Graphic matching based on constrained Voronoi diagrams

Author

Zhang, Hailin ; Yan, Hong

Author_Institution

Sch. of Electr. & Inf. Eng., Sydney Univ., NSW, Australia

Volume

1

fYear

1999

fDate

1999

Firstpage

431

Abstract

This paper presents a novel approach to the inexact graph matching problem based on local relationships. The constrained Voronoi diagram (CVD) is used to partition a graph into a set of polygons. The CVD provides a convenient and robust neighborhood representation for the graph. In the proposed method, Voronoi neighboring relationships as well as geometrical characteristics are used to detect potentially matching pairs of objects. This method is implemented and tested in the matching process for 2D cartoon key frames with encouraging results

Keywords

computational geometry; computer vision; graph theory; image matching; image segmentation; 2D cartoon image key frames; Voronoi neighboring relationships; computer vision; constrained Voronoi diagrams; geometrical characteristics; graphic matching; inexact graph matching problem; local relationships; neighborhood representation; polygons; potentially matching pairs of objects; Australia; Computational geometry; Computer vision; Data structures; Graphics; Image segmentation; Object detection; Robustness; Signal processing; Testing;

fLanguage

English

Publisher

ieee

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

Type

conf

DOI

10.1109/ISSPA.1999.818204

Filename

818204