Vision-based density traffic vehicle counting and traffic flow calculation method and system

Abstract

The invention discloses a vision-based density traffic vehicle counting and traffic flow calculation method and system. The vision-based density traffic vehicle counting and traffic flow calculation method comprises the steps of zooming an acquired continuous frame image to obtain a pyramid characteristic map, inputting the pyramid characteristic map to a trained pyramid-YOLO network to detect vehicle targets with different scales so as to obtain a boundary frame with the vehicle targets; presetting a line-crossing probability function, judging probability of each target vehicle passing through a counting line in an image with the vehicle targets, and sieving a traced vehicle according to a probability value; performing tracking track processing on the traced vehicle according to a limitation-based multi-target tracing algorithm, and counting the vehicles passing through the counting line according to an obtained track set; and calculating to obtain vehicle flow traffic volume, speed and density according to the obtained vehicle counting result. By the vision-based density traffic vehicle counting and traffic flow calculation method, the problem of a dense traffic scene is solved,a system on the basis of vehicle detection, tracing, counting and parameter estimation is proposed, vehicle detection is accurately and rapidly performed, the dense vehicle is accurately traced, and flow estimation is achieved.

Description

Technical field [0001] The present disclosure relates to the technical field of road traffic vehicle flow detection, and in particular to a visual-based density traffic vehicle counting and traffic flow calculation method and system. Background technique [0002] The statements in this section merely provide background information related to the present disclosure, and do not necessarily constitute prior art. [0003] Vision-based traffic flow parameter estimation, especially for dense traffic scenes, is difficult to accurately count due to the fact that vehicles are blocked, small in size, and denser. Current methods mainly use detection and tracking methods to count vehicles, and few of them further estimate traffic flow parameters in dense traffic scenes. [0004] Traffic flow parameters (including volume, speed, and density) play an important role in traffic management and intelligent transportation systems. Estimating these parameters usually requires accurate vehicle counting....

AI Technical Summary

Problems solved by technology

For the three traffic parameters (volume, speed, density), most of the existing calculation methods can only directly estimate one or two parameters, and the remaining parameters are not estimated or estimated indirectly based on the direct relationship between the three parameters.

Images