Calibration data collection method, calibration method and device for traffic camera

Abstract

The invention provides a calibration data collection method, calibration method and device for a traffic camera, and relates to the technical field of camera calibration. The method comprises the steps: obtaining a picture sequence of a plurality of vehicles in a preset vehicle model library from a traffic monitoring video; determining three-dimensional coordinates corresponding to a plurality ofkey points of each vehicle according to the preset vehicle model library; and determining a two-dimensional coordinate corresponding to each key point in the picture sequence according to the three-dimensional coordinate, and determining the three-dimensional coordinate of each key point and the corresponding two-dimensional coordinate as the traffic camera calibration data. According to the invention, the vehicle meeting the model in the preset vehicle model library is obtained as the calibration object through the monitoring image shot by the traffic camera; according to the camera calibration method and device, an extra calibration board and the like do not need to be arranged, the collection difficulty of the calibration data can be effectively reduced, the accuracy and reliability ofthe camera calibration data can be effectively improved, and the subsequent calibration efficiency and effectiveness of the camera are conveniently improved.

Description

technical field [0001] The present invention relates to the technical field of camera calibration, in particular to a method for acquiring calibration data of a traffic camera, a calibration method and a device. Background technique [0002] With the rapid development of modern computer technology, an emerging technology has gradually formed: machine vision technology, which mainly uses computers to simulate human visual functions to extract information from images of objective things, process and understand them, and finally use them in Practical detection, measurement and control and other fields. Intelligent monitoring is an important application field of machine vision. It uses computer-controlled cameras instead of people to carry out relevant monitoring. By analyzing and processing continuous video images, the effect of intelligent monitoring is finally realized. For example, vehicle monitoring in traffic scenarios has become an important part of modern intelligent se...

AI Technical Summary

Problems solved by technology

However, Zhang Zhengyou’s calibration method relies on a plane checkerboard calibration board with a specific known size. The distance between the surveillance camera and the ground is usually relatively large, so a larger checkerboard must be used. However, it is inconvenient and unsafe to place a large calibration board on the traffic road.

Using the vanishing point on the road to calibrate needs to determine a vanishing point in the direction of the lane line and another vanishing point perpendicular to the direction of the lane line, but the vanishing point perpendicular to the direction of the lane line is not easy to detect, and for more complex roadsides It is not easy to determine the vanishing point along the direction of the lane

Moreover, in the above-mentioned various calibration methods, it is necessary to estimate the initial value of the camera’s internal parameters, which is only applicable to specific conditions. For complex and changing traffic scenes, the internal parameters of the surveillance camera are usually not directly obtained or calculated.

Images