Design And Implementation Of Face Recognition System Based On Embedded System

Biometric identification has become a hot topic in the modern computer science. With the input of investment and technology, modern biological identification has become more mature, more accurate and fast. While face recognition on PC machine has been unable to meet these needs, the embedded microprocessor technology is constantly updated and developed. At present, some microprocessors computing speed has already left what PC needs far behind. Small embedded devices have the advantages of high integration and fast processing. Therefore, the biological recognition technology based on this platform has become the main trend for research.In this paper,a face recognition system based on embedded microprocessor is designed, the system use different peripheral circuit modules and S3C6410 chip to build the embedded development platform. After studying the traditional face detection, face enhancement and face recognition algorithm, in this paper algorithm optimized in Adaboost and PCA. And used on ARM11 system for the face detection and face recognition.A few main modules included in the system: man-machine interface module, which is based on the platform of QT; video capture module is used by v4l2 framework, and the video format conversion is implemented; face detection module based on Adaboost algorithm; face enhancement module which use image equalization method to enhance it; face recognition module which used Optimization Algorithm of PCA.By testing the whole system, the results show that the face recognition rate of the embedded system is more than 91%, which reaches the expected goal. It has comparatively higher practical value.