Dissertation > Transportation > Road transport > Technical management of traffic engineering and road transport > Computer applications in road transport and highway projects

Research on Key Techniques of Video Monitoring System Based on DSP

Author ShiJingJing
Tutor MaRongGui
School Chang'an University
Course Traffic Information Engineering \u0026 Control
Keywords Traffic Video Detection DSP Moving Target Detection BackgroundSubtraction
CLC U495
Type Master's thesis
Year 2013
Downloads 2
Quotes 0
Download Dissertation

For recent years, video-based traffic parameter detection has become one emergingfield of computer vision research. And it is an necessary part of the IntelligentTransportation System (ITS) and has been broad application prospects. Meanwhilevideo-based traffic parameter detection technique involves multi-disciplines such asimage-processing, computer vision, computer network, pattern recognition and so on. Ithas presented more significance in academic value and theoretical research.Firstly this paper begins with an overview about architecture, features, hardwarecomponents and software processes of video-based traffic parameter detection system andapplies the Digital Signal Processing (DSP) into the video traffic monitoring system, whichhas simplified the original complex circuit design greatly and increased flexibility of theoverall system design simultaneously.Secondly, before traffic parameters and events detection, this paper has done somepreprocessing works, for example, format conversion and storage of collected trafficvideo data, image coordinates determination, lane detection and segmentation, lanedistance calibration and target region identification in the traffic video. All these workscould improve the accuracy of results in traffic parameters and events detection obviously.Finally, for the vehicles movement detection in traffic video, several moving targetdetection algorithms have been discussed. Because the traffic video has its own featuressuch as fixed camera, strong real-time and so on, this paper gives an novel moving targetdetection method which is based on background subtraction and makes some improvementin background estimation and updating algorithm. The experimental results show that theproposed method has better applicability and versatility in detecting vehicles moving targetfrom the traffic video.

Related Dissertations
More Dissertations