AI Based Paddy Disease Prediction System

Rice (paddy) serves as a fundamental nutritional source for over half the global population. However, its cultivation is frequently undermined by a variety of diseases, which can lead to severe reductions in crop yield. The prompt and precise identification of these diseases is therefore paramount for deploying effective control strategies. This research introduces a novel framework based on Artificial Intelligence (AI) for the automated classification of common paddy ailments using digital leaf images. The core of this system is a deep Convolutional Neural Network (CNN) that leverages a transfer learning methodology. Specifically, a pre-trained EfficientNet model was fine-tuned using a unique, custom-curated image dataset to identify the visual characteristics of diseases like Leaf Blast, Brown Spot, and Hispa. The model’s performance was benchmarked on its ability to accurately differentiate between various disease states and healthy specimens. The resulting high accuracy underscores the system’s viability as a powerful, economical, and readily deployable diagnostic aid for farmers, facilitating more rapid and evidence-based crop management decisions.