Weighted link analysis for logo and trademark image retrieval on the Web

Author

Voutsakis, Epimenides ; Petrakis, Euripides G M ; Milios, Evangelos

Author_Institution

Dept. of Electron. & Comput. Eng., Crete Tech. Univ., Greece

fYear

2005

fDate

19-22 Sept. 2005

Firstpage

581

Lastpage

585

Abstract

Image retrieval on the Web requires that important (authoritative) images satisfying the query selection criteria are assigned higher ranking over other relevant images. PicASHOW (Lempel and Soffer, 2002) achieves this goal using link information alone. This work proposes WPicASHOW (weighted PicASHOW), a weighted scheme for co-citation analysis that incorporates within the link analysis method of PicASHOW the text and image content of the queries and of the Web pages. WPicASHOW is implemented and integrated into a fully automated Web retrieval system for logo and trademark images.

Keywords

Web sites; content-based retrieval; image retrieval; query formulation; Web page; Web retrieval system; World Wide Web; authoritative image; co-citation analysis; image content; logo image retrieval; query selection criteria; text content; trademark image retrieval; weighted PicASHOW; weighted link analysis; Content based retrieval; Data mining; Filters; Image analysis; Image retrieval; Image storage; Information retrieval; Trademarks; Uniform resource locators; Web pages;

fLanguage

English

Publisher

ieee

Conference_Titel

Web Intelligence, 2005. Proceedings. The 2005 IEEE/WIC/ACM International Conference on

Print_ISBN

0-7695-2415-X

Type

conf

DOI

10.1109/WI.2005.162

Filename

1517912