Sistem Presensi menggunakan Deteksi Objek Wajah Mahasiswa Berbasis YOLO-V5

Abstract

Attendance is a crucial aspect across various sectors, such as corporate, governmental, and educational institutions, for efficient management. The advancement of deep learning technology, particularly in the field of facial recognition, has become a primary focus in enhancing identification accuracy. In this context, visual object detection through computer vision plays a key role, with the You Only Look Once (YOLO) method emerging as a leading choice for real-time object detection across various media, including webcams, due to its speed and efficiency. This research proposes the application of YOLO-V5 in the development of a student attendance system. This approach utilizes deep learning and data augmentation to enhance the accuracy of student identification. YOLO-V5 enables efficient real-time object detection, achieving an accuracy rate of up to 95% on each frame. The implementation of the student attendance system using the YOLO-V5 method successfully detects student attendance in real-time with a high level of accuracy. This demonstrates the potential of this method to improve the efficiency of attendance management and its suitability for integration into student attendance systems. This research represents a significant advancement in the use of deep learning and computer vision to increase the accuracy and efficiency of attendance management