<![CDATA[Students' attendance at school is still recorded manually using books. This has weaknesses such as the attendance book being susceptible to damage and loss which can cause loss of important attendance data and requires a long time to recapitulate the results of the attendance recapitulation ahead of the semester increase. Therefore, this research discusses the application of the You Only Look Once (YOLO) method to recognize students' faces because YOLO is a real-time object detection method that is very fast and has high accuracy. The dataset used consists of 1,250 images with 70% train data, 20% valid data, and 10% test data which was trained with epoch 50, epoch 75, and epoch 100 resulting in an accuracy of 100% for each epoch. The model that has been trained can recognize students' faces well and can be applied to computer vision-based software to assist teachers in taking attendance and recapitulating attendance results in a certain period so that it doesn't take a long time to recap.]]>
Copyrights © 2025
