Research And Experiment On Moving Vehicle Detection And Tracking Methods

The detection and tracking for moving objects from the video of traffic monitoring is one of the research subject in intelligent transportation systems. According to the campaigns and analysis of road vehicles by traffic monitoring video, its easy to get the traffic parameters in time, such as the speed of vehicle operating, traffic flow and highway capacity. In practical road traffic, the veracity about the detection and tracking for moving objection is easy to influenced by environment. Changing about the illumination and self scale also can influence its veracity. So the veracity about the detection and tracking for moving objects is important to conclusions of highway traffic movements in surveillance video.Based on test and analyzing about vehicle detection and algorithm and vehicle tracking algorithm, combing to the characteristic about ViBe and Inter-frame difference method,and using the characteristic for particle filter, this paper put forward a effective arithmetic about moving vehicle’s detecting and tracking. The details are as follows:l.This paper analyzes common algorithm about detection moving objects, including it’s operational principles and applied ranges. This paper use the characteristic to remove the "Ghosting" scene about Inter-frame difference method, changes the ViBe’s disadvantage about "Ghosting" scene. This way also can improve detection efficiency, reduce the rate of false insurance, and get better detection results.2.This paper classify and compared the tracking algorithm of moving object, introducing the particle filter algorithm’s fundamentals and flow of work. When moving vehicles are blocked, its easy to migrate. Based on particle filter algorithm, this paper change this scene, the paper divides the tracking windows into equal sub windows, uses color histogram distance to judge the occlusion, and compares the traditional particle filter algorithm and Mean-Shift. Experiments show that the accuracy of the algorithm is significantly improved.3.In this paper, the moving target detection algorithm and tracking algorithm is proposed to detect vehicles in the traffic video series, with artificial statistical results, and other of motor vehicles in traffic video detection and tracking method, comparing the results obtained, analysis the advantages and shortcomings of the method.