SYNOPSIS ON

An Improved Convolutional Neural Network Model for

Detection of Arecanut Disease and Remedies

ABSTRACT
Arecanut, commonly referred to as betel nut, is a tropical crop. India holds the second position
globally both in its production and consumption. Throughout its lifecycle, the crop is susceptible
to various ailments, spanning from root to fruit. Presently, disease detection primarily relies on
visual inspection, requiring farmers to meticulously examine each crop at regular intervals. In
this study, we have introduced a system utilizing Convolutional Neural Networks to identify
diseases affecting the leaves, trunk, and overall health of the arecanut plant. Additionally, the
system offers recommended remedies for these identified diseases. An improved Convolutional
Neural Network (CNN) is a type of Deep Learning algorithm designed for image analysis. In our
study, we developed an improved CNN model to distinguish between healthy and diseased
arecanut plants. To facilitate training and testing, we curated a dataset comprising more than 700
images of both healthy and diseased arecanut plants, split into an 80:20 ratio for training and
testing, respectively. During model compilation, categorical cross-entropy was utilized as the
loss function, with the Adam optimizer and accuracy as the performance metric. Training
involved 50 epochs to achieve optimal validation and test accuracy while minimizing loss. Our
approach yielded promising results, accurately identifying arecanut diseases.

INTRODUCTION
Handling and meeting heightened pressures in agriculture from the environmental, economic,
marketing and public sectors makes precision agriculture an effective concept for crop
management. Because of these stresses, farmer productivity and public benefits including
ecological and environmental impacts are highly expected. To fulfill these standards the
precision requirement agricultural production is highly recommended. Agricultural automation
can take advantage of machine vision capabilities, which can be used in various tasks such as
inspection, grading, projected output, automated autonomous machine selection and guidance.
Areca nut or Betel nut (Areca catechu L.,) is a crucial Indian commercial crop grown in various
regions such as the East, the North East, the East and the West Ghats. India ranks first among
other world countries according to both crop area and total production (47 percent). Indian areca
nut production is in top rank with total world production (1.27 tones / ha). The grading of areca
nut is an important stage in price fixing. This is helping farmers get the right price for their
harvest. The research on the Areca classification is found to be divisive. An automated method
for classifying Betel nut/Areca nut is proposed under. The classification is achieved using the
Areca nut’s color, shape and texture as characteristics. A method for classifying Areca nut in two
classes was proposed based on color. There are three steps in the method: segmentation,
masking, and classification. The segmented color elements Red and Green Areca nut area is used
for ranking.

The current approach for detecting plant diseases is simply observation with the naked eye and
farmers have to carefully analyze each and every crop periodically to detect diseases, which is an
extremely challenging and time-consuming task and which requires more manpower, adequately
equipped laboratories and costly instruments and it’s not possible for early detection of the
diseases and avoid spreading of disease. Hence, there is a need for an automatic disease detection
system.

