Multi-camera high-precision three-dimensional pose tracking method for large-view-field moving target

Abstract

The invention discloses a multi-camera high-precision three-dimensional pose tracking method for a large-view-field moving target. The method comprises the following steps: firstly, calibrating to obtain internal and external parameters of each camera, and solving a three-dimensional coordinate initial value of each angular point on a checkerboard; then, by taking a standard checkerboard as a constraint, carrying out space local bundle adjustment optimization on internal parameters of the camera; and carrying out bundle adjustment optimization on the three-dimensional coordinate initial values of the angular points on the checkerboard by taking the internal parameters of the camera subjected to spatial local optimization as constraints. According to the invention, the optimization of the internal parameters of the camera is not affected by the change of the position of the measured target, and the spatial positioning precision of the measured target can be improved.

Description

technical field [0001] The invention relates to the technical field of geometric measurement, in particular to a high-precision three-dimensional pose tracking method based on a multi-camera moving target with a large field of view. Background technique [0002] In industrial production, it is often necessary to achieve high-precision tracking of the three-dimensional pose of a moving target. In the aviation manufacturing industry, the large parts of the aircraft need to be tracked in real time when they are automatically docked and assembled, and the drill gun needs to be tracked in real time when the aircraft skin is drilled and riveted. In the light pen type coordinate measuring system, it is necessary to measure the pose of the light pen. In the complex surface measurement technology, a combination of a 3D scanning measurement robot and a visual tracking system can be used. At this time, the pose of the 3D scanning probe needs to be tracked in real time, and the accurac...

AI Technical Summary

Problems solved by technology

The internal parameters of the camera obtained by using the traditional camera calibration method are the global optimal solution, but not the local optimal solution at different spatial positions, so that the reconstruction accuracy of the system in different spatial positions is not optimal; therefore, the camera calibration obtained by using the traditional method Internal parameters need to be optimized locally;

[0005] (2) The initial coordinates of 3D reconstruction are calculated by the internal and external parameters of the camera, and there is error transmission. According to the traditional beam adjustment idea, it is unreasonable to simultaneously optimize the internal and external parameters of the camera and the initial coordinates of 3D reconstruction

Images