Lexicon-based word recognition using support vector machine and Hidden Markov Model

Abd Rahim Ahmad, C. Viard-Gaudin, M. Khalid

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

13 Citations (Scopus)

Abstract

Hybrid of Neural Network (NN) and Hidden Markov Model (HMM) has been popular in word recognition, taking advantage of NN discriminative property and HMM representational capability. However, NN does not guarantee good generalization due to Empirical Risk minimization (ERM) principle that it uses. In our work, we focus on online word recognition using the support vector machine (SVM) for character recognition. SVM's use of structural risk minimization (SRM) principle has allowed simultaneous optimization of representational and discriminative capability of the character recognizer. We evaluated SVM in isolated character recognition environment using IRONOFF and UNIPEN character database. We then demonstrate the practical issues in using SVM within a hybrid setting with HMM for word recognition by testing the hybrid system on the IRONOFF word database and obtained commendable results.

Original languageEnglish
Title of host publicationICDAR2009 - 10th International Conference on Document Analysis and Recognition
Pages161-165
Number of pages5
DOIs
Publication statusPublished - 2009
EventICDAR2009 - 10th International Conference on Document Analysis and Recognition - Barcelona, Spain
Duration: 26 Jul 200929 Jul 2009

Other

OtherICDAR2009 - 10th International Conference on Document Analysis and Recognition
CountrySpain
CityBarcelona
Period26/07/0929/07/09

Fingerprint

Hidden Markov models
Support vector machines
Character recognition
Neural networks
Hybrid systems
Testing

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Ahmad, A. R., Viard-Gaudin, C., & Khalid, M. (2009). Lexicon-based word recognition using support vector machine and Hidden Markov Model. In ICDAR2009 - 10th International Conference on Document Analysis and Recognition (pp. 161-165). [5277749] https://doi.org/10.1109/ICDAR.2009.248
Ahmad, Abd Rahim ; Viard-Gaudin, C. ; Khalid, M. / Lexicon-based word recognition using support vector machine and Hidden Markov Model. ICDAR2009 - 10th International Conference on Document Analysis and Recognition. 2009. pp. 161-165
@inproceedings{68d9f12b13fe452784abbc00b03c1ff4,
title = "Lexicon-based word recognition using support vector machine and Hidden Markov Model",
abstract = "Hybrid of Neural Network (NN) and Hidden Markov Model (HMM) has been popular in word recognition, taking advantage of NN discriminative property and HMM representational capability. However, NN does not guarantee good generalization due to Empirical Risk minimization (ERM) principle that it uses. In our work, we focus on online word recognition using the support vector machine (SVM) for character recognition. SVM's use of structural risk minimization (SRM) principle has allowed simultaneous optimization of representational and discriminative capability of the character recognizer. We evaluated SVM in isolated character recognition environment using IRONOFF and UNIPEN character database. We then demonstrate the practical issues in using SVM within a hybrid setting with HMM for word recognition by testing the hybrid system on the IRONOFF word database and obtained commendable results.",
author = "Ahmad, {Abd Rahim} and C. Viard-Gaudin and M. Khalid",
year = "2009",
doi = "10.1109/ICDAR.2009.248",
language = "English",
isbn = "9780769537252",
pages = "161--165",
booktitle = "ICDAR2009 - 10th International Conference on Document Analysis and Recognition",

}

Ahmad, AR, Viard-Gaudin, C & Khalid, M 2009, Lexicon-based word recognition using support vector machine and Hidden Markov Model. in ICDAR2009 - 10th International Conference on Document Analysis and Recognition., 5277749, pp. 161-165, ICDAR2009 - 10th International Conference on Document Analysis and Recognition, Barcelona, Spain, 26/07/09. https://doi.org/10.1109/ICDAR.2009.248

Lexicon-based word recognition using support vector machine and Hidden Markov Model. / Ahmad, Abd Rahim; Viard-Gaudin, C.; Khalid, M.

ICDAR2009 - 10th International Conference on Document Analysis and Recognition. 2009. p. 161-165 5277749.

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

TY - GEN

T1 - Lexicon-based word recognition using support vector machine and Hidden Markov Model

AU - Ahmad, Abd Rahim

AU - Viard-Gaudin, C.

AU - Khalid, M.

PY - 2009

Y1 - 2009

N2 - Hybrid of Neural Network (NN) and Hidden Markov Model (HMM) has been popular in word recognition, taking advantage of NN discriminative property and HMM representational capability. However, NN does not guarantee good generalization due to Empirical Risk minimization (ERM) principle that it uses. In our work, we focus on online word recognition using the support vector machine (SVM) for character recognition. SVM's use of structural risk minimization (SRM) principle has allowed simultaneous optimization of representational and discriminative capability of the character recognizer. We evaluated SVM in isolated character recognition environment using IRONOFF and UNIPEN character database. We then demonstrate the practical issues in using SVM within a hybrid setting with HMM for word recognition by testing the hybrid system on the IRONOFF word database and obtained commendable results.

AB - Hybrid of Neural Network (NN) and Hidden Markov Model (HMM) has been popular in word recognition, taking advantage of NN discriminative property and HMM representational capability. However, NN does not guarantee good generalization due to Empirical Risk minimization (ERM) principle that it uses. In our work, we focus on online word recognition using the support vector machine (SVM) for character recognition. SVM's use of structural risk minimization (SRM) principle has allowed simultaneous optimization of representational and discriminative capability of the character recognizer. We evaluated SVM in isolated character recognition environment using IRONOFF and UNIPEN character database. We then demonstrate the practical issues in using SVM within a hybrid setting with HMM for word recognition by testing the hybrid system on the IRONOFF word database and obtained commendable results.

UR - http://www.scopus.com/inward/record.url?scp=71249127581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71249127581&partnerID=8YFLogxK

U2 - 10.1109/ICDAR.2009.248

DO - 10.1109/ICDAR.2009.248

M3 - Conference contribution

AN - SCOPUS:71249127581

SN - 9780769537252

SP - 161

EP - 165

BT - ICDAR2009 - 10th International Conference on Document Analysis and Recognition

ER -

Ahmad AR, Viard-Gaudin C, Khalid M. Lexicon-based word recognition using support vector machine and Hidden Markov Model. In ICDAR2009 - 10th International Conference on Document Analysis and Recognition. 2009. p. 161-165. 5277749 https://doi.org/10.1109/ICDAR.2009.248