Research And Implementation Of The Hybird Verification Gateway System Based On ZYNQ

The rapid development of science and technology makes life in a kind of equipment to the intelligent,traditional entrance guard system also gradually said goodbye to the single card or password identification,and the use of human biological information as the recognition feature.Face image processing technology is a typical recognition of human biological information,and embedded technology can customize the special system needed by users according to different application scenarios.This paper will study the accuracy and real time of embedded face image processing system.In order to solve the traditional single-core embedded devices in the embedded image processing the problems of insufficient resources,etc,this paper puts forward using ZYNQ architecture for the research and development of the embedded image processing system,using the ZYNQ peculiar multi-core structure of software and hardware,and the image processing algorithm respectively on the processor software and hardware circuit,so as to realize the hardware and software work together.In this paper,Xilinx’s ZedBoard development board is the main hardware development platform.The ZYNQ chip programmable logic part of the implementation of face image preprocessing algorithms and the parallelization of hardware acceleration,in the ARM processor section of the ZYNQ chip by combining OpenCV to complete the function of face recognition,and proposed for face recognition based on principal component analysis decision tree improve system performance.The integrated fingerprint recognition and face recognition system is constructed by combining the fingerprint reader and relay module.According to the experimental results,it can be seen that,based on the mixed verification system of ZYNQ platform,it also speeds up the operation time of the algorithm and improves the system’s response speed while ensuring the accuracy.