Abstract Registration of point sets is done by finding a rotation and translation that
点集的抽象配准是通过寻找旋转和平移来完成的
produces a best fit between a set of data points and a set of model points. We use
在一组数据点和一组模型点之间生成最佳拟合。我们使用
robust M-estimation techniques to limit the influence of outliers, more specifically
更具体地说,是限制异常值影响的稳健M估计技术
a modified version of the iterative closest point algorithm where we use iteratively
迭代最近点算法的一个改进版本,其中我们迭代使用
re-weighed least squares to incorporate the robustness. We prove convergence with
重新加权最小二乘法以合并稳健性。我们证明了与
respect to the value of the objective function for this algorithm. A comparison is also
关于这个算法的目标函数的值。此外,还进行了比较
done of different criterion functions to figure out their abilities to do appropriate point
做了不同的标准函数,找出自己做适当点的能力
set fits, when the sets of data points contains outliers. The robust methods prove to be
集合拟合,当数据点集合包含异常值时。鲁棒性方法被证明是有效的
superior to least squares minimization in this setting.
在这种情况下优于最小二乘法最小化。