High-fidelity triangular mesh smoothing algorithm

Abstract

The invention discloses a high-fidelity triangular mesh smoothing algorithm which comprises the following steps: step 1, classifying mesh vertexes; step 2, filtering by a double-line method; step 3, clustering adjacent patches; step 4, calculating vertex normals; the invention belongs to the technical field of triangular mesh smoothing algorithms, and particularly relates to a high-fidelity triangular mesh smoothing algorithm, which not only can effectively protect geometrical characteristic information of an original mesh, but also can effectively smooth the original mesh, and is robust.

Description

technical field [0001] The invention belongs to the technical field of triangular mesh smoothing algorithm, and specifically refers to a high-fidelity triangular mesh smoothing algorithm. Background technique [0002] 3D mesh smoothing is an important processing tool in computer graphics, because the triangular mesh obtained by the existing 3D scanning equipment or surface reconstruction algorithm will inevitably produce noise or errors visible to the naked eye, causing the source of these noises or errors There are many aspects, such as the accuracy of the scanning equipment, the use environment of the scanning equipment, and the material properties of the scanned object. These errors not only cannot objectively reflect the real data information, but also affect the subsequent graphics processing (for example: finite element analysis, fluid mechanics analysis, etc.) have obvious effects, so no matter from the aesthetic point of view or the subsequent graphics processing, th...

AI Technical Summary

Problems solved by technology

[0002] 3D mesh smoothing is an important processing tool in computer graphics, because the triangular mesh obtained by the existing 3D scanning equipment or surface reconstruction algorithm will inevitably produce noise or errors visible to the naked eye, causing the source of these noises or errors There are many aspects, such as the accuracy of the scanning equipment, the use environment of the scanning equipment, and the material properties of the scanned object. These errors not only cannot objectively reflect the real data information, but also affect the subsequent graphics processing (for example: finite element analysis, fluid mechanics analysis, etc.) have obvious effects, so no matter from the aesthetic point of view or the subsequent graphics processing, the smoothing algorithm of the triangular mesh is very necessary, and in the prior art, the existing mesh smoothing algorithm It is mainly divided into isotropy and anisotropy. The early grid smoothing algorithms are mainly isotropic, and the most representative one is Laplace smoothing. This smoothing algorithm is simple and fast, but this algorithm not only cannot Effectively retain the geometric feature information, and also shrink the size of the grid. Compared with the previous isotropic smoothing algorithm, the anisotropic smoothing algorithm can usually better protect the geometric features. The existing anisotropic smoothing algorithm mainly uses 2D The characteristics of the grid protect the anisotropic diffusion to the geometric anisotropy of the 3D grid. Although such algorithms can produce high-quality grid data, they will not be robust due to the numerical instability of the diffusion equation. In summary As mentioned above, the existing network smoothing algorithm will erase the geometric feature information of the original grid while smoothing, especially some engineering grids with obvious geometric features, and the existing network smoothing algorithm will almost lose Remove all geometric feature information

Images