Design And Implementation Of Vehicle Congestion Pre-warning System Based On Video

With the acceleration of urbanization,the continuous improvement of people’s living standards,the rapid development of transportation,people’s life has been greatly convenient,at the same time,the rapid growth of the number of motor vehicles,coupled with the continuous optimization and reconstruction of the traffic network,the traffic condition is deteriorating,traffic congestion is widespread,and traffic accidents occur constantly,which lead to personal safety,exhaust noise,traffic accidents environmental pollution and other social problems.In order to solve the problem of traffic congestion from the source,people began to use information communication technology and artificial intelligence control technology to realize the orderly operation of vehicles,roads and managers,and then intelligent transportation system(ITS)appeared.In this paper,from the current intelligent transportation system involved in vehicle detection and statistics technology as the research object,through the analysis and processing of traffic video,focus on the research and design of vehicle detection and congestion warning based on video.Firstly,this paper studies the traffic video image processing technology.The advantages and disadvantages of different video image processing technologies are compared and analyzed through experiments.In view of the environmental impact in the process of video image capture,this paper proposes video preprocessing and post-processing,which greatly improves the detection effect of video image.Secondly,this paper studies the vehicle detection technology.Aiming at the situation of moving vehicle’s ghosting blur in video image detection,this paper first adopts the target detection algorithm which combines visual background extractor(Vi Be)algorithm and Su BENSE algorithm.Vibe algorithm is used to get the foreground image of the video image,and then Su BENSE method is used to detect the adjacent foreground image.Then,through Kalman filter and image similarity,combined with the setting of virtual detection line,the coordinates of a single vehicle are segmented,and the vehicle information,accurate coordinates and traffic flow statistics are obtained by using the linked list.Finally,this paper studies the realization of vehicle congestion early warning by vehicle speed detection correction and vehicle space rate calculation based on the centroid of vehicle ROS.On this basis,the vehicle congestion early warning system is developed by using Visual C++ 6.0 and OPENCV2 image development package.The experimental results show that the vehicle congestion early warning system designed in this paper has high accuracy and reliability.