A Novel Approach for Detection and Classification of Rice Leaf Diseases

Abstract

India stands tall as one of the world’s largest rice-producing countries. A major part of Indian agriculture consists of rice as the principal food crop. Rice farming in India is challenged by diseases that can infest and destroy the crops causing detrimental losses to the farmers. Thus, the detection of diseases like “leaf smut”, “brown spot”, and “bacterial leaf blight” becomes a need of the hour. In this paper, we have proposed a way that can efficiently detect and classify these three diseases through image processing. The research can help in knowing if the rice crop is infested with the diseases or not. Images of the infected crop can be used in a real-life scenario and one can know if it is infested with any of the three diseases mentioned. The detection and classification of these diseases have been made possible using various state-of-the-art classification models, like support vector machine (SVM), random forest, KNN, naive Bayes, and neural network.