Research On Vehicle Trajectory Recognition Method Based On Video Detection

Road traffic accident has become a social public security problem.As one of the important causes of traffic accidents,vehicle illegal lane change has been paid more and more attention by the government and the community.Highway traffic flow is large,and the speed is high.At this time the random lane change will often cause great traffic safety risks,causing traffic accidents.Therefore,how to effectively detect and identify the illegal lane changing behavior of the vehicle,to prevent and reduce the loss caused by traffic accidents,has become an urgent task for the traffic management department.At present,China’s vehicle illegal road behavior identification is mostly through the traffic police on-site observation or monitoring of video for manual interpretation.This method is time-consuming and laborious and low detection accuracy.In order to solve these problems,this paper proposes a vehicle lane change track recognition method based on video detection,which can realize the identification of random lane change behavior in the unsupervised state.The identification model can judge the behavior of the vehicle lane change directly through video detection without manual interpretation.To provide evidence for the fight against not driving the provisions of the lane,free to change the lane and the highway emergency lane occupancy and other illegal acts.In this paper,we mainly study the behavior of expressway vehicle lane,take the video processing technology as the core,use the moving target detection,vehicle identification and tracking method to extract the moving vehicles of video,get their trajectory and analyze the movement characteristics,put forward a trajectory identification method based on video detection.The main work is as follows:(1)The related algorithms of video image preprocessing are studied,and the preprocessing techniques such as image gray scale transformation,smoothing filtering,image enhancement and image binarization are compared and analyzed.The experimental results are verified by experiments.Lay the foundation for follow-up testing.(2)The background reduction method is used to detect the moving vehicle.According to the characteristics of expressway traffic flow,the Gaussian hybrid background modeling algorithm is selected to extract the background,and the edge detection operator is used to calculate the image to obtain the edge contour of the moving target and the center of mass,then use the shadow extraction technology to remove the shadow in the image.An improved sparse subspace clustering algorithm is used to classify the moving objects.The experiment proves that the algorithm can accurately classify the targets in the scene.(3)The Hough transform is used to detect the lane mark and the target vehicle is tracked using the improved centroid tracking algorithm and template matching algorithm.The trajectory of the moving vehicle is extracted and the distance between the existing vehicle trajectory and the lane line is calculated.The vehicle lane is defined,to determine whether the vehicle illegal road behavior through the vehicle trajectory and the distance between the lane of the variance.Experiments show that this method can effectively detect the trajectory of vehicle lane,and the detection accuracy and efficiency are higher.