DISEASE DETECTION OF VARIOUS LEAF USING IMAGE

PROCESSING TECHNIQUES

Project Report
Submitted in partial fulfilment for the award of degree of

BACHELOR OF TECHNOLOGY
in
INFORMATION TECHNOLOGY

Submitted By (Batch-5)

K. Sri Sai Bhargavi (18L31A1245)

D. Snigdha Bhavani (18L31A1279)
B. Pujitha (18L31A1266)
S. Siddharth(18L31A1290)

Under the esteemed guidance of

Mrs G.Jyothi
Assistant Professor IT Department

i
 Department of Information Technology

CERTIFICATE
This is to certify that the Project report entitled “DISEASE DETECTION OF VARIOUS
LEAF USING IMAGE PROCESSING TECHNIQUE” is a bonafide record of project work
carried out under my supervision by K. Sri Sai Bhargavi bearing Reg.No.18L31A1245, D. Snigdha
Bhavani bearing Reg.No.18L31A1279, B. Pujitha bearing Reg.No.18L31A1266, S. Siddharth
bearing Reg.No.18L31A1290 in partial fulfilment of the degree of Bachelor of Technology in
Information Technology of Vignan’s Institute of Information Technology(A) affiliated to
Jawaharlal Nehru Technology University Kakinada, during the academic year 2018-2022.

Signature Signature

Mrs G.Jyothi Dr.G.Rajendra Kumar

(Assistant Professor) (HOD)

EXTERNAL EXAMINER

ii
 DECLARATION

We here by declare that this project report entitled “Disease Detection of various Leaf using
Image Processing Technique” has undertaken by us for the fulfilment of Bachelor of Technology in
Information Technology. We declare that this project report has not been submitted anywhere in the
part of fulfilment for any degree of any other University.

PLACE:Visakhapatnam
DATE: K. Sri Sai Bhargavi (18L31A1245)
D. Snigdha Bhavani (18L31A1279)
B. Pujitha (18L31A1266)
S. Siddharth(18L31A1290)

iii
 ACKNOWLEDGEMENT

An endeavor over a long period can be successfully with the advice and support of

many well-wishers. I take this opportunity to express our gratitude and appreciation to all of them.

I express my sincere gratitude to my internal guide , Mrs. G.Jyothi , Assistant Professor

for his/her encouragement and cooperation in completion of my project. I am very fortunate in

getting the generous help and constant encouragement from him/her.

I would be very grateful to our project coordinator , Mrs. A. Sirisha , Assistant Professor
for the continuous monitoring of my project work.I truly appreciate for her time and effort helping me.

I would like to thank our Head of the Department , Dr. G. Rajendra Kumar , Professor

and all other teaching and non - teaching staff of the department for their cooperation

and guidance during my project.

I sincerely thank to Dr. B. Arundhati , Principal of VIGNAN’S INSTITUTE OF

INFORMATION TECHNOLOGY (A) for her inspiration to undergo this project.

I wanted to convey my sincere gratitude to Dr. V. Madhusudhan Rao , Rector of

VIGNAN’S INSTITUTE OF INFORMATION TECHNOLOGY (A) for allocating the

required resources and for the knowledge sharing during my project work.

I extended my grateful thanks to our honorable Chairman Dr. L. Rathaiah for giving

me an opportunity to study in his esteemed institution.

K. Sri Sai Bhargavi (18L31A1245)

D. Snigdha Bhavani(18L31A1279)
B. Pujitha (18L31A1266)
S. Siddharth (18L31A1290)

iv
 INFORMATION TECHNOLOGY

VISION :

To be a centre of excellence in high quality education and research producing globally

competent IT professionals with ethical /human values to meet the needs of Information
Technology sector and related services.

MISSION:

 To impart high quality of education through innovative teaching – learning practices

resulting in strong software and hardware knowledge and skills to enable students to
meet the challenges of IT profession.

 To facilitate faculty and students to carry out research work by providing necessary latest
facilities and a conductive environment.

 To mould the students into effective professionals with necessary communication skills ,
team spirit , leadership qualities , managerial skills , integrity , social and environmental
responsibility and lifelong learning ability with professional ethics and human values.

Vision and Mission of the Institute: Vision of the Institute (VIIT):

We envision to be recoginzed leader in technical education and shall aim at national excellence
by creating comptent and socially conscious technical manpower for the current and future
v
 Industrial requirements and development of the nation.
Mission of the Institute (VIIT):

We intend to fulfill our stated vision by

i. Introducing innovative practices of Teaching-Learning

ii. Undertaking research and development in thrust areas

iii. Continuously collaborating with industry

iv. Promoting strong set of ethical values

v. Serving the surrounding region and nation at large

Program Educational Objectives (PEOs)

PEO1: To work in core IT companies/allied industries, educational institutions, research organizations

and/or be entrepreneurs.

PEO2: To pursue higher education/ research in the field of Information Technology

PEO3: To demonstrate communication skills , team spirit , leadership qualities , managerial skills,
Integrity , social & environmental responsibility and lifelong learning ability , professional
ethics and human values in profession/career.

PSOs: Program Specific outcomes.

PSO1: Analyze and design the solutions for data storage & computing systems.

PSO2: Implement the solutions for network and communication problems of Information Technology.

vi
 PROGRAM OUTCOMES

Engineering Knowledge:

Apply the knowledge of mathematics science engineering fundamentals and mathematics, science,
PO1 engineering fundamentals, and an engineering specialization to the solution of complex
engineering problems engineering problems.

Problem analysis:

Identify, formulate, review research Literature, and analyze complex engineering problems
PO2
reaching substantiated conclusions using first principles of mathematics, and natural sciences, and
engineering sciences

Design/development of solutions:

Design solutions for complex engineering problems and design system components or processes
PO3 that meet the specified needs with appropriate consideration for the public health and safety, and
the cultural societal, and environmental considerations

Conduct investigations of complex problems:

PO4 Use research-based knowledge and research methods including design of experiments, analysis and
interpretation of data, and synthesis of the information to provide valid conclusions

Modern tool usage:

Create, select, and apply appropriate techniques, resources, and modern engineering and IT tools
PO5 including prediction and modeling to complex engineering activities with an understanding of the
limitations.

vii
 The engineer and society:

PO6 Apply reasoning informed by the contextual knowledge to assess societal, health, safety, legal
and cultural issues and the consequent responsibilities legal and cultural issues and the
consequent responsibilities relevant to the professional engineering practice

Environment and sustainability:

PO7 Understand the impact of the professional engineering solutions in societal and environmental
contexts, and demonstrate the knowledge of, and need for sustainable development and need for
sustainable development.

Ethics:

PO8 Apply ethical principles and commit to professional ethics and responsibilities and norms of the
engineering practice.

Individual and team work:

PO9 Function effectively as an individual and as a member or leader in diverse teams and individual,
and as a member or leader in diverse teams, and in multidisciplinary settings.

Communication:
PO10
Communicate effectively on complex engineering activities with the engineering community
and with society at large, such as, being able to comprehend and write effective reports and
design documentation, and write effective reports and design documentation, make effective
presentations, and give and receive clear instructions.

Project management and finance:

PO11
Demonstrate knowledge and understanding of the engineering and knowledge and
understanding of the engineering and management principles and apply these to one’s own
work, as a member and leader in a team, to manage projects and in multidisciplinary
environments.

Life-long learning:
PO12
Recognize the need for and have the preparation and ability to engage in independent and life-
long learning in the broadest context of technological change.

viii
ABSTRACT

ix
 Agriculture plays an vital role in Indian economy but owing to changing climatic
conditions, crops often get affected, as a result of which agricultural yield decreases
drastically. If the condition gets worse, crops may get vulnerable towards infections caused
by fungal, bacterial, virus, etc. diseases causing agents. The method that can be adopted to
prevent plant loss can be carried out by real-time identification of plant diseases. Our
proposed model provides an automatic method to determine leaf disease in a plant using a
trained dataset of pomegranate leaf images. The test set is used to check whether an image
entered into the system contains disease or not. If not, it is considered to be healthy,
otherwise the disease if that leaf is predicted and the prevention of plant disease is proposed
automatically. Further, the rodent causing disease is also identified with image analysis
performed on the image certified by biologists and scientists. This model provides an
accuracy of the results generated using different cluster sizes, optimized experimentally, with
image segmentation. Our model provides useful estimation and prediction of disease causing
agent with necessary precautions.

x
 INDEX

S. No Content Page No
1 Introduction
1.1 Introduction to project 1-3
1.2 Purpose of the project
2 Literature Survey 4-6
3. System Analysis
3.1 Existing System
3.2 Proposed System
3.3 Feasibility study 7-12
3.3.1 Technical Feasibility
3.3.2 Operational Feasibility
3.3.3 Economic Feasibility
4. System specifications
4.1 Functional Requirements
4.2 Non-functional requirements 13-21
4.3 Hardware requirements
4.4 Software requirements (SRS)

5. System Design
5.1 System Architecture
5.2 UML Diagrams 22 -45
5.3 Data Flow Diagram
6. System Implementation
6.1 Project Modules 46-67
6.2 Methodology (Algorithms )
6.3 Source Code
7. System Testing
7.1 Testing Methods 68-71
7.2 Test cases
8. Experimental Results 72-77
9. Conclusion & Future scope 79-80
10. Bibliography 81-83

xi
1.INTRODUCTION

1
1.1 INTRODUCTION TO PROJECT

The research paper referenced in tells about the various techniques involved in detecting
disease in a plant leaf through image processing techniques which involves image acquisition,
pre-processing of image and segmentation steps. The paper discussed methods to determine
the health status of each plant by considering the requirement of that plant. This method was
used so that the chemicals are applied only to those plants which require this treatment. This
method will result in reducing the expense on plant health to a larger scale. Aims at using
real time applications to change captured RGB image to Grey scale images since it increases
the clarity of the images and the disease is detected more efficiently. However, the paper
discusses how diseases in a pomegranate plant can be detected by applying proper
segmentation methods on the extracted images and identify the condition of a Pomegranate
plant.

