<![CDATA[Chili plants (Capsicum sp.) are one of the important horticultural commodities in Indonesia with high economic value. However, chili productivity is often reduced due to leaf diseases such as leaf curl, yellow leaf virus, and leaf spot disease. Manual disease identification conducted by farmers still has several limitations because it requires considerable time, experience, and is prone to observation errors. Therefore, an automatic system is needed to support early detection of chili leaf diseases quickly and accurately. This study aims to develop a chili leaf disease classification system using a Convolutional Neural Network (CNN) based on leaf images. The dataset used in this study consists of chili leaf images categorized into four classes, namely healthy leaves, leaf curl, yellow leaf, and leaf spot. The research stages include dataset collection, image preprocessing, data augmentation, CNN model training, and model evaluation using a confusion matrix with performance metrics including accuracy, precision, recall, and F1-score. The results show that the CNN model is capable of classifying chili leaf diseases with satisfactory performance. Based on the evaluation results, the model achieved a precision value of 0.75, a recall value of 0.53, and a mean Average Precision (mAP@0.5) value of 0.60. The developed system is also able to display classification results along with the confidence score of the prediction. Therefore, the CNN method has strong potential to be implemented as an image-based early detection system for chili leaf diseases to assist farmers in monitoring plant conditions more effectively.]]>
