Rice Plant Leaf Disease Classification using Machine Learning Algorithm

Abstract

Rice is most important food for west Bengal and Tamil Nadu. West Bengal is in first place in Rice production. It’s crucial to identify any diseases early on and cure the damaged plants before anything else can happen to the rice plants in order to guarantee their healthy and balanced growth. Many traditional methods were there, but they are time consuming. In order to identify afflicted plants early, treat them promptly, and, most importantly, plan future preventative measures to reduce losses, it is critical to monitor illnesses, their occurrences, and frequency. This study suggested using algorithms based on machine learning to detect and classify the diseases affecting rice plants. The three groups of diseases—bacterial leaf blight (BLB), leaf smut and brown spot—included in the data sets were gathered from the Rice Leaf Database. The input image features were extracted using an algorithm Convolutional Neural Network (CNN). Decision Tree (DT), K-Nearest Neighbour (KNN) algorithm, support vector machines (SVM), Logistic Regression (LR), and KNN algorithm are used the retrieved features individually for classification. The suggested methods effectiveness is evaluated using the recall, F1 score, accuracy and precision metrics. From the analysis it shows that the accuracy of the proposed model is 96.7%.

Topics & Concepts