Shape retrieval using pyramid matching with orientation features

Author

Zhou, Rong ; Zhang, Liqing

Author_Institution

Dept. of Comput. Sci. & Eng., Shanghai Jiao Tong Univ., Shanghai, China

Volume

4

fYear

2009

fDate

20-22 Nov. 2009

Firstpage

431

Lastpage

434

Abstract

We present a shape retrieval model using pyramid matching with orientation features. The orientation features have two kinds: one is for the edge of a shape, the other is for the angle of a shape. By capturing both edge and angle characteristics of a shape, combination of the two kinds of orientation features provides an efficient representation of a shape. Pyramid matching integrates the comparison between two shapes at different resolutions. So combination of the two kinds of orientation features and pyramid matching enhance the model´s retrieval performance. Our model is tested on two well-known databases: Kimia´s 99 database and MPEG7 CE shape-1 part B database, and achieves the effective performance compared with other methods.

Keywords

feature extraction; image retrieval; Kimia 99 database; MPEG7 CE shape-1 part B database; orientation features; pyramid matching; shape angle; shape edge; shape retrieval model; Brain modeling; Computer science; Feature extraction; Gabor filters; Histograms; Image edge detection; Neurons; Shape measurement; Spatial databases; Testing; orientation features; pyramid matching; shape retrieval;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Computing and Intelligent Systems, 2009. ICIS 2009. IEEE International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-4754-1

Electronic_ISBN

978-1-4244-4738-1

Type

conf

DOI

10.1109/ICICISYS.2009.5357650

Filename

5357650