Simultaneous mixed vertical and horizontal handwritten Japanese character line detection

Tomotaka Kimura, Chinthaka Premachandra, Hiroharu Kawanaka

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Teachers consume considerable time and their energy in process of marking examination sheets. To reduce this burden on teachers, we have been developing an automatic marking system. To mark examination sheets automatically, handwritten character lines are extracted, and then the characters on those lines are recognized. In this paper, we discuss how character lines are extracted from Japanese handwritten examination sheets without ruled lines. Japanese characters can be written vertically and horizontally, so examination sheets written in Japanese are consisted of mixed vertical and horizontal (MVH) character lines. Conventional character line extraction methods cannot deal with MVH lines, because they have been developed to consider only horizontal character lines. This paper focuses on the simultaneous detection of MVH character lines. The result of experiments using appropriate examination sheet images shows that our method can detect MVH character lines effectively.

Original languageEnglish
Title of host publicationComputer Vision and Graphics - International Conference, ICCVG 2016, Proceedings
PublisherSpringer Verlag
Pages564-572
Number of pages9
Volume9972 LNCS
ISBN (Print)9783319464176
DOIs
Publication statusPublished - 2016
EventInternational Conference on Computer Vision and Graphics, ICCVG 2016 - Warsaw, Poland
Duration: 2016 Sep 192016 Sep 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9972 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

OtherInternational Conference on Computer Vision and Graphics, ICCVG 2016
CountryPoland
CityWarsaw
Period16/9/1916/9/21

Fingerprint

Experiments

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Kimura, T., Premachandra, C., & Kawanaka, H. (2016). Simultaneous mixed vertical and horizontal handwritten Japanese character line detection. In Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings (Vol. 9972 LNCS, pp. 564-572). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9972 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46418-3_50

Simultaneous mixed vertical and horizontal handwritten Japanese character line detection. / Kimura, Tomotaka; Premachandra, Chinthaka; Kawanaka, Hiroharu.

Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings. Vol. 9972 LNCS Springer Verlag, 2016. p. 564-572 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9972 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kimura, T, Premachandra, C & Kawanaka, H 2016, Simultaneous mixed vertical and horizontal handwritten Japanese character line detection. in Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings. vol. 9972 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9972 LNCS, Springer Verlag, pp. 564-572, International Conference on Computer Vision and Graphics, ICCVG 2016, Warsaw, Poland, 16/9/19. https://doi.org/10.1007/978-3-319-46418-3_50
Kimura T, Premachandra C, Kawanaka H. Simultaneous mixed vertical and horizontal handwritten Japanese character line detection. In Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings. Vol. 9972 LNCS. Springer Verlag. 2016. p. 564-572. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-46418-3_50
Kimura, Tomotaka ; Premachandra, Chinthaka ; Kawanaka, Hiroharu. / Simultaneous mixed vertical and horizontal handwritten Japanese character line detection. Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings. Vol. 9972 LNCS Springer Verlag, 2016. pp. 564-572 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{75b028d0ba37408cb2147417a3902549,
title = "Simultaneous mixed vertical and horizontal handwritten Japanese character line detection",
abstract = "Teachers consume considerable time and their energy in process of marking examination sheets. To reduce this burden on teachers, we have been developing an automatic marking system. To mark examination sheets automatically, handwritten character lines are extracted, and then the characters on those lines are recognized. In this paper, we discuss how character lines are extracted from Japanese handwritten examination sheets without ruled lines. Japanese characters can be written vertically and horizontally, so examination sheets written in Japanese are consisted of mixed vertical and horizontal (MVH) character lines. Conventional character line extraction methods cannot deal with MVH lines, because they have been developed to consider only horizontal character lines. This paper focuses on the simultaneous detection of MVH character lines. The result of experiments using appropriate examination sheet images shows that our method can detect MVH character lines effectively.",
author = "Tomotaka Kimura and Chinthaka Premachandra and Hiroharu Kawanaka",
year = "2016",
doi = "10.1007/978-3-319-46418-3_50",
language = "English",
isbn = "9783319464176",
volume = "9972 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "564--572",
booktitle = "Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings",

}

TY - GEN

T1 - Simultaneous mixed vertical and horizontal handwritten Japanese character line detection

AU - Kimura, Tomotaka

AU - Premachandra, Chinthaka

AU - Kawanaka, Hiroharu

PY - 2016

Y1 - 2016

N2 - Teachers consume considerable time and their energy in process of marking examination sheets. To reduce this burden on teachers, we have been developing an automatic marking system. To mark examination sheets automatically, handwritten character lines are extracted, and then the characters on those lines are recognized. In this paper, we discuss how character lines are extracted from Japanese handwritten examination sheets without ruled lines. Japanese characters can be written vertically and horizontally, so examination sheets written in Japanese are consisted of mixed vertical and horizontal (MVH) character lines. Conventional character line extraction methods cannot deal with MVH lines, because they have been developed to consider only horizontal character lines. This paper focuses on the simultaneous detection of MVH character lines. The result of experiments using appropriate examination sheet images shows that our method can detect MVH character lines effectively.

AB - Teachers consume considerable time and their energy in process of marking examination sheets. To reduce this burden on teachers, we have been developing an automatic marking system. To mark examination sheets automatically, handwritten character lines are extracted, and then the characters on those lines are recognized. In this paper, we discuss how character lines are extracted from Japanese handwritten examination sheets without ruled lines. Japanese characters can be written vertically and horizontally, so examination sheets written in Japanese are consisted of mixed vertical and horizontal (MVH) character lines. Conventional character line extraction methods cannot deal with MVH lines, because they have been developed to consider only horizontal character lines. This paper focuses on the simultaneous detection of MVH character lines. The result of experiments using appropriate examination sheet images shows that our method can detect MVH character lines effectively.

UR - http://www.scopus.com/inward/record.url?scp=84989884549&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84989884549&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-46418-3_50

DO - 10.1007/978-3-319-46418-3_50

M3 - Conference contribution

AN - SCOPUS:84989884549

SN - 9783319464176

VL - 9972 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 564

EP - 572

BT - Computer Vision and Graphics - International Conference, ICCVG 2016, Proceedings

PB - Springer Verlag

ER -