Research On Topological Reconstruction Of Sharp Features For Complex Surface

Surface reconstruction is the process of constructing approximate surfaces based on point clouds of laser scanning sampled data.Reconstruct high-precision original surface from complex sampling point clouds,especially point clouds with features,has become a research hotspot in recent years.This paper contains the sharp features reconstruction related issues,research on normal estimation and surface reconstruction methods,the main research contents are following:(1)A method for normal estimation and consistency processing of point clouds with shape complexity constraints is proposed.Firstly,curvature analysis is used to classify point clouds with different complexity,and clustering analysis algorithm is used to block point clouds to avoid local samples searched by Euclidean neighborhood crossing sharp features.Then the normal direction calculation of points classification with different complexity.The complexity of local samples of point clouds in non-smooth regions is quantitatively analyzed so that the local sample size can be adaptively adjusted to improve the accuracy of normal calculation.Aiming at the sharp points,a method for calculating the normal direction of sharp points based on multi-normal weight optimization is proposed.Finally,the multi-normal strategy of sharp points is used to realize the correct propagation of two adjacent normal directions and complete the normal consistency processing.The method can improve the normal calculation precision of sharp points and solve the problem of uniform processing of adjacent surface points.(2)A method for sharp features reconstruction is proposed based on intersection of meshes extension.Firstly,sharp feature points are removed by Gauss map clustering,and the remaining points are segmented by seed point growth algorithm.Then,to improve the smoothness of surfaces continuation areas,the boundary sample points after gain optimization are taken as the local sample of surfaces,and the points are extended with the extended direction of the cubic Bezier curve as the guidance.Finally,the reconstruction results of the extended flat areas are intersected,and sharp features are reconstructed by the method of surfaces trimming.The method effectively improves shape feature surfaces approximation reconstruction,and can adapt to non-uniform sampling points.(3)A method of sharp feature surfaces reconstruction based on sharp points constraint partitioning.Firstly,the denoising problem of feature points set is transformed into curve smoothing denoising to avoid sharp feature points being smoothed.Then,the feature point cloud after noise reduction optimization is subdivided into dimension-reduced Delaunay subdivisions,and the subdivisions are detected and optimized according to the adjacent topological relationship of feature points.The vertex adjacency relation of the mesh subdivided correctly is mapped to the original point set to realize the accurate reconstruction of feature surface.Finally,Cocone reconstruction of non-closed flat surface is realized by using the neighborhood point set of feature surface boundary as the auxiliary point,and interpolation reconstruction of complete surface is realized by extracting the manifold of two kinds of surface boundary meshes.The method can avoid errors such as holes in the three-dimensional reconstruction of sharp features,and improve the accuracy of interpolation and reconstruction of sharp feature surfaces.