面向非匀点云拟合的RSR移动最小二乘法  

RSR moving least squares for non-uniform points cloud fitting

作　　者：田青[1] 梁荣华[1] 毛剑飞[1] 

机构地区：[1]浙江工业大学计算机学院,杭州310023

出　　处：《计算机工程与应用》2009年第35期153-156,共4页Computer Engineering and Applications

基　　金：浙江省重大科技专项No.2007C11022~~

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。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.

关 键 词：影响域半径动态调整的移动最小二乘法(RSRMLS) 影响域半径 拟合 K-近邻 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]