Published by : International Journal of Engineering Research & Technology (IJERT)

http://www.ijert.org ISSN: 2278-0181

Vol. 9 Issue 03, March-2020

Plant Disease Detection using Image Processing

1
Mr.V Suresh, 2D Gopinath, 3M Hemavarthini, 4K Jayanthan, 5 Mohana Krishnan
1
Assistant Professor, CSE Department, Dr.NGP Institute Of Technology
2,3,4,5
UG Scholar CSE Department, Dr.NGP Institute Of Technology

Abstract –Identification of the plant diseases is the key to effectively used by farmers thereby increasing the yield
prevent the losses in the yield and quantity of the agricultural rather than visiting the expert and getting their advice.
product. The studies of the plant diseases mean the studies of The main objective is not only to detect the
visually observable patterns seen on the plant. Health disease using image processing technologies. It also directs
monitoring and disease detection on plant is very critical for
sustainable agriculture. It is very difficult to monitor the plant
the user directly to an e-commerce website where the user
diseases manually. It requires tremendous amount of work, can purchase the medicine for the detected disease by
expertise in the plant diseases, and also require the excessive comparing the rates and use appropriately according to the
processing time. Hence, image processing is used for the directions given.
detection of plant diseases by capturing the images of the Greenhouse also called a glasshouse, or, if with
leaves and comparing it with the data sets. The data set sufficient heating, a hoth house, is a structure with walls
consist of different plant in the image format. Apart from and roof made chiefly of transparent material, such as
detection users are directed to an e-commerce website where glass, in which plants requiring regulated climatic
different pesticides with its rate and usage directions are conditions are grown. As greenhouse farming is gaining
displayed. This website can be efficiently used for comparing
the MRP’s of different pesticides and purchase the required
more importance now a day’s, this paper helps the
one for the detected disease. This paper aims to support and greenhouse farmers in an effective way. Various techniques
help the green house farmers in an efficient way. can be used to review the plant disease detection and
discuss in terms of various parameters. The paper is
Keywords: Plant disease detection, Tensor flow, Green house, organized into the following sections. First section gives a
Convolution neural network, Data model, image to byte code brief introduction to the importance of plant disease
detection. Second section discusses the existing work
I. INTRODUCTION carried out recently in this area and also reviews the
techniques used. Section three includes methodologies used
India is a cultivated country and about 70% of the in our paper. Lastly, fourth section concludes this paper
Population depends on agriculture. Farmers have large along with future directions.
range of diversity for selecting various suitable crops and
finding the suitable pesticides for plant. Hence, damage to II. LITERATURE SURVEY
the crops would lead to huge loss in productivity and would
ultimately affect the economy. Alternia leaf spot, Brown spot, Mosaic, Grey spot, and
Leaves being the most sensitive part of plants Rust are five common types of apple disease that severly
show disease symptoms at the earliest. The crops need to affect apple yield. However, the existing research lacks an
be monitored against diseases from the very first stage of accurate and fast detector of apple disease for ensuring the
their life-cycle to the time they are ready to be harvested. healthy development of the apple industry. Object detection
Initially, the method used to monitor the plants from algorithms such as SSD, DSSD and R-SSD can be regarded
diseases was the traditional naked eye observation that is a as consisting of two parts: The first part is the pre-network
time-consuming technique which requires experts to model, which is used as a basic features extractor. The
manually monitor the crop fields. In the recent years, a other is an auxillary structure that utilizes multi-scale
number of techniques have been applied to develop feature map for a detection[1].
automatic and semi-automatic plant disease detection A K-means segmentation is used for partitioning
systems and automatic detection of the diseases by just the leaf image into four clusters using the squared
seeing the symptoms on the plant leaves makes it easier as Euclidean distances. The method applied for feature
well as cheaper. These systems have so far resulted to be extraction is Colour Co-occurrence method for both colour
fast, inexpensive and more accurate than the traditional and texture features[4]. Finally, classification is completed
method of manual observation by farmers using neural network detection algorithm based on back
In most of the cases disease symptoms are seen propagation methodology. The overall system disease
on the leaves, stem and fruit. The plant leaf for the detection and classification accuracy was found to be
detection of disease is considered which shows the disease around 93%.
symptoms. A number of crop types namely, fruit crops,
There are many cases where farmers do not have a vegetable crops, cereal crops and commercial crops to
fully compact knowledge about the crops and the disease detect fungal diseases on plant leaves. Different methods
that can get affected to the crops. This paper can be have been adopted for each type of crop[5].For fruit crops,
k-means clustering is the segmentation method used,