1.2 PURPOSE OF THE PROJECT

Visual symptoms in the leaves are used to detect diseases in leaves. There is need for an
automatic system used for leaf disease detection because the disease cannot be detected by
naked eye. Our technology has grown to such an extent that a machine is capable enough to
predict the disease by looking at a high definition image of that leaves at its early stage itself.
The objective of our research is to find out the disease in a leaf. This process of detection can
be performed using image processing techniques which is method of forming a signal
processing for an inputted image by a scientist.We have used R programming language for
implementing image processing of the diseased leaf and predicting the disease. The dataset
used in our experiment contains images of infected and healthy pomegranate leaves. Model
Setup In order to conduct our experimentation, the dataset of plant leaf diseases is taken from
which is composed of 5358420 images of a single leaf disease. All the photographs present
on this site are clicked by professional photographers to provide an easy access to educational
applications.

2
The organization covers invasive species, forestry, agriculture and the images chosen for this
research are selected from an entire collection of 5359752 images as shown in the figure.
Model Libraries This section provides list of R libraries that were used to implement
proposed automated disease prediction model.

Fig .1 Datasets

3
2. LITERATURE SURVEY

4
1. Jayashri Patil ,Sachin Naik et.al, In this paper the proposed model purposes
preprocessing, division , Feature extraction and characterization methods for
pomegranate organic product illnesses. During feature extraction color, texture and
morphology features are used to identify and classify pomegranate fruit diseases.
SVM, ANN, KNN, PNN classifier used to detect and fungal and viral diseases. For
image segmentation K-means clustering is used Fuzzy c means gives highest accuracy.
Future scope is developing fully automated system with collaboration of agriculture
universities and research centers for upgrading the system with new diseases.

2. Shivaputra S.Panchal, Rutuja Sonar et.al, In this paper the proposed model is disease
detection using image acquisition, image preprocessing, image segmentation,
statistical feature extraction and classification.K-mean clustering algorithm is used for
segmentation. Support vector machine is used for classification of disease Image
processing is a form of signal processing the methodology of the proposed work
contains the five stages

 Image acquisition to get dataset of disease and non diseased images.

 Enhancement enhances the contrast of images
 K-mean is used for classification of object based on a set of features into k
number of classes.
 Feature extraction is utilized.

3. M. Bhange et.al, In this paper the proposed model gives an online apparatus has been
created to recognize natural product infections by transferring organic product picture
to the framework. Highlight extraction has been finished utilizing boundaries like tone,
morphology and CCV(colour cognizance vector). Bunching has been finished
utilizing the k-implies calculation SVM is utilized for characterization as
contaminated or non tainted.

4. J.D. Pujari et.al, In this paper the proposed model has taken number of harvest types
in particular, natural product , vegetable ,oat and business yields to distinguish
5
 contagious sickness on plant leaves. Various strategies have been For natural product
crop, k-means grouping is the division technique utilized , surface elements have been
centered around and ordered utilizing ANN and nearest neighbor algorithms.

 For vegetable yields, chan-vase strategy utilized for division, nearby paired designs
for surface component extraction and SVM and k-closest calculations for order.
 For business crops have been fragmented utilizing get cut algorithm. Wavelet based
feature extraction has been taken on utilizing Mahalnobis distance and PNN as
classifiers.
 For oat crops have been portioned utilizing k-means grouping and vigilant edge
finder.Color, shape ,texture ,color texture and random transform features have been
extracted. SVM and nearest neighbor classifiers used.
5. Garima Shrestha et.al in this paper the proposed model deployed the CNN to detect
the plant disease. Creators have effectively arranged 12 plant sicknesses and the
dataset of 3000 high goal RGB images were utilized for trial and error.The network
has 3 blocks of convolutional and pooling layers. This makes the networks
computationally also the F1 score of the model is 0.12 which is very low because of
higher number of false negative predictions.adopted for each type of crop.

6. H. Ali et.al in this paper the proposed model work aims to apply ΔE color difference
algorithm to separate the disease affected area and uses color histogram and textural
features to classify diseases achieving an overall accuracy of 99.9% [9]. A variety of
classifiers have been used such as fine KNN, Cubic SVM, Boosted tree and Bagged
tree classifiers. The bagged tree classifier out-performs the other classifiers achieving
99.5%, 100%, 100% accuracy on RGB, HSV and LBP features respectively.

7. D.A. Bashish, et.al in this paper the model opted for k-means segmentation for
partitioning the leaf image into four clusters using the squared Euclidean distances.
The method applied for feature extraction is Color Co-occurrence method for both
color and texture features. Finally, classification is completed using neural network
detection algorithm based on Back Propagation methodology. The overall system
disease detection and classification accuracy was found to be around 93%.

6
3. SYSTEM ANALYSIS

7
3.1 EXISTING METHODOLOGY

1. The most plant infections have or brought about by microscopic organisms, growths and
numerous destructive hurtful infections which we can’t distinguish with our unaided eye for
this we need innovative viewpoints, a few specialists in noticing and recognizing the plant
illnesses by utilization of some computational methodologies like PC vision, Man-made
brainpower and so forth.
2. These infections will annihilate the live plants affecting the shortage of the creation to us
and keeping the ranchers life in question which is the cultural issue as well.
3. This testing issue make agricultural nations specialists costly and tedious.

3.2 PROPOSED METHODOLOGY:

This section provides various steps involved in analyzing the image given by the user.
A. Basic Framework

• Image Acquisition - The images were collected and stored in a database of files.
Training and Test set were separated according to the images which are needed to be
trained and which are to be tested. Then a web based application was prepared to
upload an image from the test set and was loaded into the R application.

• Image Enhancement - It is used to increase contrast of the images present in the

training and test sets. This enhancement is used to obtain dimensions of the image.

• Image Segmentation - Input images are first converted into greyscale, obtaining
dimensions and removing noise. Then, applying K means algorithm. Figure 2 depicts
an example of k means applied to the dataset giving the prediction for different
number of color clusters used.

8
 Fig. 3 Flowchart

• Feature Extraction - We calculated the average value of each image pixel by

calculating their separate R G B values and finding the mean for the entire image
containing various pixels.

• Image Classification - The SVM classifier is used to identify the classes, which are
closely connected to the known and trained classes. The Support vector machine
creates the optimal separating hyper plane between the classes using the training data.

• Image Accuracy- Calculate the accuracy of each image of the dataset by varying
the number of clusters and forming a plot showing the contrast between the accuracy
of results obtained by each cluster value.
Algorithms K-Means algorithm was applied for image segmentation. In this algorithm the
objects are broken into some points and depending on the frequency of these points, number
of clusters are decided. Then these clustered are portioned according to the minimum distance
each point has with the clusters. The algorithm of finding the minimum distance is then
performed. Centroid formation of the selected points continues and it discontinues when
points get repeated. In R the function that performs k means clustering uses many sub
algorithms to find k means clusters but the default used in R is “Hartigan-Wong”. It is the
most preferred algorithm because it is time efficient. Figure 3 depicts the implementation of
image enhancement (a) and segmentation with clusters 5, 7 and 11(b,c,d). SVM is used for

9
classification of pomegranate leaf. This algorithm is used to analyze the data set used and
perform the linear vector binding between the categories of the images, i.e. their respective
diseases. A hyper plane is formed by the training points that bind with each other to form
binary classifiers. These are then matched and mismatched and provides a predicted disease
according to any feature that can be extractable.

10
3.3 FEASIBILITY STUDY:
Preliminary investigation examine project feasibility, the likelihood the system will be useful
to the organization. The main objective of the feasibility study is to test the Technical,
Operational and Economical feasibility for adding new modules and debugging old running
system. All system is feasible if they are unlimited resources and infinite time. There are
aspects in the feasibility study portion of the preliminary investigation:
 Technical Feasibility
 Operation Feasibility
 Economical Feasibility

3.3.1. Technical Feasibility

The technical issue usually raised during the feasibility stage of the investigation
includes the following:
 Does the necessary technology exist to do what is suggested?
 Do the proposed equipments have the technical capacity to hold the data
required to use the new system?
 Will the proposed system provide adequate response to inquiries, regardless of
the number or location of users?
 Can the system be upgraded if developed?
 Are there technical guarantees of accuracy, reliability, ease of access and data
security?

Earlier no system existed to cater to the needs of ‘Secure Infrastructure

Implementation System’. The current system developed is technically feasible. It is a web
based user interface for audit workflow at NIC-CSD. Thus it provides an easy access to the
users. The database’s purpose is to create, establish and maintain a workflow among various
entities in order to facilitate all concerned users in their various capacities or roles.
Permission to the users would be granted based on the roles specified. Therefore, it provides
the technical guarantee of accuracy, reliability and security. The software and hard
requirements for the development of this project are not many and are already available in-
house at NIC or are available as free as open source. The work for the project is done with the

11
current equipment and existing software technology. Necessary bandwidth exists for
providing a fast feedback to the users irrespective of the number of users using the system.

