A HYBRID MACHINE LEARNING APPROACH FOR EARLY DETECTION OF PADDY BLIGHT DISEASE

Paddy blight is a widespread disease that affects various parts of the paddy plant, including leaves, bark, nodes, neck, part of rays, and leaves sheath. The symptoms of the disease manifest as pale yellow to pale green leaves with eye-shaped lesions, distorted margins, and gray or white centers. As the lesions expand, the leaves progressively wither and dry out, eventually leading to rot and death of the affected plant parts. In this study, we propose a machine learning algorithm for detecting paddy disease by analyzing changes in paddy leaves and correlating them with existing paddy images. The algorithm incorporates fuzzy logic and deep learning techniques to enhance disease detection accuracy and provide appropriate treatment recommendations. By leveraging the power of these advanced technologies, the proposed approach aims to facilitate early detection and effective management of paddy diseases, ultimately improving crop yield and ensuring food security.