ARCHIVES

E- Rice cultivation is a cornerstone of global food security, yet it remains highly vulnerable to leaf diseases that result in significant yield losses. Traditional manual detection is often labor-intensive and prone to human error, highlighting the urgent need for automated, high-precision diagnostic tools. This project proposes a robust, two-stage cascaded framework designed to streamline the detection process while enhancing accuracy. In the first stage, a binary classifier filters healthy leaves from diseased ones to reduce computational overhead. The second stage utilizes a deep learning model integrated with a hybrid attention mechanism to precisely identify specific disease classes by focusing on critical pathological features. To bridge the gap between complex AI logic and practical farming, the system incorporates Explainable AI (XAI) to build user trust and delivers both the diagnosis and recommended remedies through an accessible digital dashboard. This integrated approach ensures a reliable, transparent, and actionable solution for modern precision agriculture.