3.3.2. Operational Feasibility

Proposed projects are beneficial only if they can be turned out into information
system. That will meet the organization’s operating requirements. Operational feasibility
aspects of the project are to be taken as an important part of the project implementation.
Some of the important issues raised are to test the operational feasibility of a project includes
the following: -
 Is there sufficient support for the management from the users?
 Will the system be used and work properly if it is being developed and implemented?
 Will there be any resistance from the user that will undermine the possible
application benefits?
This system is targeted to be in accordance with the above-mentioned issues.
Beforehand, the management issues and user requirements have been taken into consideration.
So there is no question of resistance from the users that can undermine the possible
application benefits.
The well-planned design would ensure the optimal utilization of the computer resources and
would help in the improvement of performance status.

3.3.3. Economic Feasibility

A system can be developed technically and that will be used if installed must still be a good
investment for the organization. In the economical feasibility, the development cost in
creating the system is evaluated against the ultimate benefit derived from the new systems.
Financial benefits must equal or exceed the costs.
The system is economically feasible. It does not require any addition hardware or
software. Since the interface for this system is developed using the existing resources and
technologies available at NIC, There is nominal expenditure and economical feasibility for
certain.

12
4. SYSTEM SPECIFICATIONS

13
4.1 FUNCTIONAL REQUIREMENTS:

OUTPUT DESIGN

Outputs from computer systems are required primarily to communicate the results of
processing to users. They are also used to provides a permanent copy of the results for later
consultation. The various types of outputs in general are:
 External Outputs, whose destination is outside the organization.
 Internal Outputs whose destination is with in organization and they are the
 User’s main interface with the computer.
 Operational outputs whose use is purely with in the computer department.
Interface outputs, which involve the user in communicating directly.

OUTPUT DEFINITION
The outputs should be defined in terms of the following points:

 Type of the output

 Content of the output
 Format of the output
 Location of the output
 Frequency of the output
 Volume of the output
 Sequence of the output

It is not always desirable to print or display data as it is held on a computer. It should

be decided as which form of the output is the most suitable.

For Example
 Will decimal points need to be inserted
 Should leading zeros be suppressed.

14
Output Media:
In the next stage it is to be decided that which medium is the most appropriate for the
output. The main considerations when deciding about the output media are:
 The suitability for the device to the particular application.
 The need for a hard copy.
 The response time required.
 The location of the users
 The software and hardware available.
Keeping in view the above description the project is to have outputs mainly coming under the
category of internal outputs. The main outputs desired according to the requirement
specification are:
The outputs were needed to be generated as a hot copy and as well as queries to be viewed on
the screen. Keeping in view these outputs, the format for the output is taken from the outputs,
which are currently being obtained after manual processing. The standard printer is to be
used as output media for hard copies.

INPUT DESIGN

Input design is a part of overall system design. The main objective during the input design is
as given below:

 To produce a cost-effective method of input.

 To achive the highest possible level of accuracy.
 To ensure that the input is acceptable and understood by the user.

INPUT STAGES:
The main input stages can be listed as below:

 Data recording
 Data transcription
 Data conversion
 Data verification
 Data control

15
  Data transmission
 Data validation
 Data correction

INPUT TYPES:
It is necessary to determine the various types of inputs. Inputs can be categorized as follows:
 External inputs, which are prime inputs for the system.
 Internal inputs, which are user communications with the system.
 Operational, which are computer department’s communications to the system?
 Interactive, which are inputs entered during a dialogue.

INPUT MEDIA:
At this stage choice has to be made about the input media. To conclude about the
input media consideration has to be given to;
 Type of input
 Flexibility of format
 Speed
 Accuracy
 Verification methods
 Rejection rates
 Ease of correction
 Storage and handling requirements
 Security
 Easy to use
 Portability

Keeping in view the above description of the input types and input media, it can be
said that most of the inputs are of the form of internal and interactive. As
Input data is to be the directly keyed in by the user, the keyboard can be considered to be the
most suitable input device.

16
 ERROR AVOIDANCE
At this stage care is to be taken to ensure that input data remains accurate form the
stage at which it is recorded upto the stage in which the data is accepted by the system. This
can be achieved only by means of careful control each time the data is handled.

ERROR DETECTION
Even though every effort is make to avoid the occurrence of errors, still a small
proportion of errors is always likely to occur, these types of errors can be discovered by using
validations to check the input data.
DATA VALIDATION
Procedures are designed to detect errors in data at a lower level of detail. Data
validations have been included in the system in almost every area where there is a possibility
for the user to commit errors. The system will not accept invalid data. Whenever an invalid
data is keyed in, the system immediately prompts the user and the user has to again key in the
data and the system will accept the data only if the data is correct. Validations have been
included where necessary.
The system is designed to be a user friendly one. In other words the system has been
designed to communicate effectively with the user. The system has been designed with pop
up menus.

USER INTERFACE DESIGN

It is essential to consult the system users and discuss their needs while designing the
user interface:

USER INTERFACE SYSTEMS CAN BE BROADLY CLASIFIED AS:

1. User initiated interface the user is in charge, controlling the progress of the user/computer
dialogue. In the computer-initiated interface, the computer selects the next stage in the
interaction.
2. Computer initiated interfaces

In the computer initiated interfaces the computer guides the progress of the
user/computer dialogue. Information is displayed and the user response of the computer
takes action or displays further information.
17
 USER_INITIATED INTERGFACES
User initiated interfaces fall into tow approximate classes:
1. Command driven interfaces: In this type of interface the user inputs commands or queries
which are interpreted by the computer.
2. Forms oriented interface: The user calls up an image of the form to his/her screen and fills in
the form. The forms oriented interface is chosen because it is the best choice.

ERROR MESSAGE DESIGN:

The design of error messages is an important part of the user interface design. As
user is bound to commit some errors or other while designing a system the system should be
designed to be helpful by providing the user with information regarding the error he/she has
committed.

This application must be able to produce output at different modules for different
inputs.

4.2. PERFORMANCE REQUIREMENTS

Performance is measured in terms of the output provided by the application.

Requirement specification plays an important part in the analysis of a system. Only

when the requirement specifications are properly given, it is possible to design a system,
which will fit into required environment. It rests largely in the part of the users of the
existing system to give the requirement specifications because they are the people who finally
use the system. This is because the requirements have to be known during the initial stages
so that the system can be designed according to those requirements. It is very difficult to
change the system once it has been designed and on the other hand designing a system, which
does not cater to the requirements of the user, is of no use.

The requirement specification for any system can be broadly stated as given below:
 The system should be able to interface with the existing system
 The system should be accurate
 The system should be better than the existing system
18
  The existing system is completely dependent on the user to perform all the duties.

4.3 NON-FUNCTIONAL REQUIREMENTS:

1. Security:
 System needs to control the user access and session.
 It needs to store the data in a secure location and stored in a secure format.
 It requires a secure communication channel for the data.
2. Concurrency and Capacity:
System should be able to handle multiple computations executing simultaneously, and
potentially interacting with each other.
3. Performance:
Performance is generally perceived as a time expectation. This is one of the most important
considerations especially when the project is in the architecture phase.
4. Reliability:
It is necessary to ensure and notify about the system transactions and processing as simple
as keep a system log will increase the time and effort to get it done from the very beginning.
Data should be transferred in a reliable way and using trustful protocols.
5. Maintainability:
Well-done system is meant to be up and running for long time. Therefore, it will regularly
need preventive and corrective maintenance. Maintenance might signify scalability to grow
and improve the system features and functionalities.
6. Usability:
End user satisfaction and acceptance is one of the key pillars that support a project success.
Considering the user experience requirements from the project conception is a win bet, and it
will especially save a lot of time at the project release, as the user will not ask for changes or
even worst misunderstandings.
7. Documentation:
All projects require a minimum of documentation at different levels. In many cases the users
might even need training on it, so keeping good documentation practices and standards will
do this task spread along the project development; but as well this must be establish since the
project planning to include this task in the list.

19
HARDWARE REQUIREMENTS:

 Any System with 2.8 GHZ Processor

 2 GB RAM
 160 GB Hard Disk
 Keyboard
 Mouse

4.4 SOFTWARE SPECIFICATIONS(SRS):

The software, Site Explorer is designed for management of web sites from a
remote location.

INTRODUCTION

Purpose: The main purpose for preparing this document is to give a general insight into the
analysis and requirements of the existing system or situation and for determining the
operating characteristics of the system.

Scope: This Document plays a vital role in the development life cycle (SDLC) and it
describes the complete requirement of the system. It is meant for use by the developers and
will be the basic during testing phase. Any changes made to the requirements in the future
will have to go through formal change approval process.

DEVELOPERS RESPONSIBILITIES OVERVIEW:

The developer is responsible for:

Developing the system, which meets the SRS and solving all the requirements of the
system?
Demonstrating the system and installing the system at client's location after the
acceptance testing is successful.
Submitting the required user manual describing the system interfaces to work on it and
also the documents of the system.

20
 Conducting any user training that might be needed for using the system.
Maintaining the system for a period of one year after installation.

SOFTWARE REQUIREMENTS:

 Windows Operating System

 Python 3.5 or above
 Anaconda 3.0 or pycharm
 Data Sets
 Jupyter

21
5. SYSTEM DESIGN

22
 INTRODUCTION

Software design sits at the technical kernel of the software engineering process and is
applied regardless of the development paradigm and area of application. Design is the first
step in the development phase for any engineered product or system. The designer’s goal is to
produce a model or representation of an entity that will later be built. Beginning, once system
requirement have been specified and analyzed, system design is the first of the three technical
activities -design, code and test that is required to build and verify software.

The importance can be stated with a single word “Quality”. Design is the place where
quality is fostered in software development. Design provides us with representations of
software that can assess for quality. Design is the only way that we can accurately translate a
customer’s view into a finished software product or system. Software design serves as a
foundation for all the software engineering steps that follow. Without a strong design we risk
building an unstable system – one that will be difficult to test, one whose quality cannot be
assessed until the last stage.

