Image based plant disease detection

Project Code :TCPGPY2091

Objective

The main objective of this project is to design and implement an image-based plant disease detection system using both classical Machine Learning and Deep Learning algorithms. The project aims to train and evaluate Decision Tree, Random Forest, SVM, and XGBoost models using extracted image features. In parallel, Deep Learning models such as CNN, MobileNet, VGG16, VGG19, and MobileNetv2+Cbam, ResNet18+Cbam are trained directly on plant leaf images. Another objective is to perform a comparative analysis of these models based on accuracy, efficiency, and performance metrics. The project also focuses on developing a web-based application using Flask that allows users to register, log in, upload images, and view classification results. By integrating multiple algorithms into a single platform, the project aims to identify the most suitable approach for plant disease classification and provide a structured framework for research and experimentation.

Abstract

The detection of plant diseases is essential for maintaining agricultural productivity and ensuring food security. This study explores the effectiveness of deep learning (DL) algorithms in identifying plant diseases from images, using the PlantVillage dataset available on Kaggle. In Phase 1 of the project, we focus on implementing Convolutional Neural Networks (CNN) and MobileNet algorithms to classify plant diseases. The dataset, consisting of color images of various plant species affected by different diseases, serves as the foundation for our analysis. The results from applying these deep learning techniques highlight their potential for improving plant disease detection. In Phase 2 of the project, further optimizations and comparisons will be made to refine the classification accuracy. This research aims to contribute to the development of more efficient and reliable plant disease detection systems, ultimately aiding in better agricultural management practices.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.