基于主动视觉的结构光手眼系统自标定方法

为实现结构光手眼系统的标定,提出一种基于主动视觉的结构光手眼系统自标定技术。该标定技术无需使用特制靶标,只需场景中三个特征点,通过控制机器人进行四次线性无关的平移运动和两次带旋转运动,即可实现结构光手眼系统的标定。四次非线性相关的平移运动标定摄像机内参数和手眼矩阵旋转部分,两次带旋转标定手眼矩阵的平移部分,两次带旋转运动中提取的两激光条,结合特征点所在平面信息,标定光平面方程。实验结果表明,三维数据测量精度可达到±1.32mm,平面特征点间的长度测量误差为±0.73 mm;该标定方法简单,特征选取容易,对结构光手眼系统的实际工业现场使用有重要意义。

A self-calibration approach to structured light eye-in-hand system based on active vision

To achieve structured light eye-in-hand system calibration, a new self-calibration approach based on active vision is proposed. This calibration algorithm can be achieved with only three feature points in the scene without a specific target by precisely controlling the robot four linearly independent pure translational motions and two random motions with revolving. The four linearly independent pure translational motions calibrated the camera intrinsic parameters and hand-eye rotation matrix, the two random motions with revolving calibrated the hand-eye matrix"s translation section, and combined with the plane information of the feature points, two images with laser calibrated the light plane equation. Experimental results show that the 3-dimensional data measurement results can be achieved ± 1.32mm, and the error of length measurement of the plane feature points is ± 0.73mm. The calibration approach with easy feature selection is simple; and has significance to the actual use of the structure light eye-in-hand system in the field of industrial.