You are on page 1of 7

AUTOMATIC METHODS FOR CLASSIFICATION OF PLANT DISEASES USING

CONVOLUTION NEURAL NETWORK


AUTHORS
D.AASHA1 , G. AARATHY 2, D.JAMUNA3 , GUNAPALLI LAVANYA 4
GUIDED BY
MS.G. BHAVANI

ABSTRACT-Agriculture is the primary source of economic development in India. The fertility


of soil, weather conditions, and crop economic values make farmers select appropriate crops for
every season. To meet the increasing population requirements, agricultural industries look for
improved means of food production. Researchers are in search of new that would reduce
investment and significantly improve the yields. The proposed decision making system utilizes
image content characterization and supervised classifier type of neural network. Image
processing techniques for this kind of decision analysis involves preprocessing, feature
extraction and classification stage. At Processing, an input image will be resized and region of
interest selection performed if needed. Here, color and texture features are extracted from
network training and classification The system will be used to classify the test images
automatically to decide leaf characteristics. Precision is a new technology that helps in
improving farming techniques. This is one of the main
reasons to be considered for the detection of plant diseases . Plant leaf disease detection are
the noteworthy applications of precision agriculture. The main aim of this project is to identify
the diseased and healthy leaves of distinct plants by extracting features from input images using
CNN algorithm. These features extracted help in identifying the most relevant class for images
from the datasets. By using the CNN algorithm in this proposed system we can achieve 95% of
accuracy.

INTRODUCTION
OVERVIEW
India is a country with a population of approximately 1.38 billion as of April 2020. Estimates put the total
number of farmers in India somewhere between 95.8 million. It must be noted that 18% of India’s GDP is
produced from the agricultural sector. It would, thus, be safe to infer that if agriculture was revolutionized, it’d
benefit the country greatly and also apart from alleviating the conditions of local farmers, it’d also create a lot of
employment and expansion opportunities in the agricultural sectors. Research and development on pesticides,
fungicides, and herbicides have progressed very well in India. Plant diseases affect the growth and crop yield of
the plants and make social, ecological and economic impacts on agriculture. Recent studies on leaf diseases
show how they harm the plants. Plant leaf diseases also cause significant economic losses to farmers. Early
detection of the diseases deserves special attention. But, every year, due to natural reasons, crops succumb to
various known diseases and tons of produced corps is lost and this can be dealt with quick detection of plant
diseases in proper time. It will help to get over the economic conditions faced by the countries. Nowadays
technology has changed lives for the better. Due to the internet, almost everything is within reach. With the help
of a normal camera, one can easily click photos of affected parts and upload it to the system which detects the
particular disease and provides the exact treatment as well. Most of the plants are infected by various fungal and
bacterial diseases. The exponential increase of population, the climatic conditions also cause plant diseases. The
leaves require close monitoring to detect the disease.
PROBLEM STATEMNET AND SOLUTION
Plant diseases are studied in the literature, mostly focusing on the biological aspects. They make
predictions according to the visible surface of plants and leaves. Detection of diseases as soon as they appear is
a vital step for effective disease management. The detection is traditionally carried out by human experts.
Human experts identify diseases visually but they face some difficulties that may harm their efforts. In this
context, detecting and classifying diseases in an exact and timely manner is of the great importance. Advances
in artificial intelligence researches now make it possible to make automatic plant disease detection from raw
images. There are several techniques are reported by many researchers for plant disease detection and
monitoring. They used SVM to detect leaf diseases. For experiments real sample images of leaf have been
considered. In preprocessing, phase images are resized to 224*224 resolutions to deduce the computational time.
Deep learning can be thought as a learning method on neural networks. One of the advantages of deep
learning is that it can extract features from images automatically. The neural network learns how to extract
features while training. CNN is a multi-layer feed-forward neural network and is the popular deep learning
model. Plant leaf images are complex with its background and the colour information extracted from a single
colour component is limited. It causes the feature extraction method to give lesser accuracy results. Using
different colour components is promising instead of single one. In the proposed project we developed a CNN
model based on RGB components of the leaf images on Plant Village dataset .
AIM AND OBJECTIVES
To find the interaction between the diseases causing agents and host plants in relation to the overall
environment. To identify various diseases in plants. To implement a method for preventing the diseases and
providing management for reducing the losses/damages caused by diseases. Our main aim is to provide Easy to
prevent diseases on plants for the farmers. Helps in predicting new solution. Detection of plant disease through
automatic technique is as 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 leaves.

