9 VII July 2021

International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

Leaf Disease Discerning and Pesticides

Recommendation using Neural Network
Patil N S1, Mohammed Saeem2, Simran Banu3, Soundarya S R4, Asfiya Maseera J Dhanebag5
1, 2, 3, 4, 5
Department of Information Science and Engineering, BIET Davanagere

Abstract: Crop production problems are common in India which severely effect rural farmers, agriculture sector and the
country’s economy as a whole. Food production is to be compromised by various problems; one among them is leaf disease. In
Crops, leaf plays a significant job as it gives data about the amount and nature of yield ahead of time contingent on the state of
leaf. In this paper we propose the framework which takes a shot at pre-processing, feature extraction of leaf pictures from plant
dataset pursued by convolution neural system for disease classification and suggesting Pesticides utilizing Tensorflow
innovation.
Keywords: Convolution Neural Network (CNN), Tensorflow and Pesticides

I. INTRODUCTION
Agrarian efficiency is something on which economy profoundly depends. This is the one of the reasons that disease recognition in
plants assumes a significant job in agribusiness field, as having disease in plants are very characteristic. In the event that appropriate
consideration isn't taken here, at that point it causes genuine impacts on plants and because of which individual item quality, amount
or efficiency is influenced. Recognition of plant sickness through some auto-programmed strategy is useful as it diminishes a huge
work of observing in huge ranches of crops, and at beginning period itself it identifies the side effects of sicknesses for example at
the point when they show up on plant leaves.
Innovation helps individuals in expanding the generation of food. Anyway the generation of food can be influenced by number of
factor, for example, climatic change, infections, soil fruitfulness and so forth. Out of these, disease plays major job to influence the
generation of food. Agriculture plays an significant job in Indian economy. Leaf spot infections debilitate trees and bushes by
intruding on photosynthesis, the procedure by which plants make vitality that supports development and guard frameworks and
impacts survival [1].
Over 58% smallholder rancher relies upon horticulture as their head methods for occupation. In the creating scene, more than 80
percent of the agrarian creation is produced by smallholder ranchers, and reports of yield loss of more than half because of vermin
and illnesses are common[2]. The creation is diminishing step by step with different variables and one of them is sicknesses on
plants which are not identified early arrange.
Different endeavours have been created to avoid crop misfortune due to maladies. Chronicled methodologies of broad application of
pesticides have in the previous decade progressively been enhanced by coordinated irritation the board (IPM) approaches [7].
Autonomous of the methodology, distinguishing an illness effectively at the point when it initially shows up is a vital advance for
proficient illness the board. Verifiably, ailment recognizable proof has been upheld by farming augmentation associations or other
organizations, for example, neighbourhood plant facilities. In later occasions, such endeavours have also been upheld by giving data
for sickness finding web based, utilizing the expanding web infiltration around the world. Considerably more as of late, devices in
view of cell phones have multiplied, exploiting of the generally unrivalled fast take-up of cell phone innovation in all pieces of the
world [8].
There is different work is done in earlier years. Bacterial sickness lessens plants development fastly so to distinguish this kind of
infections , Identifying the ailment at an beginning time and proposing the arrangement so greatest mischief can be maintained a
strategic distance from to expand the harvest yield [4] have utilized ANN and K-intends to group the ailment and grade the ailment
for. There is a need to structure the programmed framework to identify the leaf ailment and suggest the correct pesticide.
So as to create exact picture classifiers for the reasons of plant disease determination, we required an enormous, confirmed dataset
of pictures of unhealthy and solid plants. Until as of late, such a dataset didn't exist, and significantly littler datasets were not
unreservedly accessible. To address this issue, the Plant Village venture has started gathering a huge number of pictures of solid and
ailing yield plants [9], and has made them straightforwardly and uninhibitedly accessible.

©IJRASET: All Rights are Reserved 1775

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

A. Problem Statement
Crop illnesses are a significant danger to food security; however their fast recognizable proof stays troublesome in numerous pieces
of the world due to the absence of the fundamental foundation. This problem is overcome by blend of expanding worldwide
computer infiltration and ongoing advances in neural science made conceivable by profound learning has made ready for system
helped disease finding and suggesting required pesticide.

