大规模散乱点云数据的曲率估算及计算机实现

摘要

提出一种基于Matlab的大规模散乱点云数据的曲率估算的计算机实现方法.根据规模点云数据分布的特点,对散乱点云数据进行区域划分;对每个区域内的点云数据,用二次参数曲面逼近空间邻近点集,计算各点法矢,对不协调的法矢方向进行调整,计算各点曲率.实验表明,这种方法简化了散乱点云数据的曲率的计算量,从而在普通计算机上能实现大规模散乱点云数据的求解.%The paper proposes a computer implementation of curvature estimation of cloud data with mass scattered points based on Matlab language.According to the features of cloud data distribution,cloud data with scattered points is divided into different areas.Adjacent points are approached by parametric quadric surface to calculate the normal vector of each point.Inconsistent normal vector is adjusted and the curvature of each point is calculated.Experiments show the method reduces the calculation of curvature of cloud data with scattered points.As a result,it can be realized on ordinary computers.