Tooth mesh model automatic segmentation method based on principal curvature

Abstract

The invention belongs to the technical field of three dimensional image processing, and specifically relates to a tooth mesh model automatic segmentation method based on a principal curvature. The tooth mesh model automatic segmentation method takes a tooth triangular mesh model as a curved surface to calculate the principal curvature of the mesh, and utilizes the principal curvature and a region growing method to segment the tooth curved surface for single tooth. The tooth mesh model automatic segmentation method based on a principal curvature can keep away from complicated fuzzy clustering, and can take a tooth triangular mesh model as a curved surface to calculate the principal curvature of the mesh, and utilizes the principal curvature and a region growing method to segment the tooth curved surface for single tooth. The tooth mesh model automatic segmentation method based on a principal curvature can simplify operation of a doctor and clearly show single tooth model so as to help comprehensive inspection and rectification of teeth.

Description

technical field [0001] The invention belongs to the technical field of three-dimensional image processing, and in particular relates to an automatic segmentation method of a tooth mesh model based on principal curvature. Background technique [0002] Three-dimensional mesh model segmentation is the process of decomposing a polyhedron or an orientable two-dimensional manifold composed of meshes into a group of limited number, each with a simple shape meaning, and each connected sub-part according to certain geometric and topological characteristics. Most 3D grid segmentation algorithms are based on 2D image segmentation and extended to 3D grid space, such as: watershed algorithm, K-means, Mean-shift, region growth method, etc. Introducing geometric and topological information in the segmentation will improve the segmentation effect, such as: based on Gaussian curvature, based on voxel, based on Reeb diagram, based on skeleton extraction and topology scanning, etc. using three...

AI Technical Summary

Problems solved by technology

However, the difficulty of this method lies in the selection of weight coefficients. The polygonal nature of different tooth geometries will lead to inaccurate and fast convergence of the triangular mesh model under the combined external force.

[0006] At present, problems such as over-segmentation, inaccurate edges, a large amount of manual intervention, difficult selection of parameters, or high algorithm complexity are prone to occur during segmentation.

Today it is difficult to have a segmentation method suitable for all applications, most algorithms are proposed for specific problems

Images