Proxsee: Anti-Spoofing Web-Based Application Attendance Monitoring System Using Face Liveness And Google's Facenet

Authentication is one of the significant issues in the era of information system. Current biometric methods for attendance are too intrusive. Among other things, human face Recognition (HFR)is one of known techniques which can be used for user authentication. As an important branch of biometric verification, HFR has been widely used in many applications, Such as video monitoring/surveillance system, human-computer interaction, door access control system and network security. This paper presents a stress-free non- intrusive way of taking class attendance using face as the biometric. Proxsee has a reliable anti-spoofing facial recognition system that utilizes different convolutional neural network (CNN) of Face Liveness and Face Recognition based on Google's FaceNet. ProxSeee must contain vigorous algorithm that is capable of enrolling faces to the system, identifying presentation attacks and recognizing faces matched from the database. To utilize its main functionalities, it must be run on a server with a camera that will identify attacks and monitor student attendance. To do this, the system will be set up beside the ae entrance door of a classroom and comprise of a web camera and computer. Upon taking attendance, as soon as the system detects any form of presentation attacks, it'll prompt back to rescanning. Also, recognition is limited in the frontal face. Results shows that 8/10 attempts of spoof attacks are being detected by the system. Unfortunately, only 2/10 enrolled people are being recognized correctly. Data logging is working perfectly.