基于单特征点的手眼系统摄像机标定

基于机器人的相对运动，提出了一种基于单特征点的手一眼系统摄像机内、外参数的标定方法。在摄像机视场中任意选择一个特征点，使机器人末端在保持姿态不变的条件下运动，相当于机器人不动但具有多个特征点。利用这些点可计算出摄像机的内参数和相对于特征点的外参数。大幅度改变机器人的姿态，可标定出特征点在机器人基坐标系下的位置，进而得到相对于机器人末端的外参数。实验结果验证了该方法的有效性。

A Simple Camera Calibration Method for Robot's Hand-eye System

A kind of self calibration method for hand eye system based on single point and robot's relative movement is proposed in this paper. The point, which is taken as the benchmark, can be arbitrarily selected in the view field of the camera. When the robot's end effector moves with the constant pose, the benchmark corresponds to many feature points with the unvaried end effector position. The intrinsic parameters and the extrinsic ones of the camera, relative to the benchmark, can be obtained with the feature points. With the large range adjustment of the pose of the robot's end effector, the position of the benchmark point can be calculated in the world coordinates. Then the extrinsic parameters of the camera, relative to the end effector, can be computed. Its effectiveness was verified by the experiment results.