A Review on Machine Learning Techniques for Rice Plant Disease Detection in Agricultural Research

Abstract

Disease identification in plants is important to avert the losses in the quantity and production of agricultural products. The problems in the agricultural sector are lessoned by employing various machine learning and image processing techniques. This review mainly focus on rice plant disease detection centered on image inputs of infected rice plant by using disparate ML and image processing techniques. Also, the notable ML and image processing concepts in detecting and classifying the plant diseases are discussed. Probabilistic Neural Network (PNN), Genetic Algorithms (GA), k-Nearest Neighbor Classifier (KNN) and Support Vector Machine (SVM) are the various classification techniques used in various applications in the agricultural research. Different input data yields varied quality of an outcome and so selecting a classification method is a critical task. Biological research, agriculture, etc. are the disparate fields where the plant leaf disease classifications are applied. The detailed study on the diseases of rice plant, size of image dataset, preprocessing, segmentation techniques, classifiers are presented in this paper.