The Research Of ARM11-Based Wireless Video Surveillance System And Face Detection System

With the development of computer processing, embedded technology and other related electronic industries, video surveillance system has developed to embedded wireless video surveillance from simulation monitoring, which is widely used in various fields. Face detection is an important technology in the face of face information processing, and it has been widely used in the field of video surveillance, multimedia etc. In this paper, combined with the technology of embedded video surveillance and face detection, a system of wireless video surveillance based on ARM11 and a system of face detection is proposed. The main work is as follows:Firstly, the design scheme of whole system is proposed, and the design of hardware frame and software structure is completed. The work of cutting and transplantation about linux kernel and file system has been finished, which is based on the S3C6410 chip hardware platform of ARM11, The driver of required module is compiled and loaded to ensure the smooth running of the entire hardware. The cross-compiler environment is established on the host for building the development environment for the late program designed.Secondly, on the basis of in-depth study of different face detection algorithm, Ada Boost algorithm is used for face detection and the positive and negative sample database was established because of good real-time and high accuracy. In order to make the algorithm suitable for dynamic video face detection and train a classifier for this article, to further improve the detection accuracy, the collected dynamic video is preprocessed.Thirdly, the overall design scheme of the server is proposed. Using V4L2 interface and MMAP memory mapping to achieve the video image acquisition. Based on the H.264 standard, collected video data are compressed. Using the Socket network programming under Linux and based on UDP protocol to realize the wireless transmission of video data. Using Qt development tools to develop a PC-server interface to achieve the camera,store video and other functions. At last, Open CV library is callded to achieve the real-time video face detection.Finally, experimental results show that the system in the simple lighting and background conditions has the higher detection accuracy, strong real-time, easy installation and strong stability.