loaded and then Image acquisition takes
place.
 Then the enhancement of contraction is done on the
image.
 Then we need to choose one image from the
cluster.
 Segmentation of image helps us to segregate the
dark and bright part in clustered image.
 Later the features are extracted and the values are
viewed.
 Finally the classification and identification of type
of disease is detected.

V. CONCLUSION
The accuracy and speed of detection and classification of
plant disease is important aspect for successful cultivation of
crop can be done using image processing. This attempt to
automatic disease recognition process using Advance color
image processing has helped farmers expand their economic
stability.

References

[1] P. Revathi & M. Hemalatha, Advance Computing Enrichment


Evaluation of Cotton Leaf Spot Disease Detection U sing Image Edge
detection”, IEEE-20180 (ICCCNT), Coimbatore, India, 2012.
[2] Prof. Sanjay B. Dhaygude, 2Mr.Nitin P.Kumbhar, International
Journal of Advanced Research in Electrical, Electronics and
Instrumentation Engineering Vol. 2, Issue 1, January 2013
Agricultural plant Leaf Disease Detection Using Image Processing
[3] P. Revathi & M. Hemalatha,” Classification of Cotton Leaf Spot
Diseases Using Image Processing Edge Detection Techniques”, IEEE,
International conference on emerging Trends in science, engineering
& technology, 2012, pg: No: 169-173.
[4] PradnyaRavindra Narvekar1, Mahesh Manik Kumbhar2, S. N.
Patil3,” Grape Leaf Diseases Detection & Analysis using SGDM
Matrix Method”, International Journal of Innovative Research in
Computer and Communication Engineering.
[5] ] Yan Cheng Zhang, Han Ping Mao, Bo Hu, Ming Xili “features
selection of Cotton disease leaves image based on fuzzy feature
selection techniques” IEEE Proceedings of the 2007 International
Conference on Wavelet Analysis and Pattern Recognition, Beijing,
China, 2-4 Nov. 2007.

You might also like