The Video Image Acquisition And Processing System Based On ARM

With the high-speed development of the embedded technology,image processing technology and network technology,intelligent devices have been applied to every fields of the world in recent years.Intelligent monitoring system is an important application and is used in warehouses,housing,transport,telemedicine,pest control of agriculture and forestry,disaster monitoring and other fields.Embedded system,with the advantages of low-power consumption,small-sized volume and low cost,is always applied to intelligent monitoring system.At present,bonded with embedded technology,digital image processing technology and network technology,designing a high practical,real-time and stable motoring system at a low cost has great significance.Firstly,the article analyzes the development of the technology of video image processing which is based on the embedded system in-depth.The experimental platform,which is used to realize the functions of image's taking,display and save,is built on ARM developing board.An interface is built to control the entire system with the PC-based hypertext markup language(HTML).Meanwhile,Human face recognition&marking is realized via Visual C++ software.The system uses S3C2440A,which is produced by Samsung,as processor,and selects Linux as the embedded system for its high stability.The video,which are collected by a V4L interface-based program,are transmitted through Transmission Control Protocol(TCP)-based network-socket.The functions of image capture,display and save as well as moving object monitoring are realized too.For the expediency of manipulation,an interactive control page,which is based on B/S mode and PC browser,is built:After setup the Web server Boa and the CGI program,the ARM platform can interact with PC's browser according to HTTP protocol,and can provide interfaces for other applications.The VC,based on DIB,completes the program which can load,preprocess and mark the collected image,realizes the preprocessing of image greying,enhancement treatment and binary image,and obtains equilibrium output,and finally identifies the human face and marks its region.The algorithm and running results of the system are presented.Moreover,the results reached through testing experiment prove it's real-time and accuracy.By combining the image processing technology,the embedded technology and web technology,this article explores and develops an image collecting and processing system based on ARM embedded platform.Its testing results are positive,which prove the system's excellent stability and the system-designed goals are achieved.What's more,the proposed system has practicality.