LITERATURE SURVEY
TITLE: A Convolutional Neural Network (CNN) model and Learning Vector Quantization (LVQ) algorithm
based method for tomato leaf disease detection and classification.
AUTHOR: Melike sardogan et al
DESCRIPTION:
[1] It Presents a Convolutional Neural Network (CNN) model and Learning Vector Quantization (LVQ)
algorithm based method for tomato leaf disease detection and classification. They have modeled a CNN for
automatic feature extraction and classification. Color information is actively used for plant leaf disease
researches. In our model, the filters are applied to three channels based on RGB components. It is concluded that
the proposed method effectively recognizes four different types of tomato leaf diseases. To improve recognition
rate in classification process different filters or different size of convolutions can also be used.

TITLE: An overview of the plant disease detection using different algorithms.


AUTHOR: Garima shestra et al
DESCRIPTION:
[2] “Plant Disease Detection Using CNN” Proceedings of 2020 IEEE Applied Signal Processing Conference
(ASPCON) The author gives an overview of the plant disease detection using different algorithms. A CNN
based method for plant disease detection has been proposed here. Simulation study and analysis is done on
sample images in terms of time complexity and the area of the infected region. It is done by image processing
technique.

TITLE: An Examination of morphological features and properties like color, intensity and dimensions of the
plant leaves are taken into consideration for classification.
AUTHOR: L. Sherly Puspha Annabel et al
DESCRIPTION:
[3] It examines morphological features and properties like color, intensity and dimensions of the plant leaves are
taken into consideration for classification. This paper presents an overview on various types of plant diseases
and different classification techniques in machine learning that are used for identifying diseases in different
plant leaves.The techniques used here is Artificial Neural Network for Classification of Disease Detection and
Support Vector Machine.
TITLE: The replacement of standard convolution with depth-separable convolution, which reduces the
parameter number and computation cost.
AUTHOR: Sk Mahmudul Hassan et al
DESCRIPTION:
[4] In this paper, they replaced standard convolution with depth-separable convolution, which reduces the
parameter number and computation cost. The implemented models were trained with an open dataset consisting
of 14 different plant species, and 38 different categorical disease classes and healthy plant leaves. To evaluate
the performance of the models, different parameters such as batch size, dropout, and different numbers of
epochs were incorporated.The accuracy results in the identification of diseases showed that the deep CNN
model is promising and can greatly impact the efficient identification of the diseases, and may have potential in
the detection of diseases in real-time agricultural systems.

TITLE: A methodology for the analysis and detection of plant leaf diseases using digital image processing
techniques.

AUTHOR: Yin Min Oo et al


DESCRIPTION:
[5] ‘It proposed a methodology for the analysis and detection of plant leaf diseases using digital image
processing techniques. The experimental results demonstrate that the proposed system can successfully detect
and classify four major plant leaves diseases: Bacterial Blight and Cercospora Leaf Spot, Powdery Mildew and
Rust. The main goal of this work is to develop an image processing system that can identify and classify four
types of plant disease namely Cercospora Leaf Spot, Bacterial Blight, Powdery Mildew and Rust. The main
purpose of this system is to improve the efficiency of automatic plant disease detection. Experimental results
show that the proposed system can successfully detect and classify the plant disease with accuracy of 98.2%.

TITLE: The four stage proposed system


AUHTOR: Mr. Sachin B. Jagtap et al
DESCRIPTION:
[6] The proposed system consists of four stages; the first is the enhancement, which includes HIS
transformation, histogram analysis, and intensity adjustment. The second stage is segmentation, which includes
adaptation of fuzzy c-means algorithm. Feature extraction is the third stage, which deals with three features,
namely color size and shape of spot. The fourth stage is classification, which comprises back propagation based
neural networks. Applying this model to any other crop disorder requires only spatial care to be taken in order to
acquire a sufficient set of images for training purpose as representative of these disorders. Due to integration of
this proposed system diagnosis accuracy will increase.

