<![CDATA[The purpose of this study is to increase the accuracy for rice plant disease classification by developing a hybrid optimization model using Convolutional Neural Network (CNN) in combination with Extreme Learning Machine (ELM), followed by Support Vector Machine. A key issue is to overcome with traditional expert systems that difficult, due the variation differences and complex among rice plant image data set. For feature extraction, plant images are passed through CNN and for classification ELM & SVM used. Experimental results show the best accuracy of 98.63% is attained using CNN+ELM model on images resized to 100x100 pixels and has precision, recall, F1-Score all at value=0.99 By comparison, the CNN+SVM model achieves an accuracy of 91.92% using that same image size. Top AbstractIntroductionMethodsResultsDiscussionConclusionReferencesOverall, the proposed CNN+ELM combination can classify rice plant diseases better than using only a conventional approach (CNN) through various results from devices with limited computing power. The study presents a novel plant disease detection system that can be utilized for the development of precise tools to help improve agricultural management practices.]]>
Copyrights © 2025
