Convolutional Neural Network-Based Crop Disease Detection with Farmer Mobile Application

Objective: To develop and deploy a mobile application for Indian farmers that detects crop diseases from leaf images using a trained convolutional neural network model.

Crop diseases severely impact agricultural productivity and income for Indian farmers, especially those with limited access to expert diagnosis. Early and accurate disease detection is crucial to prevent losses, but rural farmers often lack the tools or knowledge to identify diseases from leaf symptoms.

This project proposes an AI-driven solution: a convolutional neural network (CNN) trained on labeled leaf images, integrated into a user-friendly Android mobile app. Farmers can capture leaf images, receive instant disease classification and suggestions, and access disease information in local languages.

Key deliverables include: a fully implemented CNN model (using a dataset like PlantVillage or custom Indian crop datasets), an optimized backend/API for inference, and a live demo of the mobile app with real-time diagnosis, offline support, and recommendation features. The working model will be validated on field samples and demonstrated before examiners.

Industry and societal impact: The solution empowers farmers with actionable insights, reduces dependency on experts, and can scale across crops and regions. It supports digital agritech adoption, potentially improving yields and rural livelihoods.