Presentation 2015-02-19
Character Detection With Stroke Width Features From Scene Image
Yasushi OSHIMA, Tomo MIYAZAKI, Yoshihiro SUGAYA, Shinichiro OMACHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We perform the character detection from the scene image. In this paper, character regions are extracted as connected components. By using CCs extracted by Maximally Stable Extremal Regions (MSER), we calculate features for classifying CCs into character regions or other regions. We consider that there exists a characteristic such that stroke widths of characters in CC of the character regions are nearly constant. Therefore, we describe a feature vector of CCs from stroke widths, and detect CC of the character using a binary classifier by SVM.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Character Detection / Connected Component / Stroke Width
Paper # PRMU2014-133,CNR2014-48
Date of Issue

Conference Information
Committee CNR
Conference Date 2015/2/12(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 Cloud Network Robotics (CNR)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Character Detection With Stroke Width Features From Scene Image
Sub Title (in English)
Keyword(1) Character Detection
Keyword(2) Connected Component
Keyword(3) Stroke Width
1st Author's Name Yasushi OSHIMA
1st Author's Affiliation School of Engineering, Tohoku University()
2nd Author's Name Tomo MIYAZAKI
2nd Author's Affiliation School of Engineering, Tohoku University
3rd Author's Name Yoshihiro SUGAYA
3rd Author's Affiliation School of Engineering, Tohoku University
4th Author's Name Shinichiro OMACHI
4th Author's Affiliation School of Engineering, Tohoku University
Date 2015-02-19
Paper # PRMU2014-133,CNR2014-48
Volume (vol) vol.114
Number (no) 455
Page pp.pp.-
#Pages 2
Date of Issue