PROPOSED APPROACH
In this proposed system we uses CNN algorithm to extract the features of the plants. The working of CNN
Basically, First we resize every image into 224 x 224. Convolutional-neural-network (CNN) models are
implemented to identify and diagnose diseases in plants from their leaves, since CNNs have achieved
impressive results in the field of machine vision. The implemented models were trained with an open dataset
consisting of different plant species, and different categorical disease classes and healthy plant leaves. To
evaluate the performance of the models, different parameters such as batch size, dropout, and different numbers
of epochs were incorporated. After that this image feed into the Convolutional Neural Network. At last, we
flatten the final max pool layer output and feed to the next linear layer which is also called a fully connected
layer, and finally, as a final layer.
PROCESS FLOW

RESEARCH METHODOLOGY:
A. IMAGE ACQUISITION:-The first stage of any vision system is the image acquisition. Image acquisition
involves the steps to obtain the plant leaf and captured the high quality images through the camera. Images are
acquired from the internet or agriculture field. The efficiency of the concept depends upon the quality of
database images. This image is in RGB (Red, Green and Blue) form.
B. IMAGE PRE-PROCESSING:-Image pre-processing involves the steps of image enhancement, RGB to Lab
conversion, filtering etc. Here, image enhancement is carried out for increasing the contrast. Image smoothing is
done using the filtering techniques. There are different types of filtering techniques available in image
processing like median filter, average filter, Gaussian filter etc.
C. IMAGE SEGMENTATION:- Image segmentation means partitioning of image into various parts of same
features or having some similarity. The segmentation can be done using various methods like k-means
clustering, converting RGB image into HIS model etc. 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.
D. CLASSIFICATION & DETECTION OF DISEASES:- Finally, classifiers are used for the training and
testing of the datasets. These classifiers may be support vector machine (SVM), k-nearest neighbor,
Convolutional neural network, fuzzy logic based etc. These methods are used to classify and detect the leaf
diseases.

EXPERIMENTATION
Structure of CNN
Convolutional Neural Network (CNN) architecture for plant leaf disease detection using Deep Learning
technique is proposed. A CNN model is trained with the help of the Plant leaf Dataset consisting of images
comprising of different classes of both unhealthy and healthy leaves. The system was designed to detect and
recognize diseases of several plant varieties specifically apple, corn, grapes, potato, strawberry, and tomato. It is
comprised images of healthy plant leaves and infected with the diseases. At first stage of this project, the
collected dataset images are preprocessed. This preprocessing stage includes image resizing, image
enhancement and segmentation. The next process is training the preprocessed image. Neural layer is trained
using a dataset containing images of healthy and diseased plant leaves. The model serves its objective by
classifying images of leaves into diseased categories based on the features of the leaves. The leaves have texture
and visual similarities which are attributes for the identification of disease type. The next process is testing
images; here a new set of images is used for testing. When we upload the images of the leaf, our model detects
whether the given leaf image is healthy or diseased with high level of accuracy.

DATA GENERATION
List of Datasets

We analyse 54,306 images of plant leaves, which have a spread of 38 class labels assigned to them.
Each class label is a crop-disease pair, and we make an attempt to predict the crop-disease pair given just the
image of the plant leaf. In all the approaches described in this paper, we resize the images to 224 × 224 pixels,
and we perform both the model optimization and predictions on these downscaled images. Across all our
experiments, we use three different versions of the whole Plant Village dataset. We start with the Plant Village
dataset as it is, in colour; then we experiment with a Gray-scaled version of the Plant Village dataset, and finally
we run all the experiments on a version of the Plant Village dataset where the leaves were segmented, hence
removing all the extra background information which might have the potential to introduce some inherent bias
in the dataset due to the regularized process of data collection in case of Plant Village dataset. 25 Segmentation
was automated by the means of a script tuned to perform well on our particular dataset.
Images from the Dataset
We chose a technique based on a set of masks generated by analysis of the colour, lightness and saturation
components of different parts of the images in several colour spaces (Lab and HSB). One of the steps of that
processing also allowed us to easily fix colour casts, which happened to be very strong in some of the subsets of
the dataset, thus removing another potential bias. This set of experiments was designed to understand if the
neural network actually learns the “notion” of plant diseases, or if it is just learning the inherent biases in the
dataset.

TRAINING AND VALIDATION ACCURACY AND LOSS

RESULT AND DISCUSSION


Detection of plant disease through automatic technique is as 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
leaves. To find the interaction between the diseases causing agents and host plants in relation to the overall
environment. To identify various diseases in plants. To implement a method for preventing the diseases and
providing management for reducing the losses/damages caused by diseases .Our main aim is to provide Easy to
prevent diseases on plants for the farmers. Helps in predicting new solution.

