Machine Learning-Based Approach to Detect and Classify Rice Blast Disease

Rice blast disease is caused by fungus, early detection and cure using pesticides will increase the yield of the crop. Computer-aid-based image processing techniques approaches are of more interest and are finding application in the detection of the rice blast disease. For the diagnostics systems for computerized-based approaches, Machine learning utilizing an effective algorithm and classification of rice disease is critical. In this work, the grayscale images are utilized through the k-means algorithm for segmentation of the rice disease-infected area of the rice plant and the Support Vector Machine (SVM) for categorizing the type of the disease. The present study could identify five types of rice disease. Further, compared to the existing methods, the approach presented by this study has a score of 96.77% accuracy, enabling it to give greater results.