Authors:
Michal Spanel
;
Premysl Krsek
;
Miroslav Svub
and
Vit Stancl
Affiliation:
Brno University of Technology, Czech Republic
Keyword(s):
CT/MR data, Medical image segmentation, Surface reconstruction, Delaunay triangulation, Variational tetrahedral meshing, Isotropic meshing, Graph-cut segmentation.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Geometric Computing
;
Geometry and Modeling
;
Modeling and Algorithms
;
Surface Modeling
Abstract:
In this paper, a vector segmentation algorithm of volumetric data based on the 3D Delaunay triangulation is presented. A modified variational meshing method is used to adapt tetrahedral mesh to the underlying CT/MRI volumetric data. Moreover, to classify tetrahedra in the mesh into regions whose characteristics are similar, a clustering scheme viewing the mesh as undirected graph with edges weighted according to similarity of tetrahedra is discussed.