Professional Documents
Culture Documents
Potato Leaf Disease Detection.-Test1
Potato Leaf Disease Detection.-Test1
By TYITB107 Mansi Patil, TYITB112 Abhishek Pawar and TYITB145 Harshal Bhimartwar
II Literature Review :
In a research paper with the name called
“Classification of diseased potato leaves
using machine learning” the authors used algorithms which are AlexNet, VggNet,
integrated image processing and machine ResNet, LeNet and a sequential model.
learning. They achieved an accuracy of 84%
which was particularly less as compared to In a research paper named as 'Potato Leaf
our work. Disease Detection using Deep learning' the
authors used GoogleNet, Resnet50, and
In a research paper with the name 'Potato VGG16 convolutional neural network
Disease Detection using Machine Learning' architecture models to create an accurate
the authors used automated systems to detect classification system.
the diseases. These systems consisted of
2034 pictures and used image processing [3]In a research paper titled ‘Early Detection
algorithms. of Potato Leaf Diseases using CNN with
Web Application’ the authors used deep
In a research paper titled 'CNN based learning model with a multi-layer for the
Disease Detection Approach on Potato detection of diseases in the potato leaves.
Leaves' the authors used CNN with 5
inexpensive way to detect diseases in these [4]In a research paper named ‘Classification
of Diseased Potato Leaves using Machine
Learning’ the authors used different ML
algorithms such as Support Vector Machine,
K Nearest Neighbor, Naive Bayes and
Decision Tree.
III Methodology :
Late blight
IV.I DATASET :
The dataset we have used in this study
consists of 360 samples .The dataset is
divided into three class those are healthy
leaves, early blight and late blight each of Table no 1
these consists of 120 sample each .This Samples Number
dataset is collected from kaggle under the
name “PLANT VILLAGE”. Healthy 120
Total 360
80% 20%
IV.IV CLASSIFICATION
After these steps, further we have to classify
images using Convolutional Neural
Network(CNN) architecture.A convolution
neural network or CNN is a type or artificial
neural network, which is used for image
recognition and classification in a wide
range. Deep learning identifies objects in an
image using a CNN. After applying the CNN
model to the dataset the system is able to
recognize the type of leaves with the Each layer in the design of a CNN model
accuracy of 84.00%. often serves a particular function in the
image processing pipeline. An summary of
each layer is provided below:
IV.V ARCHITURE OF CNN
1. Input layer: This layer takes in the
An artificial neural network called a CNN input images and sends it to the
(Convolutional Neural Network) model following layer for processing.
which is frequently employed at image Usually, the input layer's dimensions
processing and computer vision tasks. It are fixed at the same size as the input
utilizes a technique known as convolution, image
which involves applying a number of filters
to the image to find edges, textures, and 2. Convolutional layer: Convolution is
other visual patterns, to automatically the fundamental process of a CNN
identify and extract important characteristics model, and it is carried out by this
from images. layer. Convolution is the process of
extracting features from an input
Several layers, including convolutional, image by swiping a group of filters
pooling, and fully linked layers, make up (also known as kernels). Each filter
architecture of a standard CNN model. creates a feature map that draws
Pooling layers shrink the size of the feature attention to a certain pattern or form,
like edges, corners, or blobs. A recognize which type of leaf it is either healthy
convolutional layer's output is a ,early blight or late blight with the accuracy of
collection of feature maps that indicate 84%. With this system the farmer will get to
the existence of different visual know which type of disease the leaf has and
features in the input image can find the solution on this which will help to
reduce the loss of crop and the financial loss of
3. Pooling layer: The convolution layer's the farmer also.
output feature maps are downsampled
at the pooling layer to minimize their
size and improve model performance.
Max pooling, which chooses the
highest value in each local region of the
feature map, is the most popular
pooling operation.
The necessity for extensive and diverse [2] P. Patil, N. Yaligar and S. M. Meena,
datasets, handling changes in lighting and "Comparison of Performance of Classifiers -
background, and the requirement for SVM, RF and ANN in Potato Blight Disease
real-time detection in the field are all issues Detection Using Leaf Images," 2017 IEEE
that still need to be resolved. International Conference on Computational
Notwithstanding these difficulties, there is Intelligence and Computing Research (ICCIC),
significant potential for increasing crop Coimbatore, India, 2017, pp. 1-5, doi:
10.1109/ICCIC.2017.8524301
output by using automated approaches for
detecting potato leaf disease.
