Plant Disease Prediction System Using Deep Learning and Image Analysis

Plant diseases pose a significant threat to agricultural productivity, causing substantial economic losses and food security challenges worldwide. Traditional disease identification relies on manual visual inspection, which is time-consuming, subjective, and often inaccessible to small-scale farmers. This paper presents an automated Plant Disease Prediction System utilizing deep learning techniques for disease identification from leaf images. The proposed system employs the MobileNetV2 architecture pre-trained on ImageNet, fine-tuned for classification of tomato and potato leaf diseases. The model was trained and evaluated on the PlantVillage dataset, comprising images of healthy and diseased leaves across multiple disease categories. The system achieves 90.47% classification accuracy with a precision of 91.18% and provides a user-friendly web interface for disease diagnosis. Experimental results demonstrate the ef-fectiveness of the proposed approach in enabling rapid, accurate, and accessible plant disease detection, contributing to sustainable agricultural practices and improved crop management.