Presentation 2018-02-19
A top-down approach in dot-matrix character extraction
Takuya Mizoguchi, Yoshihisa Ijiri, Masaki Suwa, Kiyoshi Kiyokawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A dot-matrix character is a character formed by a dot matrix and used in such factory automation application as product quality management. If the machine can automatically recognize dot-matrix characters it will contribute greatly to the industry applications from the viewpoints of cost, operational aspects, etc. In the task of character recognition, it is common to generate an input image for individual characters, so image segmentation into each character region is necessary. However, such existing approaches relying on pixel connectivities cannot be used for dot-matrix characters. In this research, we propose a top-down segmentation method which gradually narrows down dot-matrix character regions considering such characteristics. In this research, we evaluated the proposed segmentation method with a variety of images containing printed dot-matrix characters including difficult properties such as distortion and specular highlights which have not been verified in existing studies.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Factory Automation / Dot-matrix characters / Segmentation / Top-down / Unsupervised Learning
Paper # PRMU2017-145,CNR2017-23
Date of Issue 2018-02-12 (PRMU, CNR)

Conference Information
Committee PRMU / CNR
Conference Date 2018/2/19(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Shinichi Sato(NII) / Tetsuo Ono(Hokkaido Univ.)
Vice Chair Hironobu Fujiyoshi(Chubu Univ.) / Yoshihisa Ijiri(Omron) / Masayuki Kanbara(NAIST) / Kazunori Takashio(Keio Univ.)
Secretary Hironobu Fujiyoshi(AIST) / Yoshihisa Ijiri(NAIST) / Masayuki Kanbara(Hokkaido Univ.) / Kazunori Takashio(Panasonic)
Assistant Masato Ishii(NEC) / Yusuke Sugano(Osaka Univ.) / Wataru Mito(SECOM) / Yuka Kobayashi(Toshiba) / Tatsuya Ishihara(NTT)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Cloud Network Robotics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A top-down approach in dot-matrix character extraction
Sub Title (in English)
Keyword(1) Factory Automation
Keyword(2) Dot-matrix characters
Keyword(3) Segmentation
Keyword(4) Top-down
Keyword(5) Unsupervised Learning
1st Author's Name Takuya Mizoguchi
1st Author's Affiliation Graduate School of Information Science, Nara Institute of Science and Technology(NAIST)
2nd Author's Name Yoshihisa Ijiri
2nd Author's Affiliation TECHNOLOGY AND INTELLECTUAL PROPERTY H.Q., OMRON Corporation(OMRON)
3rd Author's Name Masaki Suwa
3rd Author's Affiliation TECHNOLOGY AND INTELLECTUAL PROPERTY H.Q., OMRON Corporation(OMRON)
4th Author's Name Kiyoshi Kiyokawa
4th Author's Affiliation Graduate School of Information Science, Nara Institute of Science and Technology(NAIST)
Date 2018-02-19
Paper # PRMU2017-145,CNR2017-23
Volume (vol) vol.117
Number (no) PRMU-442,CNR-443
Page pp.pp.1-6(PRMU), pp.1-6(CNR),
#Pages 6
Date of Issue 2018-02-12 (PRMU, CNR)