CROP DISEASE DETECTION USING DEEP LEARNING

Abstract: This project focuses on the development of a web-based system for detecting crop diseases using deep learning techniques. The system uses a Convolutional Neural Network based on the MobileNetV2 architecture trained on the PlantVillage dataset to classify healthy and diseased crop leaves. The trained model is integrated into a Flask-based web application that allows users to upload leaf images and obtain disease predictions in real time, along with confidence scores and precautionary and treatment-related information. The lightweight nature of the selected model ensures fast prediction time while maintaining reliable classification performance. By combining deep learning with web technologies, the proposed system offers an accessible and cost-effective solution for crop disease identification, reducing dependence on manual inspection and supporting timely disease management to improve crop productivity.

Keywords: Crop Disease Detection, Deep Learning, Convolutional Neural Network, MobileNetV2, Flask, PlantVillage Dataset, Image Classification, Transfer Learning.