[3]P. K. Shukla and S. Sathiya, "Early
and minimising how plant diseases affect the
Detection of Potato Leaf Diseases using
availability of food.
Convolutional Neural Network with Web
Application," 2022 IEEE World Conference
on Applied Intelligence and Computing
In conclusion, the automated diagnosis of
(AIC), Sonbhadra, India, 2022, pp. 277-282,
potato leaf disease is a rapidly developing
doi: 10.1109/AIC55036.2022.9848975.
topic that has enormous potential to help
[4] S. Sharma, V. Anand and S. Singh,
database, we can get better accuracy. In the
"Classification of Diseased Potato Leaves
future our target is to create an application to
Using Machine Learning," 2021 10th IEEE
International Conference on Communication 1312.
Systems and Network Technologies (CSNT),
Bhopal, India, 2021, pp. 554-559, doi: [9]Mahum, Rabbia & Munir, Haris &
10.1109/CSNT51715.2021.9509702. Mughal, Zaib-Un-Nisa & Awais, Muhammad
& Khan, Falak & Saqlain, Muhammad &
Mahamad, Saipunidzam & Tlili, Iskander.
[5]M. K. R. Asif, M. A. Rahman and M. H.
(2022). A novel framework for potato leaf
Hena, "CNN based Disease Detection
disease detection using an efficient deep
Approach on Potato Leaves," 2020 3rd learning model. Human and Ecological Risk
International Conference on Intelligent Assessment: An International Journal. 1-24.
Sustainable Systems (ICISS), Thoothukudi, 10.1080/10807039.2022.2064814.
India, 2020, pp. 428-432, doi:
10.1109/ICISS49785.2020.9316021. [10 ]B. Vadivel, R. Thangaraj, P. Pandiyan,
T. Aravind, K. Harish and E. Sivaraman,
"Deep Learning Models for Potato Leaf
[6]S. Bangari, P. Rachana, N. Gupta, P. S. Disease Identification: A Comparative
Sudi and K. K. Baniya, "A Survey on Analysis," 2022 International Conference on
Disease Detection of a potato Leaf Using Green Energy, Computing and Sustainable
CNN," 2022 Second International Technology (GECOST), Miri Sarawak,
Conference on Artificial Intelligence and Malaysia, 2022, pp. 58-62, doi:
Smart Energy (ICAIS), Coimbatore, India, 10.1109/GECOST55694.2022.10010611.
2022, pp. 144-149, doi:
10.1109/ICAIS53314.2022.9742963. [11] M. A. Iqbal and K. H. Talukder,
"Detection of Potato Disease Using Image
[7]S. Omar, R. Jain and V. Bali, "Leaf Segmentation and Machine Learning," 2020
Disease Detection using Convolutional International Conference on Wireless
Neural Network," 2022 International Communications Signal Processing and
Conference on Machine Learning, Big Data, Networking (WiSPNET), Chennai, India,
Cloud and Parallel Computing 2020, pp. 43-47, doi:
(COM-IT-CON), Faridabad, India, 2022, pp. 10.1109/WiSPNET48689.2020.9198563.
53-56, doi:10.1109/CCECE.2017.7946594.
[12] M. Islam, Anh Dinh, K. Wahid and P.
[8]T. -Y. Lee, J. -Y. Yu, Y. -C. Chang and J. Bhowmik, "Detection of potato diseases
-M. Yang, "Health Detection for Potato Leaf using image segmentation and multiclass
with Convolutional Neural Network," 2020 support vector machine," 2017 IEEE 30th
Indo – Taiwan 2nd International Conference Canadian Conference on Electrical and
on Computing, Analytics and Networks Computer Engineering (CCECE), Windsor,
(Indo-Taiwan ICAN), Rajpura, India, 2020, ON, Canada, 2017, pp. 1-4, doi:
pp. 289-293, doi: 10.1109/CCECE.2017.7946594.
10.1109/Indo-TaiwanICAN48429.2020.918
By TYITB107 Mansi Patil, TYITB112 Abhishek
Pawar and TYITB145 Harshal Bhimartwar