Low Cost Portable Embedded Face Recognition System Based on ARM9 Processor

Human face recognition is a widely used biological recognition technology, and a core issue of safeguard. The system is based on MagicARM2440 development board, including transplantation of Linux operating system, the development of drivers, detecting face by using face class Haar feature, and then recognizing face by using Gabor features. In the paper, a face detection and recognition system (FDRS) based on video sequences and still image is proposed. It uses the AdaBoost algorithm to detect human face in the image or frame, adopts Gabor wavelet Transforms for feature extraction and recognition in face image. The related technologies are firstly outlined. Then, the system requirements and UML use case diagram are described. In addition, the paper mainly introduces the design solution and key procedures. The FDRS's source-code is built in C++ and Intel Open Source Computer Vision Library (OpenCV). Finally Haar Transform based face recognition system implemented on ARM9 processor based hardware system. The system uses USB based camera for image acquisition and operated with Linux operating system.