5.1 SYSTEM ARCHITECTURE

Fig. 5 System Architecture

23
Programming configuration sits at the specialized part of the programming interaction and is
applied paying little mind to the advancement worldview and space of use. Configuration is
the initial phase in the advancement stage for any designed item or framework. The’s
originator will likely create a model or portrayal of a substance that will later be assembled.
Starting, when framework necessity have been indicated and examined, framework
configuration is the first of the three specialized exercises – plan, code and test that is needed
to assemble and check programming. The significance can be expressed with a solitary word
“Quality”. Configuration is where quality is encouraged in programming improvement.
Configuration furnishes us with portrayals of programming that can evaluate for quality.
Configuration is the solitary way that we can precisely make an interpretation of a client’s
view into a completed programming item or framework. Programming configuration fills in
as an establishment for all the programming steps that follow. Without a solid plan we hazard
assembling a shaky framework – one that will be hard to test, one whose quality can’t be
surveyed until the last stage. During plan, reformist refinement of information structure,
program structure, and procedural subtleties are created assessed and archived. Framework
configuration can be seen from one or the other specialized or venture the executives
viewpoint. According to the specialized perspective, plan is included four exercises –
engineering plan, information structure plan, interface plan and procedural plan.

5.2 UML DIAGRAMS

UML (Unified Modeling Language) is a general-purpose, graphical modeling language in

the field of Software Engineering. UML is used to specify, visualize, construct, and
document the artifacts (major elements) of the software system. It was initially developed by
Grady Booch, Ivar Jacobson, and James Rumbaugh in 1994-95 at Rational software, and its
further development was carried out through 1996. In 1997, it got adopted as a standard by
the Object Management Group.

What is UML

The UML stands for Unified modeling language, is a standardized general-purpose visual
modeling language in the field of Software Engineering. It is used for specifying, visualizing,
constructing, and documenting the primary artifacts of the software system. It helps in
designing and characterizing, especially those software systems that incorporate the concept
of Object orientation. It describes the working of both the software and hardware systems.

24
The Object Management Group (OMG) is an association of several companies that controls
the open standard UML. The OMG was established to build an open standard that mainly
supports the interoperability of object-oriented systems. It is not restricted within the
boundaries, but it can also be utilized for modeling the non-software systems. The OMG is
best recognized for the Common Object Request Broker Architecture (CORBA) standards.

Goals of UML

o Since it is a general-purpose modeling language, it can be utilized by all the modelers.

o UML came into existence after the introduction of object-oriented concepts to

systemize and consolidate the object-oriented development, due to the absence of
standard methods at that time.

Characteristics of UML

The UML has the following features:

C++ vs Java
o It is a generalized modeling language.

o It is distinct from other programming languages like C++, Python, etc.

o It is interrelated to object-oriented analysis and design.

Conceptual Modeling

Before moving ahead with the concept of UML, we should first understand the basics of the
conceptual model.

A conceptual model is composed of several interrelated concepts. It makes it easy to

understand the objects and how they interact with each other. This is the first step before
drawing UML diagrams.

The vocabulary of the UML includes three sorts of building square

1. Things
2. Relationships
3. Diagrams
Things in the UML
There are four sorts of things in the UML. They are
1. Structural Things
There are seven sorts of basic things.Class:Class is a description of a lot of articles that share
similar traits, activities, connections, and semantics.
25
Class:
A Class is a set of identical things that outlines the functionality and properties of an object.
It also represents the abstract class whose functionalities are not defined. Its notation is as
follows

Object:
An individual that describes the behavior and the functions of a system. The notation of the
object is similar to that of the class.

Interface:
A set of operations that describes the functionality of a class, which is implemented
whenever an interface is implemented.

Collaboration:
It represents the interaction between things that is done to meet the goal. It is symbolized as
a dotted ellipse with its name written inside it.

26
Use case:
Use case is the core concept of object-oriented modeling. It portrays a set of actions executed
by a system to achieve the goal.

Actor:
It comes under the use case diagrams. It is an object that interacts with the system, for
example, a user.

Component:
It represents the physical part of the system.

Node:
A physical element that exists at run time.

27
2. Behavioral Things
They are the verbs that encompass the dynamic parts of a model. It depicts the behavior of a
system. They involve state machine, activity diagram, interaction diagram, grouping things,
annotation things

State Machine:
It defines a sequence of states that an entity goes through in the software development
lifecycle. It keeps a record of several distinct states of a system component.

Activity Diagram:
It portrays all the activities accomplished by different entities of a system. It is represented
the same as that of a state machine diagram. It consists of an initial state, final state, a
decision box, and an action notation.

28
Interaction Diagram:
It is used to envision the flow of messages between several components in a system.

3. Grouping Things
It is a method that together binds the elements of the UML model. In UML, the package is
the only thing, which is used for grouping.
Package:
Package is the only thing that is available for grouping behavioral and structural things.

4. Annotation Things
It is a mechanism that captures the remarks, descriptions, and comments of UML model
elements. In UML, a note is the only Annotational thing.
Note:
It is used to attach the constraints, comments, and rules to the elements of the model. It is a
kind of yellow sticky note.

29
Relationships
It illustrates the meaningful connections between things. It shows the association between
the entities and defines the functionality of an application. There are four types of
relationships given below:

Dependency:
Dependency is a kind of relationship in which a change in target element affects the source
element.It depicts the dependency from one entity to another.
It is denoted by a dotted line followed by an arrow at one side as shown below,

Association:
A set of links that associates the entities to the UML model. It tells how many elements are
actually taking part in forming that relationship.
It is denoted by a dotted line with arrowheads on both sides to describe the relationship with
the element on both sides.

Generalization:
It portrays the relationship between a general thing (a parent class or superclass) and a
specific kind of that thing (a child class or subclass). It is used to describe the concept of
inheritance.
It is denoted by a straight line followed by an empty arrowhead at one side.

Realization:
It is a semantic kind of relationship between two things, where one defines the behavior to
be carried out, and the other one implements the mentioned behavior. It exists in interfaces.
It is denoted by a dotted line with an empty arrowhead at one side.

30
Diagrams
The diagrams are the graphical implementation of the models that incorporate symbols and
text. Each symbol has a different meaning in the context of the UML diagram. There are
thirteen different types of UML diagrams that are available in UML 2.0.
UML diagrams are classified into three categories that are given below:
1. Structural Diagram
2. Behavioral Diagram
3. Interaction Diagram
Structural Diagram: It represents the static view of a system by portraying the structure of a
system. It shows several objects residing in the system. Following are the structural diagrams
given below:
 Class diagram
 Object diagram
 Package diagram
 Component diagram
 Deployment diagram
Behavioral Diagram: It depicts the behavioral features of a system. It deals with dynamic
parts of the system. It encompasses the following diagrams:
 Activity diagram
 State machine diagram
 Use case diagram
Interaction diagram: It is a subset of behavioral diagrams. It depicts the interaction between
two objects and the data flow between them. Following are the several interaction diagrams
in UML:
 Timing diagram
 Sequence diagram
 Collaboration diagram

USE CASE DIAGRAM

USE CASE Chart A social graph that shows a bunch of items and their connections.

31
Purpose of Use Case Diagrams:
The principle reason for class graphs is to fabricate a static perspective on an application.
It is the solitary graph that is generally utilized for development, and it very well may be
planned with object-situated dialects. It is quite possibly the most famous UML graphs.
Following are the reason for class graphs given underneath:
1.It examinations and plans a static perspective on an application.
2.It portrays the significant obligations of a framework.
3.It is a base for part and arrangement charts.
4.It consolidates forward and figuring out
Utilization of Utilization CASE graphs:
1.Analyzing the prerequisites of a framework
2.High-level visual programming planning
3.Capturing the functionalities of a framework
4.Modeling the fundamental thought behind the framework
5.Forward and figuring out of a framework utilizing different experiments.

 User will start the application.

 After starting the application the user can select image from the trained data set.
 After selecting the image the user can upload it to the lab.
 Lab or service center will process the image by using knn algorithm.

32
  View image and send the report to the user.
 After processing the image the lab will send report to the user.
 User will receive the image with a disease detector.
CLASS DIAGRAM

Purpose of Class Diagram

The reason for class chart is to show the static perspective on an application. Class
outlines are the solitary graphs which can be straightforwardly planned with object-arranged
dialects and in this way broadly utilized at the hour of development.
UML outlines like action graph, grouping chart can just give the arrangement stream
of the application, anyway class graph is somewhat unique. It is the most mainstream UML
chart in the coder local area.
The accompanying focuses ought to be recalled while drawing a class graph
1.The name of the class graph ought to be significant to depict the part of the framework.
2.Each component and their connections ought to be distinguished ahead of time.
3.Responsibility (properties and strategies) for each class ought to be obviously distinguished
4.For each class, least number of properties ought to be determined, as pointless properties
will make the chart muddled.
5.Use notes at whatever point needed to portray some part of the chart. Toward the finish of
the drawing it ought to be reasonable to the designer/coder.
6.Finally, prior to making the last form, the graph ought to be drawn on plain paper and
improved whatever number occasions as would be prudent to make it right.

33
 User will start the application.
 After starting the application the user can select image from the trained data set.
 Then the user can select image id and everything the user can accessing from the
trained data set.
 By using image() class it will accessing the image.
 Then the image can be process by using class Get Image().
 After selecting the image the user can upload image to the knn algorithm.
 The algorithm can accessing all the information and it will send it to the user.
 It will accessing through id only. Id means the trained data set can accessing all the
numbers visible to the user purpose only.
 Lab or service center will process the image by using knn algorithm.
 After using the knn algorithm it will processing the data sets.
 By using class only the algorithm can accessing the image by Get Datasets().
 In our project we have two data sets trained datasets and test datasets.
 After checking the algorithm can send the image view to the user.
 View image and send the report to the user.
 After processing the image the lab will send report to the user.
 User will receive the image with a disease detector.
 It gives the detail information of class and its process

34
 SEQUENCE DIAGRAM

A social outline that shows a cooperation, stressing the time requesting of messages.

