Research And Design Of Intelligent Video Surveillance System Based On ARM

Video monitoring system is widely used in the production and living securitysystem, using the network technology, intelligent technology and improving theintelligent video monitoring system is a hotspot in the research of the current. My thesiscomplete an intelligent video monitoring system based on ARM which give a detaileddesign on hardware circuit and software flow charts. And also, it researches the targetdetecting and tracking algorithm of intelligent video monitoring system.In this system, i.MX27processor is the main controller with Linux real-timeoperating system running on it. Through use of digital camera, the system collects andcompresses video image and then transplants the intelligent algorithm which is writtenby software into the ARM920T kernel, so it can extracts the region of interest or targetin video image. The ARM920T kernel gets the target’s feature information feedback tothe i.MX27, adjusts video image acquisition and makes monitoring image transmitted toupper host.On the basis of the intelligent video monitoring system hardware circuit design,this paper completes the construction of the embedded Linux system softwaredevelopment environment. Then, it designs system software and application softwareon the Linux system platform and combines multithreading technology to ensure thesystem integrality and operating stability.This paper has introduced the technology of image preprocessing and imagepost-processing.To aimed at the motion target detection in static background, this paperdetermines a set of background difference method as the core of moving target detectionmethod, the algorithm gives a detail introduce in background modeling which is based on statistical average method, and analyses the method that repair error areas in theinitial modeling background. Simulation result shows that under certain conditions thisalgorithm can complete and timely detect the moving targets. Based on research movingtarget detection algorithm, this paper first introduces the tracking method based onimage matching and detailed studies the extraction of target centroid movement. Then,puts forward a method that tracking the target according to the position based on therelative velocity between background and moving target. Last, MATLAB simulationstudy shows that in the static background it can completely obtain the motioninformation of target trajectory and instantaneous velocity parameter.