B. Objectives
In Crops leaf plays an significant job as it gives data about the amount and nature of yield ahead of time contingent on the state of
leaf. We propose the framework which should accomplish the following objectives:
1) To process the data of the leaf picture from plant dataset.
2) To extract the feature from pre-processed data.
3) To accurately classify the leaf disease.
4) To suggest the pesticide based on disease classification using tenor stream innovation technique.
5) To evaluate performance analysis.

C. Methodology
The fundamental point is to structure a framework which is proficient and which give disease name and pesticides name as quick as
conceivable. For that reason we utilize two stage: first is training stage and second is testing stage. In first stage: Image procurement,
Image Pre-preparing and CNN based preparing. In second stage Image procurement, Image Pre-preparing, classification and disease
distinguishing proof and pesticides identification. For experimentation reason we have utilized PlantVillage datasets.

Fig. 1 Block Diagram of Proposed System

II. LITERATURE SURVEY

In 2017, Monzurul Islam, Anh Dinh, Khan Wahid ,Pankaj Bhowmik present an approach that coordinates picture preparing and AI
to permit diagnosing ailments from leaf pictures. This computerized technique groups sicknesses (or nonappearance thereof) on
potato plants from an openly accessible plant picture database called 'Plant Town'. Our division approach and usage of help vector
machine show sickness grouping more than 300 pictures with a precision of 95%.
In 2017, Vijai Singh A.K. Misra exhibits a calculation for picture division method which is utilized for programmed identification
and order of plant leaf illnesses. It moreover spreads overview on various infections characterization strategies that can be utilized
for plant leaf illness identification. Picture division, which is a significant angle for sickness discovery in plant leaf illness, is
finished by utilizing hereditary calculation.

©IJRASET: All Rights are Reserved 1776

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

In 2016, Sharada Prasanna Mohanty,David Hughes and Marcel Salathe Utilizing an open dataset of 54,306 pictures of ailing and
sound plant leaves gathered under controlled conditions, they train a profound convolutional neural system to distinguish 14 harvest
species and 26 ailments (or nonattendance thereof). The prepared model accomplishes an exactness of 99.35% on a held-out test set,
exhibiting the practicality of this methodology.
In 2016, Davoud Ashourloo, Hossein Aghighi, Ali Akbar Matkan, Mohammad
Reza Mobasheri, and Amir Moeini Rad uses the spectra of the contaminated and non tainted leaves in various infection
manifestations were estimated utilizing a non imaging spectroradiometer in the electromagnetic district of 350 to 2500 nm. So as to
deliver a ground truth dataset, we utilized photographs of an advanced camera to figure the infection seriousness and ailment
manifestations portions. At that point, extraordinary test sizes of gathered datasets were used to prepare each technique.
In 2016, Noa Schor, Avital Bechar,Timea Ignat present an automated location framework for joined recognition of two significant
dangers of nursery ringer peppers: Powdery buildup (PM) and Tomato spotted wither infection (TSWV). The framework depends
on a controller which encourages arriving at different location presents. A few identification calculations are created dependent on
head part investigation (PCA) and the coefficient of variety (CV). Tests find out the framework can effectively identify the plant
and arrive at the identification posture required for PM, yet it experiences issues in arriving at the TSWV discovery present.
In 2016, Lucas G. Nachtigall and Ricardo M. Araujo ponders the utilization of Convolutional Neural Systems to naturally
distinguish and characterize sicknesses, wholesome insufficiencies and harm by herbicides on apple trees from pictures of their
leaves. This errand is basic to ensure a high nature of the subsequent yields and is at present to a great extent performed by
specialists in the field, which can seriously constrain scale and include to costs.
In 2016, Davoud Ashourloo, Ali Akbar Matkan planned for building up an unearthly malady file that can distinguish the phases of
wheat leaf rust malady at different DS levels. To meet the point of the investigation, the reflectance spectra (350–2500 nm) of
tainted leaves with various side effect parts and DS levels were estimated with a spectroradiometer.
In 2015, Aakansha Rastogi,Ritika Arora,Shanu Sharma proposed the framework which takes a shot at preprocessing, highlight
extraction of leaf pictures from plant town dataset pursued by convolution neural system for grouping of ailment and suggesting
Pesticides utilizing Tensor stream innovation.
The principle two procedures that they use in our framework is android application with Java Web Services and Deep Learning.
They have use Convolution Neural Network with various layers five, four and three to prepare our model and android application as
a UI with JWS for association between these frameworks.
In 2015, Xingchun Chen and Ron made survey which examines science, worldwide dissemination and plant harm and yield
misfortunes in soyabean brought about by creepy crawly bugs, plant infections, nematodes and weeds. The connections among
bugs, weeds and illnesses are point by point. A soyabean incorporated irritation the executives (IPM) bundle of works on, covering
the yield from pre-planting to collect, is delineated. The impact of atmosphere changes on arthropod bugs, plant illnesses and weeds
are talked about.
In 2015, Rajleen Kaur,Dr. Sandeep Singh Kang proposed programmed location of ailments and ailing part present in the leaf
pictures of plants and even in the horticulture Crop creation. It is finished with headway of PC innovation which encourages in
cultivating to build the creation. Basically there is issue of recognition exactness and in neural system approach support vector
machine (SVM) is most recent classifier of that approach.
In 2014, Ms. Kiran R. Gavhale,Prof. Ujwalla Gawande,Mr. Kamal O.Hajari present about the picture handling methods utilized in
performing early recognition of plant illnesses through leaf highlights assessment. The goal of this work is to actualize picture
examination and characterization methods for extraction and characterization of leaf maladies. Leaf picture is caught and after that
handled to decide the status of each plant.
In 2010, Dheeb Al Bashish, Malik Braik, and Sulieman Bani-Ahmad We propose and assess a structure for identification of plant
leaf/stem sicknesses. Studies show that depending on unadulterated unaided eye perception of specialists to identify such ailments
can be restrictively costly, particularly in creating nations. Giving quick, programmed, modest and exact image processing based
answers for that undertaking can be of extraordinary sensible essentialness.
In 2008, Santanu Phadik ar and Jaya Sil depicts a product model framework for rice sickness recognition dependent on the tainted
pictures of different rice plants. Pictures of the tainted rice plants are caught by advanced camera and prepared utilizing picture
developing, picture division procedures to recognize contaminated pieces of the plants. At that point the tainted piece of the leaf has
been utilized for the arrangement reason utilizing neural system.