PURPOSE OF SEQUENCE DIAGRAM

1.To model significant level connection among dynamic articles inside a framework.
2.To model communication among objects inside a joint effort understanding a utilization
case.
3.It either models nonexclusive connections or some specific occasions of association.

 Firstly, the user can start the application.

 After starting the application the user can upload the image to the server.
 By using id only it can take it because the dataset having two both the trained dataset
and test dataset.
 In trained dataset having all the information about leafs and evrything.
 After upload the image it will processing the image.
 By the id number it will take the leafs and everything.
 After receiving the image it will processing the image.
 It will processing the image then the image have any disease detection.

35
  If there is any disease detection or healthy it will upload image to the user.
 The user can accesing the image, the image can view by the user.
 The user can read the id and leaf has any diseased or healthy.
 The algorithm can give meassage to the user the leaf is diseased or healthy.

COMPONENT DIAGRAM

A part chart is utilized to separate a huge item arranged framework into the more modest
segments, in order to make them more reasonable. It displays the actual perspective on a
framework, for example, executables, records, libraries, and so forth that dwells inside the
hub. The execution subtleties of a part are covered up, and it requires an interface to execute
capacity. It resembles a black box whose conduct is clarified by the gave and required
interfaces.
Why use component diagram:
The segment graphs have momentous significance. It is utilized to portray the usefulness
and conduct of the relative multitude of segments present in the framework, dissimilar to
different charts that are utilized to address the design of the framework, working of a
framework, or essentially the actual framework.
In UML, the segment graph depicts the conduct and association of segments at any moment
of time. The framework can't be imagined by any individual segment, yet it tends to be by the
assortment of parts.
Following are a few purposes behind the prerequisite of the segment chart:
1. It depicts the segments of a framework at the runtime.
2. It is useful in testing a framework.
3. It imagines the connections between a few associations.
Where to use component diagram?
1. To model the components of the system.
2. To model the schemas of a database.
3. To model the applications of an application.
4. To model the system's source code.

36
 The user can take all the details of the application.
 After taking all the datails of the application the user can start the application.
 After checking the app details by the user the it will give a message to the user as
update the app details.
 After starting the application the lab can updating all the details given by the user.
 After applying the algorithm, the algorithm can create a new source to the user.
 Once algorithm can be applied it will give message to the user like source created
successfully.
 The user can requested to the lab to enter the existing sources like all the datasets
(trained dataset and test dataset).
 After requesting the details of the existing resources it will give message from the lab
as to update the address from the algorithm.
 The image can accessing from the lab it will take reports from the lab it can accessing
all the information from the user then the reports can be send it to the user.
 After the user can request to the lab the reports will send it to the user and give a
message to the user as view reports as per your request.
 The user can accessing all from the algorithm only. By using algorithm we can set all
the datasets in the datasets.

37
  In that datasets have both the trained datasets and test datasets.
 By using this the lab can send message to the user like a message.

ACTIVITY DIAGRAM

In UML, the movement outline is utilized to show the progression of control inside the
framework as opposed to the execution. It displays the simultaneous and successive exercises.
The movement outline helps in imagining the work process starting with one action then onto
the next. It put accentuation on the state of stream and the request in which it happens. The
stream can be consecutive, extended, or simultaneous, and to manage such sorts of streams,
the action outline has thought of a fork, join, and so forth It is likewise named as an article
arranged flowchart. It incorporates exercises made out of a bunch of activities or tasks that
are applied to show the social graph.

Notification of activity diagram:

Activity diagram constitutes following notations:
Initial State:
It depicts the initial stage or beginning of the set of actions.
Final State:
It is the stage where all the control flows and object flows end.
Decision Box:
It makes sure that the control flow or object flow will follow only one path.
Action Box:
It represents the set of actions that are to be performed.

38
When to use activity diagram:
An activity diagram can be used to portray business processes and workflows. Also, it used
for modeling business as well as the software. An activity diagram is utilized for the
followings:
1. To graphically model the workflow in an easier and understandable way.
2. To model the execution flow among several activities.
3. To model comprehensive information of a function or an algorithm employed within the
system.
4. To model the business process and its workflow.
5. To envision the dynamic aspect of a system.
6. To generate the top-level flowcharts for representing the workflow of an application.
7. To represent a high-level view of a distributed or an object-oriented system

39
 The user can capturing the image.
 After capturing the image it will input image to the lab the it will accesiing all the
algorithm.
 After input image, then it will apply all th algorithms then the image is healthy it will
give success message and the image is diseased then it will give message as failure.
 Once it is failure then it will give to the input image and set all the details of the leaf
image then it will re image.
 Once it is success then it will move to the both classifier and report.
 In classifier, it will checking all the filtering, enchancement, segmentation, feature
extrraction and then it will stop the processing image.
 In report, it will checking all the filtering, enchancement, segmentation, feature
extrraction and then it will stop the processing image.
 After checking the classifier and report whether the leaf is diseased or not it will
checking all the report of the image.

40
 STATECHART DIAGRAM
The state machine diagram is also called the Statechart or State Transition diagram, which
shows the order of states underwent by an object within the system. It captures the software
system's behavior. It models the behavior of a class, a subsystem, a package, and a complete
system.
Following are the types of a state machine diagram that are given below:
1. Behavioural state machine:
The behavioral state machine diagram records the behavior of an object within the system.
It depicts an implementation of a particular entity. It models the behavior of the system.
2. Protocal state machine:
It captures the behavior of the protocol. The protocol state machine depicts the change in
the state of the protocol and parallel changes within the system. But it does not portray the
implementation of a particular component.

 In this state chart diagram, in dataset all the leaf can be inherited all the leafs can be in
the datasets ( trained dataset and test dataset ).
 In datasets we have both the trained dataset and test dataset.
 In trained dataset it has Train.csv and it can be send to the Train.record.
 In test dataset it has Test.csv and it can be send to the Test.record.

41
 Both the Train.record and Test.record these will combine with the training the
different layer models.
 These training the different layer models it will combine with the evaluating the
different layer model.

COLLABORATION DIAGRAM

 The user can input the leaf image.

 After input the image, the image can be pre processing to the lab.
 We calculated the average value of each image pixel by calculating their separate
R G B values and finding the mean for the entire image containing various pixels.
 In database all the training and test image.
 Both the training and test dataset that will send it to the knn classification.
 By using knn algorithm it will check whether the leaf is diseased or not.
 When the leaf is healthy it will give normal then it will give result as healthy image.
 When the leaf is diseased it will give abnormal them it will give result as diseased
detection.

42
 5.3 DATAFLOW DIAGRAM

A data flow diagram is graphical tool used to describe and analyze movement of data
through a system. These are the central tool and the basis from which the other components
are developed. The transformation of data from input to output, through processed, may be
described logically and independently of physical components associated with the
system. These are known as the logical data flow diagrams. The physical data flow diagrams
show the actual implements and movement of data between people, departments and
workstations. A full description of a system actually consists of a set of data flow diagrams
The idea behind the explosion of a process into more process is that understanding at
one level of detail is exploded into greater detail at the next level. This is done until further
explosion is necessary and an adequate amount of detail is described for analyst to understand
the process.
A DFD is also known as a “bubble Chart” has the purpose of clarifying system
requirements and identifying major transformations that will become programs in system
design. So it is the starting point of the design to the lowest level of detail. A DFD consists
of a series of bubbles joined by data flows in the system.

DFD SYMBOLS:
In the DFD, there are four symbols
1. A square defines a source(originator) or destination of system data.
2. An arrow identifies data flow. It is the pipeline through which the information flows.
3. A circle or a bubble represents a process that transforms incoming data flow into outgoing
data flows.
4. An open rectangle is a data store, data at rest or a temporary repository of data.

CONSTRUCTING A DFD:
Several rules of thumb are used in drawing DFD’S:

1. Process should be named and numbered for an easy reference. Each name should be
representative of the process.

43
2. The direction of flow is from top to bottom and from left to right. Data traditionally flow
from source to the destination although they may flow back to the source. One way to
indicate this is to draw long flow line back to a source. An alternative way is to repeat the
source symbol as a destination. Since it is used more than once in the DFD it is marked with
a short diagonal.
3. When a process is exploded into lower level details, they are numbered.
4. The names of data stores and destinations are written in capital letters. Process and dataflow
names have the first letter of each work capitalized

A DFD typically shows the minimum contents of data store. Each data store should
contain all the data elements that flow in and out.
Questionnaires should contain all the data elements that flow in and out. Missing
interfaces redundancies and like is then accounted for often through interviews.

SAILENT FEATURES OF DFD’S

1. The DFD shows flow of data, not of control loops and decision are controlled considerations
do not appear on a DFD.
2. The DFD does not indicate the time factor involved in any process whether the dataflow take
place daily, weekly, monthly or yearly.
3. The sequence of events is not brought out on the DFD.
TYPES OF DATA FLOW DIAGRAMS
1. Current Physical
2. Current Logical
3. New Logical
4. New Physical

CURRENT PHYSICAL:
In Current Physical DFD proecess label include the name of people or their positions or
the names of computer systems that might provide some of the overall system-processing
label includes an identification of the technology used to process the data. Similarly data
flows and data stores are often labels with the names of the actual physical media on which
data are stored such as file folders, computer files, business forms or computer tapes.

