DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING
18EEP301L-MINOR PROJECT-III
FINAL REVIEW
AGRICULTURE LEAF DISEASE DETECTION
YEAR/SEMESTER – III/V
BATCH NUMBER : 05
DATE : 29/11/2024 PRESENTED BY:
Aswinth M (927622BEE008)
GUIDED BY: Dharineesh R.S (927622BEE021)
Mr. Jasper M.E., Ganesh Boopathi A (927622BEE032)
AP/EEE Gowtham M (927622BEE035)
LIST OF CONTENTS
Abstract
Objective
Proposed System
Block Diagram
Components Used
Components Description
Working Explanation
Future Enhancements
Cost Estimation
References
ABSTRACT
This project leverages the VGG19 neural network to detect and classify
plant leaf diseases with 95% accuracy, surpassing traditional methods in
efficiency and precision. By analyzing images of diseased leaves, it enables
early detection and timely intervention to mitigate crop losses. This innovative
deep learning approach enhances agricultural productivity and demonstrates AI's
transformative potential in modern farming.
OBJECTIVE
To develop a deep learning-based system using the VGG19
neural network model for accurate and efficient detection of plant leaf
diseases. The goal is to enable early diagnosis and timely intervention
by analyzing images of diseased leaves. This system aims to minimize
disease management in farming.
PROPOSED SYSTEM
The proposed system utilizes the VGG19 neural network to detect and
classify plant leaf diseases with high accuracy. It includes image acquisition and
preprocessing, enabling efficient analysis of leaf images. A user-friendly interface
or mobile app allows users to upload images and receive instant diagnosis and
tailored treatment recommendations. The system supports real-time monitoring
and alerts for early detection across farms and integrates cloud-based processing
for scalability. Additionally, multilingual support ensures accessibility for diverse
agricultural communities.
BLOCK DIAGRAM
COMPONENTS USED
COMPONENTS
HARDWARE SOFTWARE
• Camera or Smartphone • Python (Programming)
• Computational • VGG 19
Hardware
• Web Application Frameworks
COMPONENTS DESCRIPTION
CAMERA OR SMARTPHONE
• Used for capturing high-resolution images of plant
leaves.
• Smartphones with camera capabilities or dedicated
cameras for image capture.
COMPUTATIONAL HARDWARE
• CPU (Central Processing Unit): For basic image
processing tasks.
• GPU (Graphics Processing Unit): For faster training and
inference of deep learning models (particularly for large
datasets or real-time detection).
• RAM (Random Access Memory): To support data
processing during model training or prediction.
PROGRAMMING LANGUAGE :
• Python: Primary language for implementing the deep
learning models, data processing, and integration of the
system. Libraries like TensorFlow, Keras, OpenCV, and
NumPy are all Python-based.
VGG 19
• VGG19 is software, specifically a deep learning model used for image
classification and feature extraction tasks. It is a type of convolutional neural
network (CNN) that consists of 19 layers (16 convolutional layers and 3 fully
connected layers). Developed by the Visual Geometry Group (VGG) at Oxford
University, the model is implemented in popular machine learning frameworks.
WORKING EXPLANATION
The process involves using the VGG19 model to detect plant leaf
diseases by training it on datasets like the PlantVillage Dataset and real-field
images. The data is split into training, validation, and testing sets, with data
augmentation applied to improve generalization. The model is trained by tuning
key parameters such as epochs, learning rate, and batch size. Once trained, its
performance is evaluated and applied for classification, detection, and analysis
of disease types. This system automates plant disease management, helping
protect crops from disease.
FUTURE ENHANCEMENT
Expansion of Disease Detection:
Expanding the model to detect a wider range of plant diseases, including newly
emerging ones, by incorporating additional datasets and continuously retraining
the model.
Weather and Environmental Data Integration:
Adding weather and environmental data to improve disease prediction accuracy,
as some diseases are heavily influenced by factors such as humidity, temperature,
and rainfall.
COST ESTIMATION
S.NO COMPONENTS QUANTITY COST
1 Microcontroller 1 300
2 Lockers Few 300
3 LCD Display 1 350
4 Transformer 1 200
5 Relay 1 200
5 Micellaneous Few 700
Total 2100
REFERENCES
• Mukhtar, T.; Rehman, S.U.; Smith, D.; Sultan, T.; Seleiman, M.F.; Alsadon, A.A.;
Amna; Ali, S.; Chaudhary, H.J.; Solieman, T.H.I.; et al. Mitigation of Heat Stress in
Solanum lycopersicum L. by ACC-deaminase and Exopolysaccharide Producing
Bacillus cereus: Effects on Biochemical Profiling. Sustainability 2020, 12, 2159.
• Duffy, S.; Holmes, E.C. Multiple introductions of the Old World begomovirus
Tomato yellow leaf curl virus into the New World. Appl. Environ. Microbiol. 2007,
73, 7114–7117. [PubMed]
• Moriones, E.; Praveen, S.; Chakraborty, S. Tomato Leaf Curl New Delhi Virus:
An Emerging Virus Complex Threatening Vegetable and Fiber Crops. Viruses
2017, 9, 264.
