Reconstruction of Conic Surface Based on PCA Normal Vector Calculation and Projection Method
ZHANG Qinglan
HE Xueming
LIU Ping
Abstract：For the conic surface fitting in regular surface feature reconstruction,to propose a method for conic fitting based on PCA Normal Vector estimation and projection method.In order to achieve parameterization reconstruction of the cone,the approxi-mation function model of the distance function is first used as the objective function,the cone fitting parameters to be calculated are determined,and then the initial value of the cone parameter is solved.When calculating the initial value of the parameter,the point cloud normal vector is calculated by the PCA principal component analysis method,the projection method is used to calculate the center of the bottom surface after generating the axis,and then the initial value of other parameters of the cone is solved,and finally the cone surface is reconstructed iteratively in combination with the Levenberg-Marquardt method.By analyzing the cone scattered point cloud,the initial value of the parameter obtained by this method is iterated,the number of iterations is 7 times,and the final average error of the iteration is 1.8804e-04.This method is well applied in scattered point cloud,reduces the influence of sparse data on the initial value,improves the iterative non-convergence caused by the initial value of iteration,and then obtains more accurate cone parameters.
Keywords：cone parameterprincipal components analysisnormal vectorprojectionleast square method
Publication Date：2023-12-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 2801-2806 )