Efficient Prediction of Blast Disease in Paddy Plant using Optimized Support Vector Machine

Abstract

Diagnostics of plant diseases are essential to predicting yield loss, detecting infections, studying host–pathogen interactions, and detecting host resistance. Early detection of plant leaf disease can increase yields. This is possible if there are automated systems to help farmers diagnose rice diseases from the pictures of plant leaves. Therefore, automatic leaf detection using machine learning is proposed in this study. The proposed approach has three stages, namely, pre-processing, feature extraction, and classification. Initially, the input image is converted into red, green and blue and the noise in the green band is removed using a median filter. Initially, the input image is converted into red, green and blue and the noise in the green band is removed using a median filter. Then, important features of the green band are extracted. An Optimized Support Vector Machine (OSVM) classifier uses the extracted features to classify an image as normal or pathological. To improve SVM performance, the SVM parameters are chosen optimally using the Adaptive Sunflower Optimization (ASFO) algorithm. Then, the infected region is separated using a level-set segmentation algorithm. The efficiency of our work is analyzed based on accuracy, sensitivity, and specificity, the proposed method reached the maximum accuracy of 97.54% for plant disease prediction.