An online handwriting recognition system for Turkish

Esra Vural, Hakan Erdogan, Kemal Oflazer, Berrin Yanikoglu

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

7 Citations (Scopus)

Abstract

Despite recent developments in Tablet PC technology, there has not been any applications for recognizing hand-writings in Turkish. In this paper, we present an online handwritten text recognition system for Turkish, developed using the Tablet PC interface. However, even though the system is developed for Turkish, the addressed issues are common to online handwriting recognition systems in general. Several dynamic features are extracted from the handwriting data for each recorded point and Hidden Markov Models (HMM) are used to train letter and word models. We experimented with using various features and HMM model topologies, and report on the effects of these experiments. We started with first and second derivatives of the x and y coordinates and relative change in the pen pressure as initial features. We found that using two more additional features, that is, number of neighboring points and relative heights of each point with respect to the base-line improve the recognition rate. In addition, extracting features within strokes and using a skipping state topology improve the system performance as well. The improved system performance is 94% in recognizing handwritten words from a 1000-word lexicon.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsE.H.B. Smith, K. Taghva
Pages56-65
Number of pages10
Volume5676
DOIs
Publication statusPublished - 2005
Externally publishedYes
EventProceedings of SPIE-IS and T Electronic Imaging - Document Recognition and Retrieval XII - San Jose, CA, United States
Duration: 19 Jan 200520 Jan 2005

Other

OtherProceedings of SPIE-IS and T Electronic Imaging - Document Recognition and Retrieval XII
CountryUnited States
CitySan Jose, CA
Period19/1/0520/1/05

Fingerprint

handwriting
Hidden Markov models
Topology
tablets
topology
Derivatives
pens
strokes
Experiments

Keywords

  • Handwriting
  • HMM
  • Online
  • Recognition
  • Turkish

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Vural, E., Erdogan, H., Oflazer, K., & Yanikoglu, B. (2005). An online handwriting recognition system for Turkish. In E. H. B. Smith, & K. Taghva (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5676, pp. 56-65). [07] https://doi.org/10.1117/12.588556

An online handwriting recognition system for Turkish. / Vural, Esra; Erdogan, Hakan; Oflazer, Kemal; Yanikoglu, Berrin.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / E.H.B. Smith; K. Taghva. Vol. 5676 2005. p. 56-65 07.

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

Vural, E, Erdogan, H, Oflazer, K & Yanikoglu, B 2005, An online handwriting recognition system for Turkish. in EHB Smith & K Taghva (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5676, 07, pp. 56-65, Proceedings of SPIE-IS and T Electronic Imaging - Document Recognition and Retrieval XII, San Jose, CA, United States, 19/1/05. https://doi.org/10.1117/12.588556
Vural E, Erdogan H, Oflazer K, Yanikoglu B. An online handwriting recognition system for Turkish. In Smith EHB, Taghva K, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5676. 2005. p. 56-65. 07 https://doi.org/10.1117/12.588556
Vural, Esra ; Erdogan, Hakan ; Oflazer, Kemal ; Yanikoglu, Berrin. / An online handwriting recognition system for Turkish. Proceedings of SPIE - The International Society for Optical Engineering. editor / E.H.B. Smith ; K. Taghva. Vol. 5676 2005. pp. 56-65
@inproceedings{14a2ccbba3234091b500aa3c68e52467,
title = "An online handwriting recognition system for Turkish",
abstract = "Despite recent developments in Tablet PC technology, there has not been any applications for recognizing hand-writings in Turkish. In this paper, we present an online handwritten text recognition system for Turkish, developed using the Tablet PC interface. However, even though the system is developed for Turkish, the addressed issues are common to online handwriting recognition systems in general. Several dynamic features are extracted from the handwriting data for each recorded point and Hidden Markov Models (HMM) are used to train letter and word models. We experimented with using various features and HMM model topologies, and report on the effects of these experiments. We started with first and second derivatives of the x and y coordinates and relative change in the pen pressure as initial features. We found that using two more additional features, that is, number of neighboring points and relative heights of each point with respect to the base-line improve the recognition rate. In addition, extracting features within strokes and using a skipping state topology improve the system performance as well. The improved system performance is 94{\%} in recognizing handwritten words from a 1000-word lexicon.",
keywords = "Handwriting, HMM, Online, Recognition, Turkish",
author = "Esra Vural and Hakan Erdogan and Kemal Oflazer and Berrin Yanikoglu",
year = "2005",
doi = "10.1117/12.588556",
language = "English",
volume = "5676",
pages = "56--65",
editor = "E.H.B. Smith and K. Taghva",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - An online handwriting recognition system for Turkish

AU - Vural, Esra

AU - Erdogan, Hakan

AU - Oflazer, Kemal

AU - Yanikoglu, Berrin

PY - 2005

Y1 - 2005

N2 - Despite recent developments in Tablet PC technology, there has not been any applications for recognizing hand-writings in Turkish. In this paper, we present an online handwritten text recognition system for Turkish, developed using the Tablet PC interface. However, even though the system is developed for Turkish, the addressed issues are common to online handwriting recognition systems in general. Several dynamic features are extracted from the handwriting data for each recorded point and Hidden Markov Models (HMM) are used to train letter and word models. We experimented with using various features and HMM model topologies, and report on the effects of these experiments. We started with first and second derivatives of the x and y coordinates and relative change in the pen pressure as initial features. We found that using two more additional features, that is, number of neighboring points and relative heights of each point with respect to the base-line improve the recognition rate. In addition, extracting features within strokes and using a skipping state topology improve the system performance as well. The improved system performance is 94% in recognizing handwritten words from a 1000-word lexicon.

AB - Despite recent developments in Tablet PC technology, there has not been any applications for recognizing hand-writings in Turkish. In this paper, we present an online handwritten text recognition system for Turkish, developed using the Tablet PC interface. However, even though the system is developed for Turkish, the addressed issues are common to online handwriting recognition systems in general. Several dynamic features are extracted from the handwriting data for each recorded point and Hidden Markov Models (HMM) are used to train letter and word models. We experimented with using various features and HMM model topologies, and report on the effects of these experiments. We started with first and second derivatives of the x and y coordinates and relative change in the pen pressure as initial features. We found that using two more additional features, that is, number of neighboring points and relative heights of each point with respect to the base-line improve the recognition rate. In addition, extracting features within strokes and using a skipping state topology improve the system performance as well. The improved system performance is 94% in recognizing handwritten words from a 1000-word lexicon.

KW - Handwriting

KW - HMM

KW - Online

KW - Recognition

KW - Turkish

UR - http://www.scopus.com/inward/record.url?scp=21944448472&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21944448472&partnerID=8YFLogxK

U2 - 10.1117/12.588556

DO - 10.1117/12.588556

M3 - Conference contribution

AN - SCOPUS:21944448472

VL - 5676

SP - 56

EP - 65

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Smith, E.H.B.

A2 - Taghva, K.

ER -