©IJRASET: All Rights are Reserved 1777

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

III. SYSTEM DESIGN

System design thought as the application of theory of the systems for the development of the project. System design defines the
architecture, data flow, use case, class, sequence and activity diagrams of the project development.

A. System Architecture
The below architecture diagram in figure 2 illustrates how the system is built and is the basic construction of the software7method.
Creations of such structures and documentation of these structures is the main responsible of software architecture.

Fig. 2 Architecture Diagram

Main Steps to build a architecture if proposed system are:

1) Convolution: Is the first layer to extract features from the input image and it learns the relationship between features using
kernel or filters with input images.
2) ReLU Layer: ReLU stands for the Rectified Linear Unit for a non-linear operation. The output is ƒ(x) = max (0, x). We use this
because to introduce the non-linearity to CNN.
3) Pooling Layer: it is used to reduce the number of parameters by down sampling and retain only the valuable information to
process further.
4) Flattening: We flatten our entire matrix into a vector like a vertical one. so, that it will be passed to the input layer.
5) Fully Connected Layer: we pass our flatten vector into input Layer .we combined these features to create a model. Finally, we
have an activation function such as softmax or sigmoid to classify the outputs.

B. Dataflow Diagram
Data flow diagram also referred as bubble graph. This diagram is useful for representing the system for all degree of constructions.
The figure is differentiated into parts which show maximizing data path & practical aspect. The below figure 3 shows the dataflow
of the proposed system.