CONCLUSION AND FUTURE WORK


A large part of the Indian population relies on agriculture, hence it becomes very essential to detect and
recognize the leaf diseases that results in losses, since agriculture is critical to the growth of the economy. This
project based on deep learning approach .Here CNN is utilized to build different plant leaf disease identification,
detection and recognition system. This approach utilized a minimum set of layers to identify the diseases of
several classes. The neural network is trained with Plant Village dataset. A Graphical User Interface is designed
for this system. This GUI permits the user to choose the images from the dataset. User can select any image
from the dataset and the image gets loaded, following which the prediction of the disease will be shown on the
User Interface. Convolutional neural network, trained for identifying and recognizing the plant leaf disease,
could classify and predict the diseases correctly for almost all the images with few anomalies thus and obtained
95% of accuracy.

REFERENCES

[1] Sumit Kumar,Veerendra Chaudhary,Ms.Supriya Khaitan Chandra " Plant Disease Detection Using
CNN ",May 2021

[2] Sk Mahmudul Hassan , Arnab Kumar Maji , Michał Jasi "Identification of Plant-Leaf Diseases Using
CNN and Transfer-Learning Approach",2021.

[3] Garima Shrestha,Deepsikha,Majolica Das,"Plant Disease Detection Using CNN",2020 IEEE Applied
Signal Processing Conference (ASPCON).

[4] H.Wang, G.Li, Z.Ma, X.Li, “Application of Neural Networks to Image Recognition of Plant Diseases”,
International Conference on Systems and Informatics(ICSAI2012), 2012.
[5] M.Sardogan, A.Tuncer, Y.Ozen, “Plant Leaf Disease Detection and Classification based on CNN with
LVQ Algorithm”, 3rd International Conference on Computer Science and Engineering (UBMK), Sept. 2018.

[6] L. Sherly Puspha Annabel,T. Annapoorani and P. Deepalakshmi "Machine Learning for Plant Leaf
Disease Detection and Classification – A Review" International Conference on Communication and Signal
Processing, April 2019.

[7] Bin Liu, Yun Zhang, Dong Jian He and Yuxiang Li, “Identification of Apple Leaf Diseases Based on
Deep Convolutional Neural Networks” , 2017.

[8] Melike Sardogan, Melike Sardogan, Yunus Ozen "Plant Leaf Disease Detection and Classification
based on CNN with LVQ Algorithm"3rd International conference on computer science and engineering,2018.

[9] S. Zhang, W. Huang and C. Zhang, “Tree-channel convolutional neural networks for vegetable leaf
disease recognition”, Cognitive Systems Research, 2018.

[10] A.Sehgal, S.Mathur, “Plant Disease Classification Using SOFT COMPUTING Supervised Machine
Learning”, 3rd International Conference on Electronics, Communication, and Aerospace Technology (ICECA),
Coimbatore, India 2019.

[11] M.Islam, AnhDinh, K.Wahid, P.Bhowmik, “Detection of Potato Diseases Using Image Segmentation and
Multiclass Support Vector Machine”, IEEE 30th Canadian Conference on Electrical and Computer Engineering
(CCECE), Windsor, Canada, 30 April- 3 May 2017 .

[12] Ganesan, S.Sajiv, L.Megalanleo Leo, “CIELuv Color Space for Identification and Segmentation of
Disease Affected Plant Leaves Using Fuzzy based Approach”, 3rd International Conference on Science
Technology Engineering & Management (ICONSTEM),Chennai India,March 2017.

[13] C.Khitthuk, A.Srikaew, K.Attakitmongcol, P.Kumsawat “Plant Leaf Disease Diagnosis from Color
Imagery Using Co-Occurrence Matrix and Artificial Intelligence System”, International Electrical Engineering
Congress(iEECON), Krabi, Thailand, 7-9 March 2018

[14] N.Petrellis, “A Smart Phone Image Processing Application for Plant Disease Diagnosis”, 6th
International Conference on Modern Circuits and Systems Technologies(MOCAST), Thessaloniki, Greece, 4-
6May2016

[15] A.Sehgal, S.Mathur, “Plant Disease Classification Using SOFT COMPUTING Supervised Machine
Learning”, 3rd International Conference on Electronics, Communication, and Aerospace Technology (ICECA),
Coimbatore, India, 12-14June 2019, pp.75

You might also like