Presentation 2000/2/21
Retrieval of Relevant Parts of Document Images Based on Density Distributions of Keywords
Masaaki Tsujino, koichi Kise, Keinosuke Matsumoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to improve the retrieval accuracy of document image databases, it is necessary not to test the relevance between a query and each page / document, but to test whether each page contains parts relevant to a query. In this report, we propose a method for the retrieval of relevant parts of document images. In our method, the relevance is defined based on the observation that parts of document images are relevant if they densely contain keywords in a query. The use of 2D density distributions of feature vectors enables us to represent the density of keywords as well as to locate relevant regions in document images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Passage retrieval / Document image / Density distribution / Document image database
Paper # PRMU99-226
Date of Issue

Conference Information
Committee PRMU
Conference Date 2000/2/21(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Retrieval of Relevant Parts of Document Images Based on Density Distributions of Keywords
Sub Title (in English)
Keyword(1) Passage retrieval
Keyword(2) Document image
Keyword(3) Density distribution
Keyword(4) Document image database
1st Author's Name Masaaki Tsujino
1st Author's Affiliation Dept. of Computer and Systems Sciences, College of Engineering, Osaka Prefecture University()
2nd Author's Name koichi Kise
2nd Author's Affiliation Dept. of Computer and Systems Sciences, College of Engineering, Osaka Prefecture University
3rd Author's Name Keinosuke Matsumoto
3rd Author's Affiliation Dept. of Computer and Systems Sciences, College of Engineering, Osaka Prefecture University
Date 2000/2/21
Paper # PRMU99-226
Volume (vol) vol.99
Number (no) 648
Page pp.pp.-
#Pages 8
Date of Issue