<![CDATA[The complexity of asymmetric visual patterns and overlapping ornaments in batik motifs presents challenges for automatic identification in modern applications. The limitations of conventional classification methods have created a need for computer vision techniques capable of accurate object localization. This study aimed to detect and classify various Indonesian batik motifs in real time using the YOLOv11 model. The research methodology followed the Cross-Industry Standard Process for Data Mining (CRISP-DM), consisting of six structured phases, from business understanding to system deployment. The study focused on major variations of batik motifs originating from different regions of Indonesia, with model performance evaluated using the mean Average Precision (mAP) metric. The results demonstrated that: (1) the YOLOv11 model achieved an mAP50 of 77%; (2) data augmentation effectively reduced the risk of model overfitting; and (3) the trained detection model was successfully integrated into a web-based platform, enabling users to perform real-time image testing.]]>
Copyrights © 2026
