Wulandari, Haerunnisya Makmur, Asmaul Husna Nasrullah, Nur Azizah Eka Budiarti, Satria Gunawan Zain, Abdul Wahid
Banana (Musa spp.) is a crucial commodity in tropical and subtropical regions, serving as a significant source of income for farmers and a staple food for millions of people worldwide. Due to its high export value, banana production has been increasing to meet growing market demands. However, banana plants are vulnerable to various pests and diseases, particularly on the leaves, such as Cordana, Pestalotiopsis, and Sigatoka, which can impede fruit production. Farmers typically rely on visual inspection to detect these diseases, but this method is often inaccurate due to the similarities in symptoms among these diseases. To address this challenge, digital image processing techniques combined with machine learning approaches can be utilized to enhance the accuracy and efficiency of disease detection on banana leaves. This study aims to develop a system for identifying three banana leaf diseases by employing digital image processing techniques, including threshold segmentation, color feature extraction using the LAB color space, and area analysis of segmented objects. Additionally, information fusion of extracted features was performed to improve classification accuracy. The classification results demonstrated excellent performance using Random Forest, achieving an accuracy rate of 94.75%, with precision, recall, and F1-score values of 94.37%, 93.41%, and 93.89%, respectively. The system was successfully implemented using MATLAB, enabling users to load images, perform segmentation, and easily view classification results. © 2026 Universitat Autonoma de Barcelona. All rights reserved.
Department of Computer Engineering, State University of Makassar, Makassar, Indonesia
