Presentation 2006-06-15
A statistical approach towords improving the quality of transcript mapping
Kengo TERASAWA, Toshio KAWASHIMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper we present a transcript mapping method for brush-written document images. When there exists a document image file and a document transcript file separately, transcript mapping system associate each character in the transcript file with the corresponding region in the image file. In addition to our earlier study of image similarity based matching process, we have employed the similarity information of appearance location of the frequently appearing pattern both in image and in text, which facilitates transcript mapping for relatively long document. Experimental result for the real document written in mid 19th century has confirmed the validity of our method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Brush-written Document Images / Transcript Mapping / DP Matching
Paper # DE2006-4,PRMU2006-42
Date of Issue

Conference Information
Committee PRMU
Conference Date 2006/6/8(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) A statistical approach towords improving the quality of transcript mapping
Sub Title (in English)
Keyword(1) Brush-written Document Images
Keyword(2) Transcript Mapping
Keyword(3) DP Matching
1st Author's Name Kengo TERASAWA
1st Author's Affiliation Meme Media Laboratory, Hokkaido University()
2nd Author's Name Toshio KAWASHIMA
2nd Author's Affiliation School of Systems Information Science, Future University-Hakodate
Date 2006-06-15
Paper # DE2006-4,PRMU2006-42
Volume (vol) vol.106
Number (no) 99
Page pp.pp.-
#Pages 6
Date of Issue