IJERTV9IS030114 www.ijert.org 78
(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 9 Issue 03, March-2020

texture features have been focused on and classified using Image Acquisition
ANN and nearest neighbour algorithms achieving an In this phase, images of plant leaves are gathered
overall average accuracy of 90.723%. For vegetable crops, using digital media like camera, mobile phones etc. with
chan-vase method used for segmentation, local binary desired resolution and size. The images can also be taken
patterns for texture feature extraction and SVM and k- from web.
nearest neighbour algorithm for classification achieving an The formation of database of images is completely
overall average accuracy of 87.825%. dependent on the application system developer. The image
The commercial crops have been segmented using database is responsible for better efficiency of the classifier
grab-cut algorithm. Wavelet based feature extraction has in the last phase of the detection system.
been adopted using Mahalnobis distance and PNN as
classifiers with an overall average accuracy of 84.825%. Image Segmentation
The cereal crops have been segmented using k- This phase aims at simplifying the representation
means clustering and canny edge detector. Colour, shape, of an image such that it becomes more meaningful and
texture, colour texture and random transform features have easier to
been extracted. SVM and nearest neighbour classifiers used Analyze. As the premise of feature extraction, this phase is
getting an overall average accuracy of 83.72%.A chilli also the fundamental approach of image processing.
plant leaf image and processed to determine the health There are various methods using which images
status of the chilli plant. Their technique is ensuring that can be segmented such as k-means clustering, Otsu’s
the Chemicals should apply to the diseased chilli plant algorithm and thresholding etc. The k-means clustering
only. They used the MATLAB for the feature extraction classifies objects or pixels based on a set of features into K
and image recognition. In this paper pre-processing is done number of classes. The classification is done by
using the. minimizing the sum of squares of distances between the
Fourier filtering, edge detection and objects and their corresponding clusters.
morphological operations. Computer vision extends the
image processing paradigm for object classification. Here Feature Extraction
digital camera is used for the image Capturing and Hence, iin ithis istep ithe ifeatures ifrom ithis iarea
LABVIEW software tool to build the GUI[7]. iof iinterest ineed ito ibe iextracted. i iThese ifeatures iare
The FPGA and DSP based system is developed ineeded ito idetermine ithe imeaning iof ia isample iimage.
and used for monitoring and control of plant diseases . The I
FPGA is used to get the field plant image or video data for Features ican ibe ibased ion icolour, ishape, iand
monitoring and diagnosis. itexture. iRecently, imost iof ithe researchers iare
The DSP TMS320DM642 is used to process and iintending ito iuse itexture ifeatures ifor idetection iof
encode the video or image data. The nRF24L01 single chip iplant idiseases. iThere iare ivarious imethods iof ifeature
2.4 GHz radio transmitter is used for data transfer. It has extraction ithat ican ibe iemployed ifor ideveloping ithe
two data compress and transmission method to meet user’s isystem such ias igray-level ico-occurrence imatrix
different need and uses multi-channel wireless i(GLCM), icolor icooccurrence imethod, ispatial igrey-
communication to lower the whole system cost. level idependence imatrix, iand ihistogram ibased ifeature
iextraction. iThe iGLCM imethod iis ia statistical imethod
III. METHODS OF DISEASE DETECTION ifor itexture iclassification.

The process of plant disease detection system basically Classification

involves four phases as shown in Fig 3.1. The first phase The classification phase implies to determine if
involves acquisition of images either through digital the input image is healthy or diseased. If the image is found
camera and mobile phone or from web. The second phase to be diseased, some existing works have further classified
segments the image into various numbers of clusters for it into a number of diseases. For classification, a software
which different techniques can be applied. Next phase routine is required to be written in MATLAB, also referred
contains feature extraction methods and the last phase is to as classifier.
about the classification of diseases. A number of classifiers have been used in the past
few years by researchers such as k-nearest neighbour
(KNN), support vector machines (SVM), artificial neural
network(ANN), back propagation neural network (BPNN),
Naïve Bayes and Decision tree classifiers.
The most commonly used classifier is found to be
SVM. Every classifier has its advantages and
disadvantages, SVM is simple to use and robust technique.

