Intelligent Epidemic Prevention System Based On Face Recognition

Under the background that Covid-19 is spreading across the world,life style of people has to confront a series of challenges and changes.Nowadays masks have almost become necessities for people in public places,but most access control systems(ACS)cannot recognize people wearing mask and complete the authentication procedure to deal with increasingly serious epidemic pressure.Consequently,many public entries have turned to attendant mode,which maybe bring low efficiency and high possibility of infection.Compared with automation mode,this mode needs more working hours,but working hour of supervisors are limited.Based on above problems,there needs new ACS to deal with it.According to the requirement of school condition,this paper proposes one intelligent epidemic prevention algorithm based on face recognition.This algorithm include two real-time function modules: face authentication and mask detection,it can improve the safety of ACS and protect the property for the institute in college effectively through fusing liveness detection.The main research work of this paper is as follows:(1)Proposed an identity authentication system based on face recognition.This system has the features,such as natural uniqueness,friendly verification method and higher security,meets the non-contact needs of epidemic prevention work.(2)Proposed the algorithm based on face recognition fusing on mask detection,which reminds users to wear a mask in the checking scene.(3)Proposed the methodology realizing liveness detection with color texture analysis.It can achieve face liveness detection and face authentication in the situation of monocular camera without user cooperation.In order to validate above design,Chinese Academy of Sciences Institute of Automation-Face Anti-Spoofing Datasets(CASIA-FASD)and Replay-Attack datasets are utilized as the benchmark datasets of the experiment.The Half Total Error Rate(HTER)and Equal Error Rate(EER)were9.7% and 5.5% in the liveness detection respectively.The time cost to process a frame of image in the whole process was 0.12 seconds.Above result verifies the feasibility and effectiveness of this methodology.