CURRENT LOGICAL:
44
 The physical aspects at the system are removed as mush as possible so that the current
system is reduced to its essence to the data and the processors that transform them regardless
of actual physical form.

NEW LOGICAL:
This is exactly like a current logical model if the user were completely happy with he
user were completely happy with the functionality of the current system but had problems
with how it was implemented typically through the new logical model will differ from current
logical model while having additional functions, absolute function removal and inefficient
flows recognized.

NEW PHYSICAL:
The new physical represents only the physical implementation of the new system.
RULES GOVERNING THE DFD’S
PROCESS
1) No process can have only outputs.
2) No process can have only inputs. If an object has only inputs than it must be a sink.
3) A process has a verb phrase label.
DATA STORE
1) Data cannot move directly from one data store to another data store, a process must move
data.
2) Data cannot move directly from an outside source to a data store, a process, which receives,
must move data from the source and place the data into data store
3) A data store has a noun phrase label.

SOURCE OR SINK
The origin and /or destination of data.
1) Data cannot move direly from a source to sink it must be moved by a process
2) A source and /or sink has a noun phrase land

DATA FLOW

45
1) A Data Flow has only one direction of flow between symbols. It may flow in both directions
between a process and a data store to show a read before an update. The later is usually
indicated however by two separate arrows since these happen at different type.
2) A join in DFD means that exactly the same data comes from any of two or more different
processes data store or sink to a common location.
3) A data flow cannot go directly back to the same process it leads. There must be atleast one
other process that handles the data flow produce some other data flow returns the original
data into the beginning process.
4) A Data flow to a data store means update (delete or change).
5) A data Flow from a data store means retrieve or use.

A data flow has a noun phrase label more than one data flow noun phrase can appear on a
single arrow as long as all of the flows on the same arrow move together as one package.

46
6. SYSTEM IMPLEMENTATION

47
6.1 PROJECT MODULES:

IMAGE ACQUISITION:
Different leaf images are collected and stored in a database. Images are classified into
training set and testing set. Training and testing sets were separated according to the images
which are needed to be trained and which are needed to be tested.
IMAGE ENHANCEMENT:
It is used to increase contrast of the images present in the training and test sets. This
enhancement is used to obtain dimensions of the image.
IMAGE SEGMENTATION:
Input images are first changed over into greyscale, acquiring aspects and eliminating
clamor. Then, applying K means calculation. Figure 2 portrays an illustration of k means
applied to the dataset giving the expectation for various number of variety color clusters
utilized.
FEATURE EXTRACTION:
We calculated the average value of each image pixel by calculating their separate R G B
values and finding the mean for the entire image containing various pixels.
IMAGE CLASSIFICATION:
The SVM classifier is used to identify the classes, which are closely connected to the known
and trained classes. The Support vector machine makes the ideal separating hyper plane
between the classes utilizing the preparation information.

IMAGE ACCURACY:
Calculate the accuracy of each image of the dataset by varying the number of clusters and
forming a plot showing the contrast between the accuracy of results obtained by each cluster
value.

48
6.2 METHODOLOGY(ALGORITHMS)

ALGORITHMS:

K-MEANS CLUSTERING ALGORITHM:

K-Means clustering is an unaided learning calculation that is utilized to tackle the grouping
issues in AI or information science. In this theme, we will realize what is K-implies
clustering algorithm, how the calculation works, alongside the Python execution of k-implies
grouping. K-means algorithm K-Means clustering is a Solo Learning calculation, which
bunches the unlabeled dataset into various groups. Here K characterizes the quantity of pre-
characterized bunches that should be made all the while, as though K=2, there will be two
groups, and for K=3, there will be three groups, etc. It is an iterative calculation that separates
the unlabeled dataset into k various bunches so that each dataset has a place just one
gathering that has comparable properties. It permits us to bunch the information into various
gatherings and a helpful method to find the classifications of gatherings in the unlabeled
dataset all alone without the requirement for any preparation. It is a centroid-based
calculation, where each bunch is related with a centroid The worth of k ought to be
foreordained in this calculation.
The k-implies bunching calculation basically performs two assignments:
1. Determines the best incentive for K focus focuses or centroids by an iterative interaction. 2.
Assigns every information highlight its nearest k-focus. Those information focuses which are
close to the specific k-focus, make a bunch.
The underneath outline clarifies the working of the K-implies Grouping Calculation:

49
 Fig. 6(a) Before and After k-means

How does the k-means calculation work:

The working of the K-Means calculation is clarified in the beneath steps:
Step-1: Select the number K to choose the quantity of bunches.
Step-2: Select irregular K focuses or centroids. (It very well may be other from the
information dataset).
Step-3: Allocate every information highlight their nearest centroid, which will frame the
predefined K bunches.
Step-4: Compute the fluctuation and spot another centroid of each group.
Step-5: Rehash the third steps, which means reassign each datapoint to the new nearest
centroid of each bunch.
Step-6: In the event that any reassignment happens, go to step-4 else go to Wrap up.
Step-7: The model is prepared.
We should comprehend the above strides by thinking about the visual plots:
Assume we have two factors M1 and M2. The x-y hub disperse plot of these two factors is
given underneath:

Fig. 6(b) x-y axis scatter plot

Let’s take number k of groups, i.e., K=2, to recognize the dataset and to place them into
various bunches. It implies here we will attempt to bunch these datasets into two unique
groups.

RANDOM FOREST ALGORITHM:

Random Forest is a popular machine learning algorithm that belongs to the supervised
learning technique. It can be used for both Classification and Regression problems in ML. It

50
is based on the concept of ensemble learning, which is a process of combining multiple
classifiers to solve a complex problem and to improve the performance of the model.
Random Forest is a classifier that contains a number of decision trees on various subsets of
the given dataset and takes the average to improve the predictive accuracy of that
dataset."Instead of relying on one decision tree, the random forest takes the prediction from
each tree and based on the majority votes of predictions, and it predicts the final output.
The greater number of trees in the forest leads to higher accuracy and prevents the problem
of overfitting.
The below diagram explains the working of the Random Forest algorithm:

Fig. 6(c) Majority Prediction

Assumptions for Random Forest

Since the random forest combines multiple trees to predict the class of the dataset, it is
possible that some decision trees may predict the correct output, while others may not. But
together, all the trees predict the correct output. Therefore, below are two assumptions for a
better Random forest classifier:
 There should be some actual values in the feature variable of the dataset so that the
classifier can predict accurate results rather than a guessed result.
 The predictions from each tree must have very low correlations.
Why use Random Forest?
Below are some points that explain why we should use the Random Forest algorithm:
 It takes less training time as compared to other algorithms.
 It predicts output with high accuracy, even for the large dataset it runs efficiently.
 It can also maintain accuracy when a large proportion of data is missing.
How does Random Forest algorithm work?

51
 Random Forest works in two-phase first is to create the random forest by combining N
decision tree, and second is to make predictions for each tree created in the first phase.
The Working process can be explained in the below steps and diagram:
Step-1: Select random K data points from the training set.
Step-2: Build the decision trees associated with the selected data points (Subsets).
Step-3: Choose the number N for decision trees that you want to build.
Step-4: Repeat Step 1 & 2.
Step-5: For new data points, find the predictions of each decision tree, and assign the new
data points to the category that wins the majority votes.

Implementation Steps are given below:

 Data Pre-processing step
 Fitting the Random forest algorithm to the Training set
 Predicting the test result
 Test accuracy of the result (Creation of Confusion matrix)
 Visualizing the test set result.

SUPPORT VECTOR MACHINE(SVM) ALGORITHM:

Support Vector Machine or SVM is one of the most popular Supervised Learning algorithms,
which is used for Classification as well as Regression problems.
The goal of the SVM algorithm is to create the best line or decision boundary that can
segregate n-dimensional space into classes so that we can easily put the new data point in the
correct category in the future. This best decision boundary is called a hyperplane.
SVM chooses the extreme points/vectors that help in creating the hyperplane. These
extreme cases are called as support vectors, and hence algorithm is termed as Support Vector
Machine. Consider the below diagram in which there are two different categories that are
classified using a decision boundary or hyperplane:

52
 Fig. 6(d) SVM

Types of SVM
SVM can be of two types:
 Linear SVM: Linear SVM is used for linearly separable data, which means if a
dataset can be classified into two classes by using a single straight line, then such data
is termed as linearly separable data, and classifier is used called as Linear SVM
classifier.
 Non-linear SVM: Non-Linear SVM is used for non-linearly separated data, which
means if a dataset cannot be classified by using a straight line, then such data is
termed as non-linear data and classifier used is called as Non-linear SVM classifier
How does SVM works?
Linear SVM:

The working of the SVM algorithm can be understood by using an example. Suppose we
have a dataset that has two tags (green and blue), and the dataset has two features x1 and x2.
We want a classifier that can classify the pair(x1, x2) of coordinates in either green or blue.
Consider the below image:

53
 Fig. 6(e) Coordinates
So as it is 2-d space so by just using a straight line, we can easily separate these two classes.
But there can be multiple lines that can separate these classes. Consider the below image:

Fig. 6(f) Hyperplane

Hence, the SVM algorithm helps to find the best line or decision boundary; this best
boundary or region is called as a hyperplane. SVM algorithm finds the closest point of the
lines from both the classes. These points are called support vectors. The distance between the
vectors and the hyperplane is called as margin. And the goal of SVM is to maximize this
margin. The hyperplane with maximum margin is called the optimal hyperplane.

