<![CDATA[This study aims to develop an object detection model based on YOLOv5 to classify debit card types. With the advancement of financial technology, the need for automated systems to identify debit cards has become essential to enhance transaction efficiency and security. The research methodology involves five main stages: dataset collection, data preprocessing through labeling and resizing to 640 x 640, dataset augmentation, YOLOv5 model training, and model evaluation. The dataset used consists of three categories of debit cards, with a total of 300 images. The results demonstrate that the YOLOv5 model achieves excellent performance with a mean average precision (mAP) of 92.7% and an object loss value of 0.08. The high mAP value indicates the model’s capability to accurately recognize objects, while the low object loss value reflects minimal detection errors during testing. In conclusion, YOLOv5 has proven to be reliable for application in debit card detection systems. This study provides significant contributions to the development of automation systems in the financial sector, particularly in improving the efficiency and accuracy of identification processes. It is hoped that this research will serve as a foundation for further studies with broader datasets, the application of more advanced augmentation techniques, and the utilization of more sophisticated hardware to enhance model performance.]]>
Copyrights © 2025
