Rice Leaf Disease Classification Using SVM-LSTM Technique for Agricultural System

Rice, a crop widely produced in the nation, is susceptible to numerous diseases during its growth period. To prevent damage and enhance the production of crops in the rice field, rice diseases must be identified promptly. Farmers find it difficult to correctly identify diseases like these by manual due to a lack of knowledge and expertise. Timely detection of diseases and rapid application of necessary treatments to afflicted plants are crucial for ensuring the vigorous and normal growth of rice plants. The precise detection of diseases in plants is essential in contemporary agricultural systems. Consequently, artificial intelligence could potentially utilized to identify conditions in rice leaves. We propose an innovative support vector machine classifier with long short-term memory (LSTM) model designed to classify common rice leaf diseases. This study examines four distinct rice leaf diseases such as blast leaf disease, bacterial blight leaf, sheath blight leaf, and brown spot leaf disease. The proposed classifier attains an accuracy of 96%. The results of disease diagnosis in rice highlight the efficacy of the proposed method.