基于角度投影法的稀疏点云圆柱面拟合

Cylindrical Fitting of Sparse Point Clouds Based on Angular Projection

在孔轴类零件测量中,针对关节臂式坐标测量机获得稀疏点云去进行圆柱拟合时,存在多义性的问题,提出了一种角度投影查找圆柱拟合初值的方法。首先计算点云重心与测量点的角度来确定投影范围,然后在该范围上进行点云投影,通过投影的圆度计算来查找出最佳的投影方向,将该方向作为圆柱的轴向初值代入误差方程进行迭代计算,拟合最佳圆柱。实验结果表明该算法拟合收敛性好、精度高。

In the measurement of hole shaft parts,when obtaining a sparse point cloud for cylindrical fitting by the articular arm coordinate measuring machine,there is a problem of ambiguity,and a method of angle projection to find the initial value of the cylindrical fitting is proposed,first calculate the angle of the point cloud center of gravity and the measurement point to determine the projection range,and then perform the point cloud projection on the range,find the best projection direction through the roundness calculation of the projection,and use the direction as the axial initial value of the cylinder to replace the error equation for iterative calculation,and fit the best cylinder.Experimental results show that the algorithm has good fitting convergence and high accuracy.