Face recognition algorithms for embedded entrance guard system

Jun Li 

Institute of Computer Science, Guangdong Polytechnic Normal University, 510665, Guangzhou, P.R. China

Biometric identification technology is used to identify individuals based on their unique physiologica1 characteristics under sampling and measurement. The hardware platform of our entrance guard system is based on SAMSUNG S3C6410 embedded development board with ARM11 processor, and we adopt ordinary camera as facial image capture device, 7-inch touch-screen LCD display as an input, electronic door locks as executing components by the onboard port control. The software platform is based on Linux operating system to design and optimize face recognition algorithm, which is combined with local binary pattern (LBP) algorithm, principal component analysis (PCA) algorithm, and ridge regression algorithm. Also, the face recognition system is divided into two parts, namely, a face database training part and a face recognition part. Precisely, the face database training part acquires face information through the processes of face detection, feature extraction and stores the face information to serve as the detection basis of the face recognition part; then the face recognition part is connected with the entrance guard system and sends out an instruction for executing related action to the entrance guard system according to the face recognition result. Specially, the embedded face recognition entrance guard system is contactless and easy to collect the facial database, less in power consumption, low in cost, and has easy installation and stable performance.