激光测距仪与相机信息融合过程中位姿标定方法

激光测距仪与视觉图像的融合研究在最近的工业机器人等应用领域中受到越来越多的关注,其中对这两种仪器的位姿标定有利于对二者信息的有效融合。以单点激光测距仪和相机的位姿融合为主要研究内容,提出两种确定两者相对位置和方向的标定方法,在建立光斑相机坐标的基础上分别以可见光斑和不可见光斑为研究内容,分别通过光斑的坐标和平面约束对两种情况建立标定方程,并结合实验验证两种标定方法的有效性和可靠性,实验结果表明:在有限次测量次数的情况下标定结果能够使重投影误差达到3个像素以内,而且运算时间大大减小。

Calibration method of laser range finder and camera in information fusion process

The fusion of laser range finder (LRF) and visual image has received more and more attention in recent applications such as industrial robots. The calibration of the poses of these two instruments was conducive to the effective fusion of the two information. The single-point laser range finder and the camera's pose fusion were the main research contents. Two calibration methods for determining the relative position and direction of the two were proposed. Based on the coordinates of the spot camera, the spot was visible and the spot was invisible as the research content. The research content establishes the calibration equations for the two cases by the coordinates of the spot and the plane constraint respectively, and the validity and reliability of the two calibration methods were verified by experiments. The experimental results show that the calibration results can be made in the case of a limited number of measurements. The re-projection error is within 3 pixels, and the computation time is greatly reduced.