India is the highest producer of arecanut with a production of around 3.3lakh tones and a total
acreage under cultivation of 2.64lakh hectares, with Karnataka and Kerala accounting for nearly
72 per cent of the total production. Commonly found diseases in areca tree are Mahali Disease
(Koleroga), Bud Rot Disease, Stem Bleeding, Yellow Leaf spot, Yellow Disease which occurs
due to continuous rainfall and climatic changes, these diseases must be controlled in the primary
stage of infection otherwise it may cause difficult to control in the final stage which may lead to
loss to the former. To avoid this, we can use Machine Learning to detect disease and suggest
remedies to it. Arecanut disease detection can be done by looking for a spot on the affected nut,
leaves, and trunk. Machine Learning specifically Convolution Neural Networks is the method
we're using to detect the diseases. In this project we are going to identify Mahali disease (Kole
roga), Stem Bleeding and yellow leaf spot diseases and suggest solutions for the detected
disease.
 LITERATURE SURVEY
Dhanuja K C (2020) Proposed a system for disease detection of arecanut using image
processing technology and here the author followed texture-based grading of arecanut. For
training and testing the model a total 144 arecanut samples used which includes 49 Good, 46
Poor and 49 Negative samples [1] and K-Nearest Neighbor (KNN) algorithm is used to detect
the diseases in arecanut.
Ajit Danti, Suresh (2012) [2] Suggested a technique for segmentation and classification of raw
arecanut. In this paper the novel method is proposed for classification of arecanut into two
classes based on color which includes three steps: (i) Segmentation, (ii) Masking and (iii)
Classification. And here in this paper classification is done based on red and green color
components of the segmented region of arecanut.
Manpreet Sandhu, Pratik Hadawale [3] In this paper the author has developed an automated
system for detecting disease using leaf image classification. The presence of spots or rotting
areas in the plant leaf will be detected automatically by using machine learning algorithms. Here
unmanned aerial vehicle (UAV) with a camera that automatically captures images of the leaves
to automate the process of capturing images of the leaves.
Ashish Nage, V.R. Raut [4] This paper focuses on the approach based on image processing for
detection of diseases of plants. Here is this paper the author developed an Android application
that helps farmers in identifying plant disease by uploading a leaf image to the system which
uses a convolution neural network algorithm to identify the disease in the leaf.
Anandhakrishnan MG, Joel Hanson, Annette Joy [5] In this, a dataset is created for training a
neural network for image classification. The dataset was collected manually from the field. In
this paper to get better feature extraction, image reprocessing was done, which includes noise
removal, intensity normalization, removing reflections, and masking portions of the image. Then,
using these processed images, a deep convolutional neural network model was trained to classify
the images and TensorFlow library was used for numerical computations.
Patil Supriya, Mule Gitanjali, Labade Vidya [6] In this paper detect pomegranate diseases and
also suggest the solution on diseases. The proposed system consists of image pre-processing,
segmentation, extraction of features and classification. In image pre-processing, images are
resized. In segmentation, color segmentation is carried out. Color, morphology and texture
features (Gabor filter) are used for the feature extraction. Minimum distance classifier is used for
classification purposes.
Swathy Ann Sam, Siya Elizabeth Varghese, Pooja Murali [7] In this paper the authors have
used different algorithms (SVM, KNN, Decision tree, CNN) for detection of diseases in leaves.
This project works on uploading a captured image of sample to the system and algorithm will
detect weather the sample is affected by any diseases or not, if it is affected by any disease, it
will print the detected disease, for this detection the authors have used CNN which gave an
accurate percentage of 86%.

AIM AND OBJECTIVES

Aim

The aim of the project is to detect the disease in arecanut fruit, Stem and leaves using improved
Convolutional Neural network and Image processing techniques and finding the remedies to the
diseases.

Objectives

 To collect datasets that contain healthy and diseased images of arecanut and their leaves.
 Design and develop an algorithm for early detection of disease in arecanut that can avoid
the spreading of diseases.
 Develop a deep learning based algorithm that would suggest solutions for the detected
diseases.
  Developing a web application for early detection of disease, prediction and providing
remedies to the above solution.

EXISTING SYSTEM
Researchers and innovators constantly explore new methods and technologies for disease
detection in plants, including arecanut, aiming to improve accuracy, efficiency, and ease of
implementation for farmers.

Visual Inspection: The traditional method involves manual examination by agricultural experts
or farmers to identify visual symptoms of diseases, such as leaf discoloration, lesions, spots, or
unusual patterns.

Laboratory Techniques: These methods involve taking samples from plants showing
symptoms, culturing pathogens in a lab, and using techniques like microscopy, PCR (Polymerase
Chain Reaction), ELISA (Enzyme-Linked Immunosorbent Assay), or DNA sequencing to
identify specific pathogens causing diseases.

Image Processing: Utilizing cameras and image processing algorithms to capture images of
plants and then employing computer vision techniques, machine learning, or deep learning to
analyze these images for disease symptoms. This method allows for automated and rapid
analysis of large-scale crops.

Machine Learning: Creating machine learning platforms that integrate various data sources
(images, environmental factors, and historical data) to diagnose plant diseases. These systems
continuously learn and improve their accuracy over time.

