Design And Implementation Of Video Based Vehicle Detection And Tracking System

Vehicle detection and tracking in video images is a key technology and main research direction in the field of intelligent transportation system.It has great research value and application value.The vehicle detection module can detect vehicles in the video sequence frame in real time,obtain target parameters,and pass the information to the vehicle tracking module,which matches the vehicle and tracking to obtain the trajectory of the vehicle.In practical applications,when the target vehicle is blocked or the image size on the image changes significantly with its movement,it will cause the loss of tracking and cause the tracking of the moving target to fail.How to improve the performance of tracking is the primary problem that needs to be solved.In response to the above problems,this paper proposes a Kalman tracking algorithm based on video image processing.This video-based vehicle detection and tracking system from the perspectives of software requirements analysis,software design,software implementation,and testing.The specific work of the paper is as follows:First,combined with video-based moving target detection,recognition and tracking technology,a hybrid Gaussian model,Kalman filter technology,and target feature recognition technology are used to study video-based vehicle detection and tracking methods.Form a system of detection and tracking.Second,analyze the realization goals of the video-based vehicle detection and tracking system,and analyze the system's data,software and other requirements,detail the working principle of the vehicle detection and tracking system,and design the overall architecture of the system.In the VS2015 integrated development environment,the OpenCV computer vision library,combined with the open source Kalman tracking algorithm,was used to implement a video-based vehicle detection and tracking system.The system is mainly composed of vehicle detection interface and vehicle tracking interface,which can realize the basic functions of displaying vehicle detection and vehicle tracking.Third,test the video-based vehicle detection and tracking system.The test environment and test plan of the system are given.The video taken by the mobile phone is used to test the vehicle detection module and the vehicle tracking module respectively,and the effect of the system on tracking performance is analyzed by certain parameters.The results show that the vehicle detection and tracking system in this paper can accurately detect the vehicles in the video frame sequence and track them in real time.In the case of severe occlusion,the algorithm in this paper can still detect the location of the target vehicle,ensuring the continuity of the tracking process,improving the performance of long-term tracking,and verifying the robustness and effectiveness of the algorithm.