Title :
A new skeletonization algorithm based on constrained Delaunay triangulation
Author :
Zou, Ju Jia ; Chang, Hung-Hsin ; Yan, Hong
Author_Institution :
Sch. of Electr. & Inf. Eng., Sydney Univ., NSW, Australia
Abstract :
A new skeletonization algorithm based on the constrained Delaunay triangulation (CDT) is proposed in this paper. The CDT partitions a shape into a set of nonoverlapping triangles which represent the shape´s local symmetry properties and interconnecting relationships between branches. The skeleton of the shape is generated from the skeletons of the triangles. Methods for removing skeletonization artefacts are provided. An outstanding feature of the algorithm is that the structural information at intersections of a shape is preserved in its skeleton
Keywords :
image representation; image segmentation; image thinning; mesh generation; constrained Delaunay triangulation; local symmetry properties; nonoverlapping triangles; shape intersections; shape segmentation; skeletal representation; skeletonization algorithm; skeletonization artefacts removal; structural information; Australia; Digital images; Image segmentation; Joining processes; Partitioning algorithms; Pixel; Shape; Signal processing; Skeleton;
Conference_Titel :
Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
1-86435-451-8
DOI :
10.1109/ISSPA.1999.815823
