Applying Image Classification for Detect Leaf Disease: Case Study for Porang Plant

Abstract

The aim of this research is to identified and classification of leaf disease for Porang plant. In this study, the symptoms analysis of the plant leaf is applied under GLCM and artificial neural network to classified and identified object in real time. This work describes an automated method for detecting Porang leaf diseases that is based on four steps: preprocessing, segmentation, feature extraction, and classification. The experimental findings demonstrated that the proposed method can efficiently and accurately to detect of Porang leaf disease. The experimental results show that on unseen images with complicated background conditions. The regression coefficient and root mean square error prediction were 0.905 and 0.5. As a result, the proposed method is accurately and performs in real time.