Deep Learning Model for Predicting Rice Plant Disease Identification and Classification for Improving the Yield

Rice (Oryza sativa) stands as a cornerstone of global nutrition, feeding over half the world's population. However, its yield and quality are perennially threatened by a plethora of diseases, challenging farmers to identify and combat them timely. Traditional methods, rooted in visual inspection, have been both labor-intensive and error-prone. Addressing this, the study integrated Neural Architecture Search (NAS) with reinforcement learning, further enhanced by transfer learning, to devise a model capable of automated rice leaf disease detection. The proposed RL-based NAS model displayed a remarkable accuracy of 96.99%, a significant leap from traditional inspection methods and even basic computational models. When dissected for specific diseases like Bacterial Leaf Blight and Rice Blast, the model's proficiency remained consistent, underscoring its robustness and adaptability. This convergence of advanced machine learning with agriculture promises a paradigm shift in farming practices. By enabling precise, timely, and efficient disease detection, the model offers potential benefits ranging from improved yields to cost savings. This study, thus, not only highlights the transformative power of integrating technology into agriculture but also paves the way for further innovations that can revolutionize sustainable farming globally.