Real - Time Plant Disease Detection by Ai

Abstract

Timely and precise identification of plant diseases is essential for improving agricultural yield, reducing financial losses, and supporting sustainable farming practices. In this study, a lightweight real-time plant disease detection system intended for edge devices such as smartphones and embedded platforms is presented. The proposed framework employs an optimized convolutional neural network along with model compression methods to enable efficient offline inference on real-time field images, ensuring high accuracy with minimal latency and reduced computational demand. The system is specifically tailored for use in rural and underdeveloped areas where internet connectivity is limited, offering a reliable, practical, and scalable approach for real-time crop health monitoring and agricultural decision support.

Code and data availability

The paper describes a real-time plant disease detection system using a lightweight CNN trained on PlantVillage and a custom detection.py script, but provides no public dataset, image, code, or model deposit with an authors' URL. PlantVillage is a cited third-party benchmark, not a paper-specific asset, and no code or T

No evidence-backed public reproduction asset is currently recorded.