Research And Design Of Embedded Entrance Guard System Used Human Fece Recognition

Along with the computer and biological engineering technology rapid development, the requirements to the rapid and effective automatic identity verification are increasingly urgent. So the body’s biological characteristics for identification have become the first choice of security system. Compared to other biological characteristics, face feature is the contact resistance, good stability and difficult to counterfeit, and easy for users to accept. So face recognition technology has a broad application prospect in the market.This paper combines face recognition technology and embedded technology. In face recognition system realization of problems, mainly from three aspects of the hardware system design, embedded face recognition software design and face recognition process, this paper studies the face recognition technology in Hi3515high performance audio/video processor embedded realization. Hardware part introduces the Hi3515processor based on face recognition system hardware environment, mainly including gravity sensing module, face acquisition module, Hi3515center processing module, access control module, etc. Embedded face recognition software design and implementation includes gravity sensing part, face recognition part and door control part. In the fields of computer vision, complete face recognition process includes face image acquisition, image preprocessing, face detection, feature extraction, recognition or confirmation, etc. In order to verify the reliability and real-time feature of an embedded face recognition access control system, the hardware and software design of the system using Hi3515microprocessor was proposed. The system was based on Hi3515which provided high-performance audio/video services, and studied the implementation of face recognition algorithm based on ORL and embedded systems by image preprocessing of Gabor transform and TV+L’model, face detection of skin color space (YCrCb color space) and feature extraction of the principal component analysis (PCA) method. This face recognition algorithm is implemented in the embedded access control system and then has controlled the whole embedded entrance guard system. The corresponding experimental results show that the algorithms in embedded platform are optimized. The problem of high-reliability and real-time of the embedded system is solved.