Presentation 2004/7/7
A Study on an Adaptive Metadata Generation Method for Image Databases using Content Based Image Retrieval
Hideyasu SASAKI, Yasushi KIYOKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we present an adaptive filter for metadata indexing that is implemented on content-based image retrieval (CBIR). We have implemented an automatic filter for metadata indexing that is optimized to its domain of retrieval target . A CBIR system returns the similarity degree to a requested image data against fundamental image files given with well-defined sample indexes. The implemented filter applies the returned values to those sample indexes as weights. The sample indexes given with weights forms a newly defined metadata to the requested image file. In the case of ranking conflict, our method generates metadata that are adaptive to selected domains to identify query images within at most two geometric types of visual objects. We report several experimental results on the adaptive metadata generation filter and its feasibility to be applied to the more advanced search engines.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Metadata / Visual Features / Content-Based Retrieval / Image Data / Adaptive Metadata Indexing / CBIR
Paper # DE2004-75
Date of Issue

Conference Information
Committee DE
Conference Date 2004/7/7(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on an Adaptive Metadata Generation Method for Image Databases using Content Based Image Retrieval
Sub Title (in English)
Keyword(1) Metadata
Keyword(2) Visual Features
Keyword(3) Content-Based Retrieval
Keyword(4) Image Data
Keyword(5) Adaptive Metadata Indexing
Keyword(6) CBIR
1st Author's Name Hideyasu SASAKI
1st Author's Affiliation Graduate School of Media and Governance, Keio University()
2nd Author's Name Yasushi KIYOKI
2nd Author's Affiliation Faculty of Environmental Information, Keio University
Date 2004/7/7
Paper # DE2004-75
Volume (vol) vol.104
Number (no) 177
Page pp.pp.-
#Pages 6
Date of Issue