Fig. 3 Dataflow Diagram

©IJRASET: All Rights are Reserved 1778

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

C. Flowchart

Fig. 4 Flowchart

D. General Explanation of Flowchart

1) The input test image is acquired and pre-processed in the next stage and then it is converted into array form for comparison.
2) The selected database is properly segregated and pre-processed and then renamed into proper folders.
3) The model is properly trained using CNN and then classification takes place.
4) The comparison of the test image and the trained model take place followed by the display of the result.
5) If there is a defect or disease in the plant the software displays the disease along with the remedy

IV. IMPLEMENTATION
Pre-processing and Training the model (CNN): The database is Pre-processed such as Image reshaping, resizing and conversion to
an array form. Similar processing is also done on the test image. A database consisting of different plant species is obtained, out of
which any image can be used as a test image for the software. The train database is used to train the model (CNN) so that it can
identify the test image and the disease it has .CNN has different layers that are Dense, Dropout, Activation, Flatten, Convolution2D,
and MaxPooling2D. After the model is trained successfully, the software can identify the disease if the plant species is contained in
the database. After successful training and pre-processing, comparison of the test image and trained model takes place to predict the
disease.

A. Image Acquisition
For training, Image is taken from database. And for testing, you can take image from camera at real time but in this work, we made
a particular folder on mobile from that image will be fetched by android application and send through java web services i.e. tomcat
server to server side system on which pre-processing is done and later on algorithm test that particular image.

B. Image Pre-Processing
Image should be processed before sending to the algorithm for testing and training purpose. For that purpose, in this work image is
scaled or resize into 150 x 150 dimensions. As we used color image so that we don’t need any color conversion techniques and that
pre-processed image is directly passed to algorithm for training and testing purpose.

C. Convolutional Neural Network (CNN)

Once pre-processing is done, then CNN is used for training purpose and after that we get trained model. That CNN method is
written with help of tensor flow. By using this model, we classify the image that system is getting after pre-processing of testing
image. Then we get particular disease name or healthy leaf name if there is no disease on that leaf and that disease name is send to
android application and with the help of that disease name we get particular pesticide name which help farmer to take respective
action in order to decrease percentage of disease.

©IJRASET: All Rights are Reserved 1779

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

CNN image classifications takes an input image, process it and classify it under certain categories Computers sees an input image as
array of pixels and it depends on the image resolution. Based on the image resolution, it will see h x w x d( h = Height, w = Width, d
= Dimension ). Eg., An image of 6 x 6 x 3 array of matrix of RGB (3 refers to RGB values) and an image of 4 x 4 x 1 array of
matrix of grayscale image.
Technically, deep learning CNN models to train and test, each input image will pass it through a series of convolution layers with
filters (Kernels), Pooling, fully connected layers (FC) and apply Softmax function to classify an object with probabilistic values
between 0 and 1

1) Convolution Layer: Convolution is the first layer to extract features from an input image. Convolution preserves the
relationship between pixels by learning image features using small squares of input data. It is a mathematical operation that
takes two inputs such as image matrix and a filter or kernel.

Fig. 5 Image matrix multiplies kernel or filter matrix.

2) Pooling Layer: Pooling layers section would reduce the number of parameters when the images are too large. Spatial pooling
also called subsampling or downsampling which reduces the dimensionality of each map but retains important information.
3) Fully Connected Layer: The layer we call as FC layer, we flattened our matrix into vector and feed it into a fully connected layer
like a neural network.

Fig. 6 After pooling layer, flattened as FC layer

In the above figure 6, the feature map matrix will be converted as vector (x1, x2, x3, …). With the fully connected layers, we
combined these features together to create a model. Finally, we have an activation function such as softmax to classify the outputs.

©IJRASET: All Rights are Reserved 1780

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

V. RESULTS

Fig. 7 Model training and validation accuracy

From the plot of accuracy in figure 7, we can see that the model could probably be trained a little more as the trend for accuracy on
both datasets is still rising for the last few epochs. We can also see that the model has not yet over-learned the training dataset,
showing comparable skill on both datasets.

Fig. 8 Model training and validation loss

From the plot of loss in figure 8, we can see that the model has comparable performance on both train and validation datasets
(labelled test). If these parallel plots start to depart consistently, it might be a sign to stop training at an earlier epoch.

Fig. 9 Predicted result of test image

Figure 9 shows the results of test image predicted. Results contain the details of disease identified and also prefer the pesticide for
the identified disease along with the description of the usage of pesticide.

