Real object surface sampling point set normal estimation method based on local Poisson curved surface constraint

Abstract

The invention provides a scattered point cloud normal estimation method based on local Poisson curved surface constraint. The method is characterized in that a spatial index structure is constructed for a sampling point set; based on the structure, a local sample of a target sample point and a local reconstruction sample containing an auxiliary point are rapidly obtained; then constructing a Frenet standard frame of a target sample point according to the local sample; a Poisson grid curved surface is constructed for a local sample in the Frenet standard frame; and querying the nearest neighborgrid surface patch of the target sample point from the Poisson grid curved surface, determining the weight value of the neighborhood surface normal direction according to the shape and size of the first-order neighborhood surface of the vertex of the grid surface patch, calculating the nearest neighbor grid surface patch vertex normal direction, and taking the weighted sum of the nearest neighborgrid surface patch vertex normal directions as a robust estimation result of the sample point normal direction. An example proves that the method is suitable for scattered point cloud data of a complex curved surface, has a good normal calculation result for point clouds with noise and non-uniform sampling, and realizes normal smooth transition of sample points.

Description

technical field [0001] The invention provides a method for estimating the normal direction of a sampling point set on a physical surface constrained by a local Poisson surface, which can be used for robust estimation of the normal direction of sampling data on the physical surface, and belongs to the field of digital design and manufacturing. Background technique [0002] The sampling point set of the physical surface has the characteristics of simplicity, flexibility, and no need to maintain topological consistency, and has been widely used in the fields of industrial manufacturing, cultural relics protection, and medicine. The normal direction is one of the important attributes of the sampling point set, and the accuracy and robustness of its calculation results will directly affect the post-processing effect of the sampling point set. [0003] For the normal direction of any sample point in the sample point set of the physical surface, it can be estimated based on the app...

AI Technical Summary

Problems solved by technology

[0004] In order to obtain a neighborhood point set that can accurately reflect the local shape of the surface, the local sample acquisition speed of the above-mentioned improved algorithm is reduced to varying degrees, and the acquisition speed of local samples seriously restricts the efficiency of normal estimation. In addition, to obtain the optimal neighborhood point The sample attributes required by the set are usually unknown, so this type of algorithm is difficult to apply widely

The above-mentioned method based on the improved local surface approximation method to improve the robustness of normal calculation usually lacks enough flexibility when dealing with complex surfaces, and the robustness needs to be improved

Images