Font Size: a A A

The Research And Implementation Of Monitoring Center Of Bus Multi-channel Video Surveillance System

Posted on:2012-04-22Degree:MasterType:Thesis
Country:ChinaCandidate:T BaiFull Text:PDF
GTID:2218330338463052Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
With rapid development of public transport, a more intelligent bus video monitoring system of multi-channel is required. Designing monitoring center of multi-channel bus monitoring system is significant. The center has many functions, such as multi-channel show at same time, multi-channel record at same time,the display of the recording material and the intelligent video algorithm.Bus multi-channel monitoring system of bus terminal is DM365 DVR, the monitoring center using MFC programming method to realize. This paper firstly describes bus video monitoring system of multi-channel monitoring center function structure, then emphatically introduces the monitoring center based on PC control software realization. Separately studied the monitoring center login bus terminal, streaming video transmission, decoding, display, control, and uses some key techniques, the decoding part using FFMpeg techniques, display part used DirectDraw technology, network transmission part using the Windows Socket programming techniques, the basic functions of general monitoring system are realized, such as multiple real-time monitoring, multi-channel storage, played h.264 video files, control the bus terminal etc.In addition, puts forward a people counting algorithm with vertical perspective, which combined body head information detection and foreground target tracking, among which the head count information detection using an improved method of detection of circular random Hough transform, using the movement target tracking characteristic parameters matching method.Finally, the monitoring center software and the toll performance of the algorithm are tested, and the result shows that the monitoring center software decoding and display effect is fluent, the people counting algorithm can accurately count the number of persons at video monitoring scene.
Keywords/Search Tags:Bus Video Surveillance System, Surveillance Center, Video Decode, People Counting, Hough Transform
PDF Full Text Request
Related items