基于参数优化的多层次单元划分曲面重建

Parameter optimization based multi-level partition of unity surface reconstruction

为了提高现有多层次单元划分(MPU)曲面重建的稳定性,提出一种基于参数优化的MPU曲面重建算法。分析包围球半径的比例系数α和包围球内点云最小数目Nmin对曲面重建的影响。根据局部隐函数逼近的条件,对Nmin进行调整;同时,结合不同形态特征的点云模型对覆盖密度的要求,对α进行调整,使得α和Nmin达到一个最优组合,进而使得曲面重建算法更稳定。实验结果表明,该算法能快速准确地对不同点云模型进行参数选取,从而得到更理想的曲面重建效果。

In order to improve the stability of surface reconstruction with Multi-level Partition of Unity（MPU）,an algorithm of MPU based on parameter optimization was proposed.Firstly,the influences of proportion coefficient α of the bounding sphere radius and the minimal number Nmin of point clouds in bouding sphere were analyzed.Then,the Nminwas adjusted by the approximation of condition implicit function and the α was adjusted by the requests of coverage density in different models.Finally,the method of more stable surface reconstruction can be achieved by optimal combination selection of Nminand α.The experiments prove that the method proposed is more stable and accurate in different models.