Attendance System based on Deep Learning Face Recognition without Queue

Management of the presence of a lecturer and student in the classroom is one of the factors that determine the quality of learning in a university. Today, there are many researchers developing a system for managing student attendance in classrooms based on facial recognition technology, to replace the previous system. However, there is a new problem coming, when students come to the classroom simultaneously, then comes the queuing problem that often disrupts the learning process in the classroom. To overcome this problem in this article we will discuss the prototype system for managing student attendance in classroom based on deep learning facial recognition technology without queue. Feature extraction uses 128-d facial embeddings from FaceNet, and code implementation uses libraries originating from opencv and dlib. The installation of the device is done in such a way that it allows the process of identifying the faces of students present in the classroom can be done during the learning session in the class is still ongoing. The results show that this system is able to detect the presence of students in the classroom well.