<![CDATA[Aksara Ulu Rejang is a cultural heritage of Bengkulu that is starting to be forgotten due to the lack of literacy and digital documentation. To preserve it, this study developed an automatic detection system for aksara Ulu Rejang using the You Only Look Once (YOLO) method, a deep learning technique that is effective in real-time object detection. The research data were images of aksara Ulu Rejang from the Bengkulu Museum. The research process included data acquisition, image preprocessing (resize and augmentation), annotation with Roboflow, model training in Google Collaboratory, and evaluation using Mean Average Precision (mAP). The dataset consisted of 710 images with 253 letter and punctuation classes. The results showed that the YOLO model was able to detect the script well, obtaining an mAP value of 0.793 at an IoU threshold of 0.5–0.95. This model is expected to be the basis for developing a digital Ulu script recognition system to support the preservation of Bengkulu culture.]]>
Copyrights © 2026
