Web-Based Sustainable Detection and Treatment Recommendation System for Wheat Plant Diseases Using Convolutional...

Area of Science:

• Agricultural Science
• Plant Pathology
• Computer Vision

Background:

• Wheat rust diseases, including yellow rust (YR) and brown rust (BR), pose a significant threat to global food security, causing substantial yield losses.
• Early and accurate diagnosis of these fungal infections is crucial for effective crop management and minimizing economic impact.

Purpose of the Study:

• To develop and evaluate a web-based system utilizing a Convolutional Neural Network (CNN) for the rapid identification and classification of wheat diseases, specifically YR and BR.
• To provide farmers with an accessible and user-friendly tool for timely disease diagnosis and management.

Main Methods:

• A dataset of labeled images of healthy wheat plants, YR, and BR was curated.
• A CNN model was trained on this dataset for image classification.
• The trained CNN model was integrated into an intuitive web application, including a recommendation module for disease treatment.

Main Results:

• The CNN model achieved a high classification accuracy of 96% in identifying wheat diseases.
• The web system successfully integrated disease identification with a treatment recommendation module.
• The user-friendly interface makes the tool practical for farmers.

Conclusions:

• The developed CNN-based web system offers a reliable and efficient solution for the early detection and management of wheat rust diseases.
• This technology empowers farmers with a practical tool to protect wheat crops, contributing to improved agricultural productivity and food security.
• The system's ability to recommend treatments enhances its utility beyond mere diagnosis, supporting comprehensive crop health management.