Traffic parameter estimation and control system based on machine vision

Abstract

With the rapid development of urbanization in the world, it has brought enormous pressure on urban traffic management and control such as traffic congestion. An excellent urban traffic management and control system consists of three critical aspects: obtaining traffic parameters, developing traffic control scheme, and evaluating traffic control scheme. Intersection signal timing is one of the most important parts in urban traffic control. This paper proposed an intersection signal timing system based on traffic video which consists of three parts: acquisition of video-based traffic parameters, calculation of traffic flow-based signal timing scheme, and evaluation of intersection signal timing scheme. In the first part, we used advanced techniques such as deep learning and image processing to obtain traffic parameters such as traffic flow, vehicle type, composition of different vehicle types, and speed of vehicles passing through a scene in a traffic video. In the second part, we calculated the signal timing scheme of the video at the traffic scene through the obtained traffic flow information with Webster method. In the third part, the detailed traffic parameters and signal timing scheme were input into the VISSIM software for traffic microscopic simulation, which was used to evaluate the signal timing scheme. The experimental results show that the accuracy of the detailed traffic flow information obtained by the proposed system can reach more than 90%, the accuracy of composition of different vehicle types can be achieved more than 98%, and the vehicle speed accuracy can reach more than 95%. Therefore, the system improves the reliability and adaptability of the whole signal timing network. At the same time, the simulation results show that the proposed system integrates the acquisition of traffic parameters and the calculation and evaluation of signal timing schemes, and provides a good solution for solving research problems and actual needs such as signal timing optimization.