THE DETECTION OF FACE RECOGNITION AS EMPLOYEE ATTENDANCE PRESENCE USING THE YOLO ALGORITHM (YOU ONLY LOOK ONCE)

Abstract

face detection is a fundamental and important process in the field of facial recognition. The aim of this face detection was to determine the presence and mark the position of the face through an image called a bounding box. The problem that we often encounter in attendance machines is that it is often difficult and takes a few seconds to perform facial recognition with a low level of accuracy. Therefore, an attendance system that is faster and more accurate in recognizing faces is needed with the aim of increasing better accuracy. This research was undertaken by applying the You Only Look Once (yolo) algorithm with several test scenarios to see the performance generated by the system, because yolo is one of the fastest and most accurate methods for object detection and even exceeds 2 times the capabilities of other algorithms. The yolo (you only look once) algorithm is an architecture of deep learning and an algorithm developed to detect an object in real-time. Regarding to the classification performance measurement from the training data, it indicated that the accuracy value has reached 90% thus it can be concluded that the system can work wellt.