PROPOSED SYSTEM
The project's aim is to build a fully automated image classification system to positively identify
different diseases in areca nut. In this project, we first perform disease detection on multiple
areca nut, we used algorithms that provide good output at all detection stages. They have a
comparatively higher predictability and a lower false positive rate.
To detect various diseases of arecanut we created our dataset which consists of healthy and
diseased images of arecanut and their leaves. The database is preprocessed, which includes
image reshaping, resizing, and array conversion.

The test image is also subjected to similar processing. Images are resized to 256*256 resolutions
and converted to an array before training the improved CNN model. A Convolutional Neural
Network (ConvNet/CNN) that can take in an image as input, assign importance (learnable
weights and biases) to various aspects/objects in the image then learn from the results and be
able to differentiate one from the other. As compared to other classification algorithms, the
amount of pre-processing needed by a ConvNet is significantly less. Though filters in primitive
methods are hand-engineered, ConvNets can learn these filters/characteristics with enough
preparation. Figure 3. shows the working design of our model.

SYSTEM REQUIREMENTS AND SPECIFICATIONS

The System Requirements Specification document describes all data, functional and
behavioural requirements of the software under production or development. A functional
requirement document defines the functionality of a system or one of its subsystems. It also
depends upon the type of software, expected users and the type of system where the software
is used. On-functional requirement is a requirement that specifies criteria that can be used to
judge the operation of a system, rather than specific behaviours.

Hardware Requirements

The selection of hardware is very important in the existence and proper working of any
software. When selecting the hardware, the size and requirements are also important to run
the software. The minimum hardware requirements are as follows:

 Processor: Intel CORE i3

 RAM: 4 GB

 Disk space: minimum 256 GB

Software Requirements

 Operating System (Windows, MacOS).

 Python, HTML, CSS, JavaScript.

 Java / Python Framework and SQL database

 An Internet Browser (Google Chrome, Microsoft Edge etc).

 Code Editor (Visual Studio code/PyCharm).

 The package manager PIP (pip is a python package-management system written in Python
used to install and manage software packages).

REFERENCES

1. Dhanuja K C. Areca Nut Disease Detection using Image Processing Technology.

International Journal of Engineering Research 2020 V9. 10.17577/IJERTV9IS080352.
2. Mallaiah, Suresha & Danti, Ajit & Narasimhamurthy, S. Classification of Diseased
Arecanut based on Texture Features. International Journal of Computer Applications.
2014.
3. Manpreet Sandhu, Pratik Hadawale, Saumaan Momin, Prof. Ajitkumar Khachane. Plant
Disease Detection using ML and UAV. International Research Journal of Engineering
and Technology 2020 V7.
4. Mr. Ashish Nage, Prof. V. R. Raut, Detection and Identification of Plant Leaf Diseases
based on Python, INTERNATIONAL JOURNAL OF ENGINEERING RESEARCH &
TECHNOLOGY (IJERT),2019, Volume 08, Issue 05.
5. Anandhakrishnan MG Joel Hanson, Annette Joy, Jeri Francis, Plant Leaf Disease
Detection using Deep Learning and Convolutional Neural Network, International Journal
of Engineering Science and Computing, Volume 7, Issue No.3, March 2017
6. Manisha Bhange, H.A. Hingoliwala, Smart Farming: Pomegranate Disease Detection
Using Image Processing, Procedia Computer Science, Volume 58,2015, Pages 280-288,
ISSN 1877-0509.
7. Swathy Ann Sam, Siya Elizebeth Varghese, Pooja Murali, Sonu Joseph John, Dr. Anju
Pratap. Time saving malady expert system in plant leaf using CNN, 2020, Volume 13,
Issue No 3.
8. Detection and classification of areca nuts with machine vision Kuo-Yi Huang 2012
9. Classification of Diseased Areca nut based on Texture Features International Journal of
Computer Applications (0975 – 8887) Recent Advances in Information Technology,
2014
10. Segmentation and Classification of Raw Arecanuts Based on Three Sigma Control Limits
December 2012.
11. C.L. Hsieh, S.F. Cheng, T.T. Lin, Application of image texture analysis and neural
network on the growth stage recognition for head cabbage seedlings, Journal of
Agricultural Machinery 6 (2) (1997) 1–13 (in Chinese).