The Research Of Moving Object Tacking Technology

Video surveillance systems have been widely used in many fields at recent years,but they are limited to record the scenes.An intelligent surveillance system that can automatically analyze the video and alert for possible anomalous behaviors would be greatly appreciated.An intelligent surveillance system aims to automate object detection,recognition,tracking and activity under standing in dynamic scenes that has the very great potential value in the enhancement of social security.Detection and tracking of moving object in image sequence is also the most complicated direction of computer vision and digital image processing techniques. It is an ample project that including video collection,digital image processing and application development.This paper mainly aims to research on moving object detection and tracking.In motion target detection,the moving objects detection algorithms usually used in stationary scene be discussed.In motion target tracking,the popular tracking methods and target tracking algorithm are detailed studied,then effective improved algorithms are presented considering existing problems.In the application of motion target detection and motion tracking in intelligent transportation,this paper suggests the system structure design procedure,realizes the Camshift algorithm using VisualC++6.0 and OpenCv(Open Source.Computer Vision Library),and gives program code and experiment results.OpenCv(Open Source Computer Vision Library) is a library for digital image processing and computer vision.It is developed by the visual interactivity group in Intel's microprocessor research Lab with C++ language.It can be used in window system and Linux system.The library is open source and can be free download from Intel's Website.OpenCv provides various functions to extract frame from image sequence and video source and many of the standard image processing algorithms, these functions can be directly used in the video program for specific development projects. At the end of this paper,we use this system doing a lot of experiment under various contexts,and the experiments results are well amid.The results demonstrate that the system can do automatic motion detection and tracking with high real-time performance and moderate robustness.Since we develop this system under windows platform,how to migrate to other systems or embedded platform and further improve the versatility and robustness has become the focus future research work.