Design Of Face Recognition System Based On OpenCV

With the rapid development of science and technology,the requirements of convenience and security in identity authentication are getting higher and higher.Traditional identification methods have been unable to meet people's needs,and technique of biometric identification has become a hot research direction for personal verification.In many biometric identification techniques,face recognition technology attracts many scholar's research and now it is already a hot computer vision technology because of non-mandatory,concurrency,simplicity and so on.However,with the popularity of intelligent devices,system attacked by forged face becomes more and more easy.Some unscrupulous people use photos or video to attack the authentication system and engage in some illegal acts.Therefore,it is important to verify the authenticity of the face.The technology of face liveness detection has become the focus of the research.The main contents of this paper are face detection,face recognition,system interface and functional testing.Face detection is the basis of the entire system,and information can be identified after face detection.Face recognition is mainly to identify the information of the face,and it is the core of the entire system.The system interface is the most intuitive description of the system function and also is a necessary condition for the system integrity.Functional testing is a necessary step to verify the feasibility of the system.In this paper,it designs a face liveness recognition system which combines living detection and face recognition technology.The face recognition system is based on OpenCV open source visual library and uses Visual Studio and Qt as the development platform.The system uses Haar classifier as the face detector,combines CMU open source feature point calibration model and LBP recognition algorithm to build a live face recognition system.The camera collect face video into the system and switch interface function can achieve the face detection,live recognition and face recognition function.The whole paper is designed in double work mode which uses living detection module flexibly.That is,calling living verification function when the system is not under supervised and not calling living verification function when the system is under supervised.The research shows that the dual operating mode of the system can reduce the unnecessary time and make great significance to the further development of the system.