<![CDATA[This study aims to develop a classification system for Indonesian traditional masks using the Convolutional Neural Network (CNN) method. Traditional masks exhibit rich visual diversity that reflects the cultural identities of various regions in Indonesia; however, manual identification is time-consuming and prone to errors. The system developed in this study is capable of classifying five types of masks—Cirebon, Balinese, Malangan, Dayak, and Betawi masks—based on digital images with a high level of accuracy. The proposed CNN model achieved an accuracy of 92.3% on the test dataset, with an average macro F1-score of 0.91. Data preprocessing and augmentation techniques, including rotation, flipping, and brightness adjustment, effectively enhanced the model’s performance by reducing the risk of overfitting. These results demonstrate the strong potential of deep learning technology in supporting cultural heritage preservation through the digitalization and automated classification of Indonesian traditional masks.]]>
Copyrights © 2025