©IJRASET: All Rights are Reserved 1781

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VII July 2021- Available at www.ijraset.com

VI. CONCLUSION
The proposed work was developed taking in mind the benefits of the farmers and agricultural sector .The developed system can
detect disease in plant and also provide the remedy that can be taken against the disease. By proper knowledge of the disease and the
remedy can be taken for improving the health of the plant .The proposed work is based on python and gives an accuracy of around
78%.The accuracy and the speed can be increased by use of Googles GPU for processing.

REFERENCES
[1] Monzurul Islam, Anh Dinh, Khan Wahid ,Pankaj Bhowmik“Detection of Potato Diseases Using Image Segmentation and Multiclass Support Vector Machine”,
IEEE 30th Canadian Conference on Electrical and Computer Engineering (CCECE),2017.
[2] Vijai Singh A.K. Misra ,"Detection of plant leaf diseases using image segmentation and soft computing techniques",INFORMATION PROCESSING IN
AGRICULTURE 4 (2017) 41–49,Elsevier.
[3] Sharada Prasanna Mohanty,David Hughes and Marcel Salathé,"Using Deep Learning for Image-Based Plant Disease Detection,"April 15, 2016.
[4] Davoud Ashourloo, Hossein Aghighi, Ali Akbar Matkan, Mohammad Reza Mobasheri, and Amir Moeini Rad,"An Investigation Into Machine Learning
Regression Techniques for the Leaf Rust Disease Detection Using Hyperspectral Measurement,"IEEE JOURNAL OF SELECTED TOPICS IN APPLIED
EARTH OBSERVATIONS AND REMOTE SENSING, VOL. 9, NO. 9, SEPTEMBER 2016.
[5] Noa Schor, Avital Bechar,Timea Ignat, Aviv Dombrovsky, Yigal Elad,and Sigal Berman,"Robotic Disease Detection in Greenhouses: Combined Detection of
Powdery Mildew and Tomato Spotted Wilt Virus",IEEE ROBOTICS AND AUTOMATION LETTERS, VOL. 1, NO. 1, JANUARY 2016.
[6] Lucas G. Nachtigall and Ricardo M. Araujo ,“Classification of Apple Tree Disorders Using Convolutional Neural Networks”, IEEE 28th International
Conference on Tools with Artificial Intelligence, 2016.
[7] Davoud Ashourloo, Hossein Aghighi, Ali Akbar Matkan, Mohammad Reza Mobasheri, and Amir Moeini Rad,"An Investigation Int0 Machine Learning
Regression Techniques for the Leaf Rust Disease Detection Using Hyperspectral Measurement,"IEEE JOURNAL OF SELECTED TOPICS IN APPLIED
EARTH OBSERVATIONS AND REMOTE SENSING, VOL. 9, NO. 9, SEPTEMBER 2016.
[8] Aakansha Rastogi,Ritika Arora,Shanu Sharma,“Leaf Disease Detection and garding using computer vision Technology & fuzzy logic",IEEE-IC on Signal
processing and Integrated network,2015.
[9] Xingchun Chen and Ron, “Monitoring Soybean Disease and Insect Infection Patterns in Nebraska”, International Conference on Granular Computing, IEEE
2015.
[10] Rajleen Kaur,Dr. Sandeep Singh Kang,"An Enhancement in Classifier Support Vector Machine to Improve Plant Disease Detection",IEEE 3rd International
Conference on MOOCs, Innovation and Technology in Education (MITE),2015.
[11] Ms. Kiran R. Gavhale,Prof. Ujwalla Gawande,Mr. Kamal O.Hajari,"Unhealthy Region of Citrus Leaf Detection Using Image Processing
Techniques",International Conference for Convergence of Technology - 2014,IEEE.
[12] Dheeb Al Bashish, Malik Braik, and Sulieman Bani-Ahmad,”A framework for detection and classification of plant and stem diseases”, International
Conference on Signal and Image Processing,2010.
[13] Santanu Phadik ar and Jaya Sil, “Rice Disease Identification using Pattern Recognition”, Proceedings of 11th International Conference on Computer and
Information Technology,2008.
[14] Hughes DP, Salathé M (2015) An open access repository of images on plant health to enable the development of mobile disease diagnostics. CoRR
abs/1511.08060.

©IJRASET: All Rights are Reserved 1782