三维重建中的线结构光标定

Calibration of Line-Structured Light in 3D Reconstruction

针对现有光平面标定的不足,改进一种光平面标定法。利用改进过的Steger中心提取算法,对投射在不同高度或者角度的棋盘格空白处的2条及以上的光条进行亚像素中心提取,然后利用张正友相机标定法中的棋盘格世界坐标zw=0的约束条件,计算出光条中心点的相机坐标,再利用最小二乘法拟合成光平面,并采用了与传统直射式三角法不同的方法求被测物的高度。针对光平面标定精度没有明确标准,提出一种新的评价方式。试验结果表明,该算法的精度和鲁棒性都不错。对于三种高度不同的物块,均方根误差最大是0.063 mm,最小为0.023 mm。

In view of the shortcomings of the existing light plane calibration,a method of calibrating line-structured light is improved.Firstly,this method uses the improved Steger algorithm to extract the subpixel center of 2 or more laser stripes projected at different heights or angles of the checkerboard calibration board.Secondly,using the constraint of the checkerboard grid world coordinates=O in Zhang Zhengyou's camera calibration method,the camera coordinates of the center points of the laser stripes are calculated,and then the ordinary least squares is used to calculate the light plane.A different method from the traditional direct-beam trigonometric method is used to find the height of the object and a new evaluation method is proposed because of no clear standard for the accuracy of the light plane calibration.Experimental results show that the accuracy and robustness of the algorithm are good.For the three blocks with different heights,the maximum of root mean square error is 0.063 mm and the minimum is 0.023 mm.