A video-based traffic parameter measurement method

Abstract

The invention relates to a traffic parameter measuring method based on videos. A camera is installed above the front of a measuring road junction; a reference entity a, a reference entity b, a reference entity c and a reference entity d are selected on a road, the actual distances between all line segments are measured, actual data of all line segments are input into a video analysis module, and the mapping relation between a system image coordinate system and a world coordinate system is determined so that a system can conveniently measure and calculate traffic parameters like traffic flow and vehicle speeds. Due to the adoption of a whole set of reasonable and effective camera coordinate positioning, vehicle tracking and vehicle counting methods, the traffic parameter measuring method based on the videos is wide in measuring range, reasonable and scientific in measuring reference point setting, and accurate and fast in vehicle track judgment; due to the adoption of the method, equipment can be conveniently installed, measuring can be easily, fast and reliably carried out, and the acquired traffic parameters are accurate and are instant; the method has outstanding technical features in the field of video traffic parameter detection, and provides large-capacity traffic information and reliable regulation violation evidence for fair and reasonable traffic monitoring.

Description

technical field [0001] The invention relates to the field of traffic monitoring using image recognition, in particular to a method for measuring traffic parameters based on video Background technique [0002] In the known technology, traffic road parameter detection includes manual detection, buried geographical induction coil, geomagnetic detection, pressure measurement, infrared measurement, ultrasonic detector, laser speedometer, GPS measurement, video detection and other methods. Due to video collection and measurement, information It has become the main direction of today's traffic measurement, and it is one of the important topics in the application of computer vision and image pattern recognition technology in the field of intelligent transportation. , It is an important link to realize the intelligentization of traffic management. [0003] Among today's traffic video measurement methods, the US Patent "Video-Based Vehicle Detection and Tracking Method Using Spatiote...

AI Technical Summary

Problems solved by technology

"However, its detection range is small, and it needs to be calibrated by external objects before it can be measured; when analyzing and judging the vehicle trajectory, it needs to be judged according to the tracking of the vehicle's external feature points; when calculating the average speed of the vehicle, it is based on the longest characteristic of the vehicle The extraction of the chain is calculated; these shortcomings make the measurement method complex, the change and uncertainty of the measurement intermediary reference object are large, which greatly reduces the measurement accuracy and reliability

Images