Based On Vehicle Tracking Trajectory In Parking And Retrograde Detection Research

With the rapid development of the national economy, the intelligent transportation is becoming more and more important. As an important part of intelligent transportation, traffic incident detection to ease traffic pressure, reduce the number of traffic accidents has a very important role. Traffic incident detection based on video technology, due to its significant advantages(simple installation, wide coverage, intuitive), has increasingly become the focus of academic research in various countries. On the basis previous research, this paper mainly studies the vehicle detection and vehicle tracking, and draw the trajectory according to the coordinates of the center of mass, respectively, then the use of vehicle tracking trajectory for il egal parking detection and illegal retrograde detection.Firstly, in moving vehicles detection, by comparing the background extraction algorithms, then using multi frame averaging method to extract video background. Analysis inter- frame difference, optical flow method, background subtraction method, considering the actual testing result, using color background subtraction method of vehicle detection, after the image binary segmentation, morphology processing and mass filling, using HSV color space of vehicle shadow removal, the sequence enables vehicle target to complete. Secondly, in terms of moving vehicle tracking, on the basis of common tracking methods are introduced, using kalman filter and vehicle mass center features, to forecast the vehicle characteristic value of the next moment, after the vehicle search matching, realize the tracking. By analyzing the centroid coordinates, to extract the vehicle tracking trajectory and real- time display on the monitor. Based on the above research, detection of parking and retrograde identification.For illegal parking, using centroid distance variation and velocity change is judged, when the distance between the centroid and the vehicle speed is less than a certain threshold value, combined with the specific parking time determination. For retrograde parking, when the vehicle is traveling in the opposite direction and the roads in a predetermined direction, Y-axis coordinate of the centroid of the vehicle changes to be determined. Finally, using the test video on the parking and retrograde detection, has obtained the good effect.