A novel method for representing industrial inspection pattern

Author

Sun, Yung-nien ; Tsai, Ching-tsorng

Author_Institution

Inst. of Inf. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan

fYear

1990

fDate

24-27 Sep 1990

Firstpage

596

Abstract

A critical step in industrial pattern inspection is to design an effective and flexible method for representing the inspected pattern. The authors propose a novel representation method called pattern attributed hypergraph (PAHG). The PAHG is used to describe the pictorial information of the input pattern, which includes attributed nodes, connectivity between nodes (intrarelation), regional primitives, and neighborhood relations between regions (spatial interrelation). It is found that the complexity of graph matching with the PAHG structure is much lower than the corresponding complexity of the graph structure used by A.M. Darwish and A.K. Jain (1988). The reduced complexity is approximately 1/K² of the original, where K is the number of regions in the inspected pattern. It provides a better and faster approach for industrial pattern inspection

Keywords

computer vision; graph theory; inspection; computer vision; connectivity; graph matching; industrial inspection pattern; input pattern; neighborhood relations; pattern attributed hypergraph; pattern representation; pictorial information; regional primitives; spatial interrelation; Artificial intelligence; Computational complexity; Computer vision; Image edge detection; Image segmentation; Inspection; Manufacturing industries; Smoothing methods; Sun; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Communication Systems, 1990. IEEE TENCON'90., 1990 IEEE Region 10 Conference on

Print_ISBN

0-87942-556-3

Type

conf

DOI

10.1109/TENCON.1990.152680

Filename

152680