Font Size: a A A

Design Of Embedded Intelligent Video Surveillance System

Posted on:2010-10-26Degree:MasterType:Thesis
Country:ChinaCandidate:M LiFull Text:PDF
GTID:2178360302959089Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
With the rapid development of Internet and embedded technology, the research of embedded equipment with network function is becoming a hotspot field. On the other hand, how to ensure sensitive public places in all weather conditions, automatic, real-time monitoring are important points which are paid more attention by people in the world and intelligent video surveillance is the most useful way to solve this problem.The paper brings forward a design plan of low-cost embedded network interface equipment used in remote video surveillance system against medium low-end users.According to this project, these researches as follow have been done.(1)This system is built by embedded processor named ARM9 as the core of hard platform, which has USB port and Network module so that it can achieve data acquisition and real-time remote data transmission. The paper adopts Linux operation system as software platform which is free source and designed for embedded processors.(2)For remote surveillance functions, use the Internet as a delivery platform, the paper completes the design of video capture, nertwork transmission which is real-time, low-cost, long-distance, based on TCP/IP protocol, video display on LCD functions.(3)According to large-scale video surveillance system, a large number of video data management strategy NVR we improve on its function. Video data network synchronization and video matrix output function has been added. Net off synchronization to ensure interruption of the network is not the loss of essential video data, matrix output function allows users to more easily monitor the video viewing. (4)At the software design of intelligent video surveillance, motion detection baced on background subtraction and frame subtraction is used in order to get the profile of object moving, use star vector to get a moving object extraction eigenvector then human identification based on SVM is achieved.In this paper, according to intelligent video surveillance, based on the characteristics of users, we use embedded processor for image acquisition and processing, use internet which is wide-area, low-cost, real-time as the data transmission platform, enable this embedded network transmission system to realize a popular demand and low-cost implementation.
Keywords/Search Tags:Embedded, ARM9, V4L, Motion Detection, Support Vector Machine, Human identification
PDF Full Text Request
Related items