Fig. 6(g) Optimal hyperplane

LOGISTIC REGRESSION ALGORITHM:

 Logistic regression is one of the most popular Machine Learning algorithms, which
comes under the Supervised Learning technique. It is used for predicting the
categorical dependent variable using a given set of independent variables.
 Logistic regression predicts the output of a categorical dependent variable. Therefore
the outcome must be a categorical or discrete value. It can be either Yes or No, 0 or 1,
true or False, etc. but instead of giving the exact value as 0 and 1, it gives the
probabilistic values which lie between 0 and 1.

54
  Logistic Regression is much similar to the Linear Regression except that how they are
used. Linear Regression is used for solving Regression problems, whereas Logistic
regression is used for solving the classification problems.
 In Logistic regression, instead of fitting a regression line, we fit an "S" shaped logistic
function, which predicts two maximum values (0 or 1).
The below image is showing the logistic function:

Logistic Function (Sigmoid Function):

 The sigmoid function is a mathematical function used to map the predicted values to
probabilities.
 It maps any real value into another value within a range of 0 and 1.
 The value of the logistic regression must be between 0 and 1, which cannot go beyond
this limit, so it forms a curve like the "S" form. The S-form curve is called the
Sigmoid function or the logistic function.
 In logistic regression, we use the concept of the threshold value, which defines the
probability of either 0 or 1. Such as values above the threshold value tends to 1, and a
value below the threshold values tends to 0.
Assumptions for Logistic Regression:
 The dependent variable must be categorical in nature.
 The independent variable should not have multi-collinearity.
Logistic Regression Equation:
The Logistic regression equation can be obtained from the Linear Regression equation. The
mathematical steps to get Logistic Regression equations are given below:
 We know the equation of the straight line can be written as:

 In Logistic Regression y can be between 0 and 1 only, so for this let's divide the
above equation by (1-y):

55
  But we need range between -[infinity] to +[infinity], then take logarithm of the
equation it will become:

The above equation is the final equation for Logistic Regression.

Steps in Logistic Regression:
To implement the Logistic Regression using Python, we will use the same steps as we have
done in previous topics of Regression. Below are the steps:
 Data Pre-processing step
 Fitting Logistic Regression to the Training set
 Predicting the test result
 Test accuracy of the result(Creation of Confusion matrix)
 Visualizing the test set result.

6.3 SOURCE CODE

SAMPLE CODING

#-----------------------------------

# GLOBAL FEATURE EXTRACTION

#-----------------------------------

from sklearn.preprocessing import LabelEncoder

from sklearn.preprocessing import MinMaxScaler

import numpy as np

import mahotas

import cv2

import os

import h5py

#--------------------

56
# tunable-parameters

#--------------------

images_per_class = 800

fixed_size = tuple((500, 500))

train_path = "dataset/train"

h5_train_data = 'output/train_data.h5'

h5_train_labels = 'output/train_labels.h5'

bins =8

# Converting each image to RGB from BGR format

def rgb_bgr(image):

rgb_img = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

return rgb_img

# Conversion to HSV image format from RGB

def bgr_hsv(rgb_img):

hsv_img = cv2.cvtColor(rgb_img, cv2.COLOR_RGB2HSV)

return hsv_img

# image segmentation

# for extraction of green and brown color

57
def img_segmentation(rgb_img,hsv_img):

lower_green = np.array([25,0,20])

upper_green = np.array([100,255,255])

healthy_mask = cv2.inRange(hsv_img, lower_green, upper_green)

result = cv2.bitwise_and(rgb_img,rgb_img, mask=healthy_mask)

lower_brown = np.array([10,0,10])

upper_brown = np.array([30,255,255])

disease_mask = cv2.inRange(hsv_img, lower_brown, upper_brown)

disease_result = cv2.bitwise_and(rgb_img, rgb_img, mask=disease_mask)

final_mask = healthy_mask + disease_mask

final_result = cv2.bitwise_and(rgb_img, rgb_img, mask=final_mask)

return final_result

# feature-descriptor-1: Hu Moments

def fd_hu_moments(image):

image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

feature = cv2.HuMoments(cv2.moments(image)).flatten()

return feature

# feature-descriptor-2: Haralick Texture

def fd_haralick(image):

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

haralick = mahotas.features.haralick(gray).mean(axis=0)

return haralick

58
# feature-descriptor-3: Color Histogram

def fd_histogram(image, mask=None):

image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)

hist = cv2.calcHist([image], [0, 1, 2], None, [bins, bins, bins], [0, 256, 0, 256, 0, 256])

cv2.normalize(hist, hist)

return hist.flatten()

# get the training labels

train_labels = os.listdir(train_path)

# sort the training labels

train_labels.sort()

print(train_labels)

# empty lists to hold feature vectors and labels

global_features = []

labels = []

# loop over the training data sub-folders

for training_name in train_labels:

# join the training data path and each species training folder

dir = os.path.join(train_path, training_name)

# get the current training label

current_label = training_name

59
# loop over the images in each sub-folder

for x in range(1,images_per_class+1):

# get the image file name

file = dir + "/" + str(x) + ".jpg"

# read the image and resize it to a fixed-size

image = cv2.imread(file)

image = cv2.resize(image, fixed_size)

# Running Function Bit By Bit

RGB_BGR = rgb_bgr(image)

BGR_HSV = bgr_hsv(RGB_BGR)

IMG_SEGMENT = img_segmentation(RGB_BGR,BGR_HSV)

# Call for Global Fetaure Descriptors

fv_hu_moments = fd_hu_moments(IMG_SEGMENT)

fv_haralick = fd_haralick(IMG_SEGMENT)

fv_histogram = fd_histogram(IMG_SEGMENT)

# Concatenate

global_feature = np.hstack([fv_histogram, fv_haralick, fv_hu_moments])

60
 # update the list of labels and feature vectors

labels.append(current_label)

global_features.append(global_feature)

print("[STATUS] processed folder: {}".format(current_label))

print("[STATUS] completed Global Feature Extraction...")

# get the overall feature vector size

print("[STATUS] feature vector size {}".format(np.array(global_features).shape))

# get the overall training label size

print("[STATUS] training Labels {}".format(np.array(labels).shape))

# encode the target labels

targetNames = np.unique(labels)

le = LabelEncoder()

target = le.fit_transform(labels)

print("[STATUS] training labels encoded...")

# scale features in the range (0-1)

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler(feature_range=(0, 1))

rescaled_features = scaler.fit_transform(global_features)

print("[STATUS] feature vector normalized...")

print("[STATUS] target labels: {}".format(target))

print("[STATUS] target labels shape: {}".format(target.shape))

# save the feature vector using HDF5

h5f_data = h5py.File(h5_train_data, 'w')

61
h5f_data.create_dataset('dataset_1', data=np.array(rescaled_features))

h5f_label = h5py.File(h5_train_labels, 'w')

h5f_label.create_dataset('dataset_1', data=np.array(target))

h5f_data.close()

h5f_label.close()

# training

#-----------------------------------

# TRAINING OUR MODEL

#-----------------------------------

import h5py

import numpy as np

import os

import glob

import cv2

import warnings

from matplotlib import pyplot

from sklearn.model_selection import train_test_split, cross_val_score

from sklearn.model_selection import KFold, StratifiedKFold

from sklearn.metrics import confusion_matrix, accuracy_score, classification_report

from sklearn.linear_model import LogisticRegression

from sklearn.tree import DecisionTreeClassifier

from sklearn.ensemble import RandomForestClassifier

from sklearn.neighbors import KNeighborsClassifier

from sklearn.discriminant_analysis import LinearDiscriminantAnalysis

from sklearn.naive_bayes import GaussianNB

62
from sklearn.svm import SVC

#from sklearn.externals import joblib

import joblib

warnings.filterwarnings('ignore')

#--------------------

# tunable-parameters

#--------------------

num_trees = 100

test_size = 0.20

seed =9

train_path = "dataset/train"

test_path = "dataset/test"

h5_train_data = 'output/train_data.h5'

h5_train_labels = 'output/train_labels.h5'

scoring = "accuracy"

# get the training labels

train_labels = os.listdir(train_path)

# sort the training labels

train_labels.sort()

if not os.path.exists(test_path):

os.makedirs(test_path)

63
# create all the machine learning models

models = []

models.append(('LR', LogisticRegression(random_state=seed)))

models.append(('LDA', LinearDiscriminantAnalysis()))

models.append(('KNN', KNeighborsClassifier()))

models.append(('CART', DecisionTreeClassifier(random_state=seed)))

models.append(('RF', RandomForestClassifier(n_estimators=num_trees, random_state=seed)))

models.append(('NB', GaussianNB()))

models.append(('SVM', SVC(random_state=seed)))

# variables to hold the results and names

results = []

names = []

# import the feature vector and trained labels

h5f_data = h5py.File(h5_train_data, 'r')

h5f_label = h5py.File(h5_train_labels, 'r')

global_features_string = h5f_data['dataset_1']

global_labels_string = h5f_label['dataset_1']

global_features = np.array(global_features_string)

global_labels = np.array(global_labels_string)

h5f_data.close()

h5f_label.close()

# verify the shape of the feature vector and labels

64
print("[STATUS] features shape: {}".format(global_features.shape))

print("[STATUS] labels shape: {}".format(global_labels.shape))

print("[STATUS] training started...")

