Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (35): 153-156.DOI: 10.3778/j.issn.1002-8331.2009.35.047

• 图形、图像、模式识别 • Previous Articles     Next Articles

RSR moving least squares for non-uniform points cloud fitting

TIAN Qing,LIANG Rong-hua,MAO Jian-fei   

  1. College of Computer Science,Zhejiang University of Technology,Hangzhou 310023,China
  • Received:2009-07-24 Revised:2009-08-24 Online:2009-12-11 Published:2009-12-11
  • Contact: TIAN Qing


田 青,梁荣华,毛剑飞   

  1. 浙江工业大学 计算机学院,杭州 310023
  • 通讯作者: 田 青

Abstract: In order to handle the non-uniform sampling points fitting,this paper presents an adaptive adjustment of the radius of influence domain with the moving least squares(RSRMLS).Based on the moving least squares(MLS) fitting,this method can automatically adjust the size of radius for MLS according to the consistency of the sampled data points.This approach is compared with traditional MLS on the same sampling points,and this method has more approximate fitting results than the traditional MLS.

摘要: 针对传统的移动最小二乘法在非均匀分布的采样点集拟合中的不足,提出了影响域半径动态调整的移动最小二乘法(RSRMLS)。在传统移动最小二乘法(MLS)的基础上,根据拟合子区域采样点数据稀疏情况,该方法可自动调整MLS的半径区域大小。通过对相同数据点集的拟合比较,提出的RSRMLS拟合效果明显优于传统MLS。

CLC Number: