Font Size: a A A

Research And Design Of A Intelligent Video Surveillance System Based On Embedded Hardware Platform

Posted on:2011-05-17Degree:MasterType:Thesis
Country:ChinaCandidate:R YanFull Text:PDF
GTID:2178360302493762Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
With the constant development of video surveillance system both in the theory and technology, it has been widely used by people in various places. Intelligent video surveillance system is the main technology development direction of video surveillance system and its excellent performance attracts many users, nowadays many scientists and engineers worldwide spend lot of their time into this technology's research and development, and they already have made a lot of achievements. Intelligent video surveillance system is an innovation of video surveillance technology. Its biggest innovation is introducing the machine intelligent into video surveillance research and free people from the boring monitors. Intelligent video surveillance system can reduces the possibility of leakage alarm appeared result in the factors of people, so it has great social and economic value.This paper firstly gives a brief introduction to the development of video monitoring system, as well as the intelligent video surveillance system research focus and existing problems, a typical structure diagram of the intelligent video surveillance system is also shown in this paper; Secondly it analysis of various common target detection algorithms which can be used in the system, after that we chose background subtraction as this article's detection algorithm; Finally this paper develop a actual intelligent video surveillance system based on the hardware development platform provided by our company partner, the system performance test results are given in the last in the paper. This article mainly includes the following aspects:(1)When background subtraction method is used to detect moving objects, illumination changes can easily impact the detection. In order to deal with this problem, a novel algorithm which synthesizes the methods of background subtraction and adjacent-frame difference is proposed.(2)This article analyze the system requirements of the project then determine the software structure, it also gives a proper module partition and load distribution for the detection algorithm based on the hardware platform. (3)This article development an actual intelligent video surveillance instrument based on ADSP-BF561 kernel, the surveillance instrument and the intelligent video surveillance management software running in a PC constitute a complete intelligent video surveillance system.(4)ln order to test the Intelligent video surveillance system we built a test platform base on the current conditions in the laboratory, then target detection module, target tracking module, target classification module are tested in this paper.Intelligent video surveillance developed by this article can run basic intelligence analysis, such as suspicious target detection, virtual line detection, virtual alert area detection, due to the project budget and time limit we can't realize smoke detection, and other advanced function, we think after the completion of the entire system can get huge economic and social benefits.
Keywords/Search Tags:Intelligent video surveillance, Moving Target detection, Target recognition, The Gaussian mixture model, ADSP-BF561
PDF Full Text Request
Related items