Research And Implementation Of Traffic Parameters Detection System Based On Video

With the rapid economic development of our society, the urban population is increasing, the number of urban vehicles is rapid increasing, the supply of urban roads has been far from satisfying the needs of the development of road traffic, all kinds of traffic problems have become increasingly prominent. Long-term practice of developed countries shows that approach of solving the growing traffic demand by relying solely on building more roads and expanding the network scale, has been difficult to adapt to the rapid development of modern society. However, dynamic information-based Intelligent Transportation System (ITS) is one of the most effective ways which are widely recognized for the improvement of urban road traffic conditions to "protect the safety, increase efficiency, improve the environment, energy conservation". With the rapid development of video technology, the implementation of traffic parameter detection system based on computer video detection technology has become possible. The computer video detection technology has become one of the key and hot issues in the study field of modern traffic engineering.The main work is divided into two parts:vehicle detection based on video and traffic parameter detection based on video.First, this paper Vehicle detection is the premise of the video traffic parameters detection. Complex background, diversity of moving targets, changes of weather and lighting and other factors affect the accuracy of vehicle detection in different degrees. To this end, we propose vehicle detection algorithms based on frame difference methods and feature methods. The method achieves robust detection of vehicle in different weather and lighting based on the difference in the frame and the edge, symmetry, shadows and other features of the vehicle.Secondly, according to the vehicle detection, this paper proposes calculation methods of video-based traffic flow, speed and other traffic parameters and a vehicle identification method based on camera calibration. This method realizes vehicle classification using camera calibration and calculating the actual width of the detected vehicle.Finally, using the video of the traffic in the real environment, the proposed vehicle detection algorithm and the traffic parameter detection algorithm are tested. The results show the effectiveness of vehicle detection and the accuracy of traffic parameter.