AgriVision: A Deep Learning Framework for EcoFriendly Crop Health Monitoring

Abstract

Agriculture remains the cornerstone of India’s economy, accounting for approximately 17% of the national GDP and offering livelihoods to more than 60% of the population. However, challenges like crop diseases, changing climate conditions, and sunsustainable farming practices continue to threaten agricultural productivity and food security. With technological advancements becoming increasingly accessible, integrating AI and drone- based monitoring systems has emerged as a viable solution for improving crop health management and promoting sustainable agriculture. This work introduces AgriVision, an AI-powered crop health monitoring system designed to detect plant diseases in their early stages using dronecaptured imagery and a Convolutional Neural Network (CNN) model. Unlike traditional methods that rely on manual inspection and lab testing, AgriVision enables real-time, non-invasive detection and provides actionable insights through a userfriendly dashboard built using Flask and Streamlit. The system also incorporates environmental data via the OpenWeather API to provide timely weather forecasts and treatment suggestions. By promoting precision agriculture through techniques like targeted spraying and use of natural pest control methods, AgriVision aims to reduce crop loss, minimize chemical usage, and support sustainable agricultural practices and long-term food security. Through the seamless integration of hardware, AI algorithms, and intuitive interfaces, this system empowers farmers with the tools needed to make informed decisions, enhancing yield, resilience, and sustainability in the face of agricultural challenges.

Plant phenotyping relevance

ドローン画像とCNNにより植物病害を早期検出するシステム開発が中心で、植物の病害状態を画像から推定するため、植物フェノタイピング手法として採用する。

Code and data availability

植物フェノタイピング解析を再現する公開資産であることを、入力本文と直接リンクから確認できなかったため保留しました。

No evidence-backed public reproduction asset is currently recorded.