The Real-time Detection System For Traffic Flow Base On Video And Image Process

With the advancement of urbanization,the problem of traffic congestion is serious increasingly.Traffic information is an intuitive reflection of road conditions.Timely grasping of traffic information is of great significance to the reasonable allocation of traffic resources and alleviating congestion.It also provides data support for the development of intelligent transportation.To obtain road traffic information in real time,this paper designs an FPGA-based vehicle flow and vehicle queue length detection system.The main research contents of this paper are as follows:Firstly,the system requirements is analyzed,that introduces the system’s overall structure design and software implementation process,and briefly introduces the development platform and simulation tools.Secondly,image preprocessing,vehicle presence detection algorithms,vehicle flow detection algorithms,and vehicle queue length detection algorithms are studied in this paper.Regarding the problem that the traditional Sobel edge detection threshold selection algorithm is not conducive to hardware implementation,an improved Sobel edge detection algorithm based on the overall and local brightness is given.For the traffic flow statistics algorithm,on the basis of the traditional detection window-based statistics algorithm,a data extraction algorithm and a frame number correction algorithm are added to improve the detection accuracy.The vehicle queue length extraction algorithm is designed,and the algorithm is optimized according to the characteristics of FPGA hardware.Finally,the involved algorithms are elaborated,and Matlab is used for algorithm verification.Thirdly,ISE14.7 software is used to implement each module,and Modelsim software is used to simulate and verify each module,verifying the correctness of the traffic statistics module,and complete the main peripheral hardware selection and circuit design.Finally,the vehicle queue length detection module was tested on the board.The experiment proved that the system’s queue length detection result was more accurate.The traffic flow detection platform designed in this paper is a complete hardware property,with a high operating rate.It can obtain road traffic information in real-time and provide a reference for road traffic resource optimization and traffic light signal control,so it has specific practical significance.Figure [64] table [4] reference [80]...