Fig.3.1 Phases of plant disease detection system

IJERTV9IS030114 www.ijert.org 79
(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 9 Issue 03, March-2020

IV. OVERVIEW OF PLANT DISEASE

Plant diseases are generally caused by infectious

agents such as fungi, bacteria, and viruses. Signs of plant
disease are observable evidence of infection and symptoms
are the visible effects of these kinds of disease. Fungal
infections cause signs like visible spores, mildew, or mold
and the basic symptoms are like leaf spot and yellowing.
Fungal diseases are plant infections caused by
fungi. Fungi can be single or multicellular, but either way
infect plants by stealing nutrients and breaking down
tissue. Fungal diseases are the most common infection in
plants. There are some characteristic symptoms, or
observable effects of the disease, in plants.
Fungi infections can be recognized by symptoms
like spots on plant leaves, yellowing of leaves, and birds- Fig 4.2 Leaf affected by bacteria
eye spots on berries. With some fungal diseases, the
organism itself can actually be viewed on the leaves appear Viruses are infectious particles that are too small
as a growth and as a mold, to be detected by a light microscope. They invade host cells
and hijack host machinery to force the host to make
millions of copies of the virus.
Viral diseases don't show any signs in plants since
viruses themselves cannot be seen even with a light
microscope. However, there are symptoms that the trained
eye can observe. A mosaic leaf pattern, yellowed, or
crinkled leaves are all
characteristic of viral infection. This classic pattern of
discoloration is where many plant viruses get their name,
such as the tobacco mosaic virus. Also, decreased plant
growth is also commonly seen in viral infections.

Fig 4.1 Leaf affected by fungal infection

These may a malformations on stems or the

underside of leaves. These direct observations of the
disease-causing organism are called signs of infection
Bacteria are single-celled, prokaryotic organisms. Bacteria
are everywhere and many can be beneficial, but some can
cause disease both in humans and plants.
Fig 4.3 Leaf affected by virus
The signs of bacteria are often harder to detect
than fungi, since bacteria are microscopic. Upon cutting an So, these are our observation on how to classify the various
infected stem, a milky white substance may appear, plant disease and how to be cautious about that.
called bacterial ooze. This is one sign of a bacterial
infection. Other signs include water-soaked lesions, which V. PROPOSED SYSTEM
are wet spots on leaves that ooze bacteria.
Eventually, as the disease progresses, the lesions Proposed system have an end-to-end Android application
enlarge and form reddish-brown spots on the leaves. A with TFLite. Proposed system opted to develop an Android
common symptom of bacterial infection is leaf spots or application that detects plant diseases. It has the algorithms
fruit spots. Unlike fungal spots, these are often contained and models to recognize species and diseases in the crop
by veins on the leaf. leaves by using Convolutional Neural Network. Proposed
system use Colab to edit source code.
A dataset of 54,305 images of diseased and healthy
plant leaves collected under controlled conditions Plant
Village dataset. The images cover 14 species of crops,
including: apple, blueberry, cherry, grape, orange, peach,

IJERTV9IS030114 www.ijert.org 80
(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 9 Issue 03, March-2020

