With the continuous development of intelligent information processing technology,face recognition has become the major research of identity authentication,and was widely used in people's daily life.The traditional face recognition system based on the computer platform is big and used inflexible,but Embedded ARM processor has the characteristics of small size,low power consumption,high performance and low cost.The research and development of face recognition system based on ARM platform have great value of application.In this paper a stable and accurate embedded face recognition system is designed and implemented.Firstly,an embedded platform based on the Samsung Exynos 4412 Crotex-A9 development board is established.Then the embedded Linux operating system,Qt graphical interface and OpenCV computer vision library are migrated to the platform.Secondly,the face detection based on Adaboost algorithm is deeply analyzed and realized.The face recognition methods based on Eigenfaces,Fisherfaces and LBPH are respectively realized and compared.By further studying the face image preprocessing method,the recognition performance is further improved.Finally,a complete embedded face recognition system was designed in this paper to implement the function of USB camera video acquisition and display and face recognition.Then the four functional modules of the system was tested,including the establishment of face database,the training of signature database,the import of configuration files and face recognition.The system identification rate,running speed and other performance indicators are tested and analyzed in this paper.The experimental results show that the face recognition system has a complete function,a friendly interface,convenient use and efficient recognition. |