From Binary to Multi-Classification: A Novel Framework for the Detection and Severity Classification of Rice Sheath Rot Disease

A multi-classification model for the rice sheath rot disease is proposed in this research work. The model is built on six distinct severity levels and makes use of a dataset of 10,000 images with Deep Learning (DL) algorithms. Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) have been used in the model to obtain a high level of accuracy in the identification and classification of rice sheath rot disease. The accuracy of the proposed model was measured at an average of 95.14%, which is much greater than the accuracy attained by typical machine learning models. Rice sheath rot is a disease that can assist in minimizing crop losses and boosting agricultural output. The findings suggest that the proposed model is helpful in precisely recognizing and managing the disease. In future work, the dataset will be extended for investigating the possibility of integrating this model with a mobile application so that it is more widely accessible.