pepper, potato, raspberry, soy, squash, strawberry and VI. RESULTS AND DISCUSSION
tomato. It contains images of 17 basic diseases, 4 bacterial
diseases, 2 diseases caused by mold (oomycete), 2 viral There two different conditions for training and
diseases and 1 disease caused by a mite. 12 crop species testing. One is under the lab conditions, which means that
also have healthy leaf images that are not visibly affected by the model is tested with the images from the same dataset
disease. Our dataset contains solutions for several plant from which it is used for both training and testing. The
textures such as, other condition is that field condition; this means that our
model has tested with the images taken from the real world
1. Apple scab conditions (land). Since the lighting conditions and
2. Apple black rot background properties of the images are totally different
3. Apple cedar apple rust when we take samples from the real field, there is a chance
4. Cherry including sour powdery mildew that our model to produce a very low accuracy, when
5. Cherry including sour healthy comparing to the accuracy values acquired during the lab
6. Corn maize cercospora leaf spot gray leaf spot conditions. So to overcome this impact, we had an idea of
7. Corn maize common rust having a mixed variety of images during the training phase
8. Corn maize northern leaf blight (heterogeneity).
9. Corn maize healthy
10. Grape black rot
11. Grape esca black measles
12. Grape leaf blight isariopsis leaf spot
13. Grape healthy
14. Orange haunglongbing citrus greening
15. Peach bacterial spot
16. Peach healthy
17. Pepper bell bacterial spot
18. Pepper bell healthy
19. Potato early blight
20. Potato late blight
21. Squash powdery mildew
22. Strawberry leaf scorch
23. Tomato bacterial spot
24. Tomato early blight
25. Tomato late blight
26. Tomato leaf mold
Fig.6.1 Performance
27. Tomato septoria leaf spot
28. Tomato spider mites two spotted spider mite The accuracy of Real-time detection of apple leaf disease
29. Tomato target spot using deep learning approach based on improved
30. Tomato yellow leaf curl virus convolution neural networks is less when compared to the
31.Tomato mosaic virus proposed system because it detects multiple diseases in a
single system.
Data generators that will read pictures in our source
folders, convert them to `float32` tensors, and feed them Input Faster R- Proposed system
(with their labels) to our network is set up. (apple) CNN
As data that goes into neural networks should usually Scab 58.82 70.82
be normalized in some way to make it more amenable to Black rot 68.12 82.68
Cedar apple rust 90.34 94.96
processing by the network. In our case, we will pre-process
our images by normalizing the pixel values to be in the `[0,
1]` range (originally all values are in the `[0, 255]` range). VII. CONCLUSION
We will need to make sure the input data is resized to
224x224 pixels or 299x299 pixels as required by the Thus an application built for the identification of
networks. You have the choice to implement image disease affected plants and healthy plants is done and this
augmentation or not. proposed work is focuses on the accuracy values during the
Apart from just detecting the plant disease using real field conditions, and this work is implemented by
the above methods our system directs the user to an e- having several plant disease images.
commerce website. This website displays all the pesticides Overall this work is implemented from scratch
that are available for the detected disease with its MRP and produces a decent accuracy. The future work is to
rate. Along with this the directions to use it is also available increase the number of images present in the predefined
in the website. Thus by comparing the rate and features of database and to modify the architecture in accordance with
the pesticides the user can purchase it the dataset for achieving better accuracy.

IJERTV9IS030114 www.ijert.org 81
(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 9 Issue 03, March-2020

REFERENCES [6] Punajari, J.D., Yakkundimath, R., Byadgi, A.S., ‘Image

Processing Based Detection of Fungal Diseases In Plants’,
International Conference on Information and Communication
[1] Peng Jiang , Yuehan Chen ,Bin Liu , Dongjian He , Chunquan
Technologies, Volume 46, pp. 1802-1808, 2015
Liang ,’ Real-Time Detection of Apple Leaf Diseases Using Deep
[7] Zulkifli Bin Husin, Abdul Hallis Bin Abdul Aziz, Ali Yeon Bin
Learning Approach Based on Improved Convolutional Neural
Md Shakaff Rohani Binti S Mohamed Farook, “Feasibility Study
Networks’, ( Volume: 7 ), pp. 06 May 2019
on Plant Chili Disease Detection Using Image Processing
[2] Zhou, R., Kaneko, S., Tanaka, F., Kayamori, M., Shimizu, M.,
Techniques”, 2012 Third International Conference on Intelligent
‘Disease detection of Cercospora Leaf Spot in sugar beet by
Systems Modelling and Simulation.
robust template matching’, Computers and Electronics in
[8] Chunxia Zhang, Xiuqing Wang, Xudong Li, “Design of
Agriculture, Volume 108, pp.58-70, 2014
Monitoring and Control Plant Disease System Based on
[3] Barbedo, J.G.A., Godoy, C.V., ‘Automatic Classification of
DSP&FPGA”, 2010 Second International Conference on
Soybean Diseases Based on Digital Images of Leaf Symptoms’,
Networks Security, Wireless Communications and
SBI AGRO,2015
TrustedComputing.
[4] Barbedo, J.G.A., ‘A review on the main challenges in automatic
[9] Omrani, E., Khoshnevisan, B., Shamshirband, S., Saboohi, H.,
plant disease identification based on visible range images’, 2016
Anuar, N.B., Nasir, M.H.N., ‘Potential of radial basis function-
,Biosystems Engineering, Volume 144, pp. 52-60.
based support vector regression for apple disease detection’,
[5] Bashish, D.A., Braik, M., Ahmad, S.B., ‘A Framework for
Journal of Measurement, pp. 233-252, 2014
Detection and Classification of Plant Leaf and Stem Diseases’,
[10] Gharge, S., Singh, P., ‘Image Processing for Soybean Disease
International Conference on Signal and Image Processing, pp.
Classification and Severity Estimation’, Emerging Research in
113-118, 2010
Computing, Information, Communication and Applications, pp.
493-500, 2016

IJERTV9IS030114 www.ijert.org 82
(This work is licensed under a Creative Commons Attribution 4.0 International License.)