Computer aided detection and classification of plant leaf diseases using Deep Convolutional Neural Network

Abstract

Agriculture is the backbone of the Indian economy and the production rate of agricultures is based on detection and classification of plant leaf diseases. The conventional machine learning methods are failed to meet the maximum classification accuracy. Therefore, this article is focused on implementation of transfer learning models for feature extraction. Initially, non-local means filters (NLMF) are used to perform the preprocessing operation, which removed the different types of noises and also enhances the region of plant diseased region. Then, hybrid k-means clustering (HKMC) is used to localize the diseased region by segmentation operation. In addition, log of gradients descriptor (LOG) is used to extract the diverse features from the dataset. Finally, deep convolutional neural network (DCNN)is used to classify the different types of plant diseases. Further, the simulation results show that the proposed method resulted in superior performance as compared to existing approaches for both subjective and objective analysis.