Presentation 1999/1/21
Detection of Telop and Flip Frames and Extraction of Characters from News Images
Katsumi Matsuura, Seiichi Takao, Yoshiaki Sugiyama, Yasuo Ariki,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to classify news articles into topics, it is required to recognize telop and flip characters. In this study, we propose a method to detect frames including telop and flip, and then to extract the characters. At first, frames are detected in which character regions appear and disappear. Next, stable frame sections are extracted where character regions are stable. Finally the character regions are extract based on the line density and peripheral features. We applied our method to news images and obtained better results.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) retrieval of image data / telop and flip characters / extraction of frame sections / characters region extraction
Paper # PRMU98-188
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/1/21(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) Detection of Telop and Flip Frames and Extraction of Characters from News Images
Sub Title (in English)
Keyword(1) retrieval of image data
Keyword(2) telop and flip characters
Keyword(3) extraction of frame sections
Keyword(4) characters region extraction
1st Author's Name Katsumi Matsuura
1st Author's Affiliation Faculty of Science and Technology, Ryukoku University()
2nd Author's Name Seiichi Takao
2nd Author's Affiliation Faculty of Science and Technology, Ryukoku University
3rd Author's Name Yoshiaki Sugiyama
3rd Author's Affiliation Faculty of Science and Technology, Ryukoku University
4th Author's Name Yasuo Ariki
4th Author's Affiliation Faculty of Science and Technology, Ryukoku University
Date 1999/1/21
Paper # PRMU98-188
Volume (vol) vol.98
Number (no) 527
Page pp.pp.-
#Pages 8
Date of Issue