Afnan Afnan, Cheng Chen, Putri Nazwa Safira, Maulisa Oktiana, Melinda Melinda
Sustainable food production, as emphasized in the United Nations Sustainable Development Goal 2 (Zero Hunger), requires timely and accurate detection of plant diseases to reduce crop losses and improve agricultural productivity. This paper presents an automated chili leaf disease classification system based on a ResNet50 convolutional neural network to support early disease identification in chili cultivation. Manual inspection of chili leaves is labor-intensive and prone to error due to the small size and visual similarity of leaf symptoms across different diseases. To address this challenge, a novel dataset consisting of 1,500 chili leaf images (after augmentation) was constructed, covering five leaf-condition classes: curly leaf, spotted leaf, whitefly-infected leaf, yellow virus leaf, and healthy leaf. The ResNet50 model was trained and evaluated under varying learning rates (10-3 and 10-4) and training epochs ranging from 20 to 100 to assess its classification performance. Experimental results indicate that the ResNet50 configuration with a learning rate of 10-3 and 100 training epochs achieves the best performance, yielding an accuracy of 99.3%, precision of 99.4%, recall of 99.4%, and F1-score of 99.2% on the constructed dataset. These findings confirm the effectiveness of deep learning-based image classification for automated agricultural inspection and highlight its potential contribution to sustainable food production through enhanced crop health monitoring and early disease detection. © 2026 IEEE.
Universitas Syiah Kuala, Department of Electrical and Computer Engineering, Banda Aceh, Indonesia; Lubar College of Business, University of Wisconsin-Milwaukee, Milwaukee, United States
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock