A STUDY ON DISEASE DETECTION AND REMEDY IDENTIFICATION IN LEAVES

Abstract

Plant diseases significantly impact agricultural productivity, leading to major crop losses. Early detection and timely treatment are essential to minimize damage. This project introduces a Machine Learning-based mobile application for detecting diseases in tomato, grape, mango, and corn leaves using Convolutional Neural Networks (CNNs) with over 95% accuracy. Once detected, the system suggests appropriate treatments like fungicide application, pruning, or improved irrigation.
Implemented with TensorFlow, OpenCV, and the PlantVillage dataset, the app allows farmers to capture leaf images for real-time diagnosis and treatment recommendations. It also features a mapping system for locating nearby remedy stores and a stock management system for shop owners to update product availability and prices. By integrating AI, smart farming, and marketplace features, this project enhances efficiency, reduces crop losses, and improves overall agricultural productivity.