# split the training and testing data

(trainDataGlobal, testDataGlobal, trainLabelsGlobal, testLabelsGlobal) =

train_test_split(np.array(global_features),

np.array(global_labels),

test_size=test_size,

random_state=seed)

print("[STATUS] splitted train and test data...")

print("Train data : {}".format(trainDataGlobal.shape))

print("Test data : {}".format(testDataGlobal.shape))

trainDataGlobal

# 10-fold cross validation

for name, model in models:

kfold = KFold(n_splits=10, random_state=seed, shuffle=True)

cv_results = cross_val_score(model, trainDataGlobal, trainLabelsGlobal, cv=kfold,

scoring=scoring)

results.append(cv_results)

names.append(name)

msg = "%s: %f (%f)" % (name, cv_results.mean(), cv_results.std())

print(msg)

# boxplot algorithm comparison

65
fig = pyplot.figure()

fig.suptitle('Machine Learning algorithm comparison')

ax = fig.add_subplot(111)

pyplot.boxplot(results)

ax.set_xticklabels(names)

pyplot.show()

clf = RandomForestClassifier(n_estimators=num_trees, random_state=seed)

import seaborn as sns

sns.heatmap(cm ,annot=True)

print(classification_report(testLabelsGlobal,y_predict))

from sklearn.metrics import accuracy_score

accuracy_score(testLabelsGlobal, y_predict)

import cv2

import numpy as np

import matplotlib.pyplot as plt

import numpy as np

import matplotlib.pyplot as plt

from scipy import ndimage

from sklearn.cluster import KMeans

from sklearn.preprocessing import MinMaxScaler

from matplotlib.colors import hsv_to_rgb

img = cv2.imread('./image.jpg')

plt.imshow(img)

plt.show()

img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

66
plt.imshow(img)

plt.show()

hsv_img = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)

plt.imshow(hsv_img)

lower_green = np.array([25,0,20])

upper_green = np.array([100,255,255])

mask = cv2.inRange(hsv_img, lower_green, upper_green)

result = cv2.bitwise_and(img, img, mask=mask)

plt.subplot(1, 2, 1)

plt.imshow(mask, cmap="gray")

plt.subplot(1, 2, 2)

plt.imshow(result)

plt.show()

lower_brown = np.array([10,0,10])

upper_brown = np.array([30,255,255])

disease_mask = cv2.inRange(hsv_img, lower_brown, upper_brown)

disease_result = cv2.bitwise_and(img, img, mask=disease_mask)

plt.subplot(1, 2, 1)

plt.imshow(disease_mask, cmap="gray")

plt.subplot(1, 2, 2)

plt.imshow(disease_result)

plt.show()

final_mask = mask + disease_mask

final_result = cv2.bitwise_and(img, img, mask=final_mask)

plt.figure(figsize=(15,15))

67
plt.subplot(1, 2, 1)

plt.imshow(final_mask, cmap="gray")

plt.subplot(1, 2, 2)

plt.imshow(final_result)

plt.show()

sift = cv2.SIFT_create()

kp = sift.detect(img,None)

img=cv2.drawKeypoints(img,kp,img)

plt.imshow(img)

68
7. SYSTEM TESTING

69
7.1 TESTING METHODS:

Framework tests are intended to approve a completely created framework with the end
goal of expecting that it meets its prerequisites. There are three sorts of framework testing
they are
Alpha Testing:
Alpha testing alludes to the framework testing that is conveyed by the client inside the
association alongside the engineer. The alpha tests are directed in controlled way.
Beta Testing:
Beta testing is the framework performed by a chose gathering of clients, the designer is
absent at the site and the client will educate the issues that are experienced during testing.
The product designer makes the essential changes and submits to the client.
Acceptance Testing:
Acceptance testing is the framework testing performed by the client to whether to
acknowledge the conveyance of the framework.
Unit Testing:
Unit testing centers check exertion around the littlest unit of the product structure the module.
Utilizing the point by point structure depiction as a guide, significant control ways are tried to
reveal blunders with the limit of the module for the accompanying modules. All the
announcements in the module are executed atleast once. From this we can guarantee that
every autonomous way through the control structures are worked out.
Integration Testing:
Incorporation testing is a precise procedure for building the program structure and to lead
tests for revealed blunders with interfacing. In this framework, Top-Down combination is
performed for the development of program structures.
Validation Testing:

70
This testing is performed to guarantee that the framework capacities in a way that can be
sensibly by the clients. Here information is approved first at interface(authentication), before
sending it to the server.

7.2 TEST CASES:

Fig. 7(a) Module not found error

71
An error occurs when a module is not found as shown in the above figure an error occured
because “sklearn” module is not found.

Fig. 7(b) Image path error

An error occurs when the path of the image is not defined correctly as shown in the figure.

72
8. EXPERIMENTAL RESULTS

73
Machine Learning algoritm comparison:

Fig. 8(a) Machine Learning Algorithm comparison

By using machine learning we calculate all the linear regression,Linear Discriminant

Analysis, k-nearest neighbour, classification and regression trees for machine learning, radio
frequency, navie bayes and support vector machine.
Each and every algoarithm has given some value. After comparisiom of linear regression,
linear discriminant analysis, n nearest neighbour, classification and regression trees for
machine learning, radio frequency, navie baised and support vector machine it is given some
value after processing the machine learning algoarithm comparision.

Details of Diseased Leaf:

Fig. 8(b) Details of diseased leaf

This screen is to details of the diseased plant. It will set the details where the diseases leaf
place it will arranging the leaf with calculation model.

74
Selected Image:

Fig. 8(c) Selected Image

In project we can select one image from the data sets. In dataset we have the all leafs in that,
we can select and add it one in the project.

Diseased or Healthy:

Fig. 8(d) Diseased or Healthy

After we can select the image it will checking whether the leaf is diseased or healthy.

75
Converted image into hsv image:

Fig. 8(e) HSV image

After checking the leaf is diseased or not then it will be checking HUE SATURATION
VALUE(HSV).

After converting hsv image then it will convert into gray scale:

Fig. 8(f) Converting hsv into gray scale

76
Gray Scale Image:

Fig. 8(g) Gray Scale Image

Diseased image spots:

Fig. 8(h) Diseased image spots

77
Fig. 8(i) Diseased image

78
9. CONCLUSION & FUTURE SCOPE

79
 Our research innovated a computer aided segmentation and classification method.
Main focus of our project is to recognize diseases on the pomegranate leaf. At first
preprocessing is done. Second stage occurs with K-Means algorithm being applied on all the
images contained in the dataset. Third stage involves feature extraction that includes color
features and shape features. This is followed by classification of diseases with the help of rich
trained data repositories. The statistical parameters are used as features for classification. The
work can be used to identify the condition of the pomegranate leaf and for the identification
of the diseased leaf or healthy leaf of the pomegranate plant. This project can be broadly
expanded. The future scope of this project includes extending it to various leaves and their
respective diseases. Also, this research is useful in finding out weather the accuracy of the
result is sufficient enough to rely on the prediction made by the automated analysis of leaf
disease. Special application can be built for easy access in which the researcher may upload a
real time image of the diseased leaf and it may get the name and cure of the disease. Many
more improved classification algorithms can be applied to predict the leaf diseases.

80
10. BIBILIOGRAPHY

81
[1] Nikita Goel, Dhruv Jain and Adwitiya Sinha, “Prediction Model for Automated Leaf
Disease Detection & Analysis”,IEEE 2018.

[2] Shivaputra S.Panchal, Rutuja Sonar, “Pomegranate Leaf Disease Detection Using Support
Vector Machine”,International Journal Of Engineering And Computer Science ISSN: 2319-
7242 Volume 5 Issues 6 June 2016, Page No. 16815-16818.

[3] J Shirahatti, R Patil and P Akulwar, “A survey paper on plant disease identification using
machine learning approach”,IEEE 2018.

[4] Jayashri Patil, Sachin Naik , “Pomegranate fruit diseases detection using image
processing techniques:”A Review",IT in Industry, Vol. 9, No.2, 2021.

[5] SD Khirade and AB Patil, “Plant disease detection using image processing”,IEEE 2015.

[6] Pujari, J.D., Yakkundimath, R., Byadgi, A.S., “Image Processing Based Detection of
Fungal Diseases In Plants”, International Conference on Information and Communication
Technologies, Volume 46, pp. 1802-1808, 2015.

[7] K. Sakthidasan Sankaran,N. Vasudevan and V. Nagarajan, “Plant Disease Detection and
Recognition using K means Clustering”,IEEE 2020.

[8] G. Shrestha, Deepsikha, M. Das and N. Dey, “Plant Disease Detection Using CNN”, 2020
IEEE Applied Signal Processing Conference (ASPCON), 2020.

[9] Bijaya Kumar Hatuwal, Aman Shakya, and Basanta Joshi, “Plant Leaf Disease
Recognition Using Random Forest, KNN, SVM and CNN”,Polibits 2020.

[10] Bhange, M., Hingoliwala, H.A., “Smart Farming Disease Detection Using Image
Processing”, Second International Symposium on Computer Vision and the Internet, Volume
58, pp. 280-288, 2015.

82
[11] U.Shruthi, V.Nagaveni; B.K. Raghavendra, “A Review on Machine Learning
Classification Techniques for Plant Disease Detection” ,International Conference on
Advanced Computing & Communication Systems(ICACCS), 2019.

[12] L.Sherly Pushpa Annabel, T. Annapoorani, P. Deepalakshmi, “Machine Learning for

Plant Leaf Disease Detection and Classification – A Review”, International Conference on
Communication and Signal Processing(ICCSP), Chennai.

[13] Jay Trivedi, Yash Shamnai, Ruchi Gajjar, “ Plant Leaf Disease Detection Using Machine
Learning”, International Conference on Emerging Technology Trends in Electronics
Communication and Networking, 2020.

[14] Adhao Asmita Sarangdhar, V.R. Pawar, “Machine learning regression technique for
cotton leaf disease detection and controlling using IoT”, International Conference of
Electronics, Communication and Aerospace Technology(ICECA), 2017.

[15] Amrita S. Tulshan; Nataasha Raul, “ Plant Leaf Disease Detection using Machine
Learning”, International Conference on Computing, Communication and Networking
Technologies(ICCNT), 2019.

83