Segmentation and Multi-Layer Perceptron: An Intelligent Multi-classification model for Sugarcane Disease Detection
Rishabh Sharma, Vinay Kukreja
Abstract
Sugarcane disease detection has been an active area of research for past decades, due to the increasing demand and supply of the crop, the higher production level led to a hike in the number of diseases encountered in the crop. Sugarcane red rot (SRR) is one of those diseases which has a draconian effect on sugarcane production and to eliminate that factor a multi-layer perceptron (MP) based deep learning (DL) model has been developed to build a system for identification and classification of 2000 image dataset of SRR disease based on 5 different disease levels. 99.12% of binary classification and 99.15%% of best multi-classification accuracy have been encountered along with a comparison of various levels of SRR disease. The proposed model has been proved to be an efficient model in terms of accuracy results for the classification of an image.