Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition

Jawad H. Alkhateeb, Olivier Pauplin, Jinchang Ren, Jianmin Jiang

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

This paper presents a comparative study of two machine learning techniques for recognizing handwritten Arabic words, where hidden Markov models (HMMs) and dynamic Bayesian networks (DBNs) were evaluated. The work proposed is divided into three stages, namely preprocessing, feature extraction and classification. Preprocessing includes baseline estimation and normalization as well as segmentation. In the second stage, features are extracted from each of the normalized words, where a set of new features for handwritten Arabic words is proposed, based on a sliding window approach moving across the mirrored word image. The third stage is for classification and recognition, where machine learning is applied using HMMs and DBNs. In order to validate the techniques, extensive experiments were conducted using the IFN/ENIT database which contains 32,492 Arabic words. Experimental results and quantitative evaluations showed that HMM outperforms DBN in terms of higher recognition rate and lower complexity. 

LanguageEnglish
Pages680-688
Number of pages9
JournalKnowledge Based Systems
Volume24
Issue number5
DOIs
Publication statusPublished - 1 Jul 2011

Fingerprint

Bayesian networks
Hidden Markov models
Classifiers
Learning systems
Feature extraction
Hidden Markov model
Classifier
Experiments
Machine learning

Keywords

  • off-line handwritten recognition
  • hidden markov model (HMM)
  • dynamic bayesian network (DBN)
  • performance evaluation
  • IFN/ENIT database

Cite this

Alkhateeb, Jawad H. ; Pauplin, Olivier ; Ren, Jinchang ; Jiang, Jianmin. / Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition. In: Knowledge Based Systems. 2011 ; Vol. 24, No. 5. pp. 680-688.
@article{af9beac17d3645ec8a613f17a9e24536,
title = "Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition",
abstract = "This paper presents a comparative study of two machine learning techniques for recognizing handwritten Arabic words, where hidden Markov models (HMMs) and dynamic Bayesian networks (DBNs) were evaluated. The work proposed is divided into three stages, namely preprocessing, feature extraction and classification. Preprocessing includes baseline estimation and normalization as well as segmentation. In the second stage, features are extracted from each of the normalized words, where a set of new features for handwritten Arabic words is proposed, based on a sliding window approach moving across the mirrored word image. The third stage is for classification and recognition, where machine learning is applied using HMMs and DBNs. In order to validate the techniques, extensive experiments were conducted using the IFN/ENIT database which contains 32,492 Arabic words. Experimental results and quantitative evaluations showed that HMM outperforms DBN in terms of higher recognition rate and lower complexity. ",
keywords = "off-line handwritten recognition, hidden markov model (HMM), dynamic bayesian network (DBN), performance evaluation, IFN/ENIT database",
author = "Alkhateeb, {Jawad H.} and Olivier Pauplin and Jinchang Ren and Jianmin Jiang",
year = "2011",
month = "7",
day = "1",
doi = "10.1016/j.knosys.2011.02.008",
language = "English",
volume = "24",
pages = "680--688",
journal = "Knowledge Based Systems",
issn = "0950-7051",
number = "5",

}

Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition. / Alkhateeb, Jawad H.; Pauplin, Olivier; Ren, Jinchang; Jiang, Jianmin.

In: Knowledge Based Systems, Vol. 24, No. 5, 01.07.2011, p. 680-688.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition

AU - Alkhateeb, Jawad H.

AU - Pauplin, Olivier

AU - Ren, Jinchang

AU - Jiang, Jianmin

PY - 2011/7/1

Y1 - 2011/7/1

N2 - This paper presents a comparative study of two machine learning techniques for recognizing handwritten Arabic words, where hidden Markov models (HMMs) and dynamic Bayesian networks (DBNs) were evaluated. The work proposed is divided into three stages, namely preprocessing, feature extraction and classification. Preprocessing includes baseline estimation and normalization as well as segmentation. In the second stage, features are extracted from each of the normalized words, where a set of new features for handwritten Arabic words is proposed, based on a sliding window approach moving across the mirrored word image. The third stage is for classification and recognition, where machine learning is applied using HMMs and DBNs. In order to validate the techniques, extensive experiments were conducted using the IFN/ENIT database which contains 32,492 Arabic words. Experimental results and quantitative evaluations showed that HMM outperforms DBN in terms of higher recognition rate and lower complexity. 

AB - This paper presents a comparative study of two machine learning techniques for recognizing handwritten Arabic words, where hidden Markov models (HMMs) and dynamic Bayesian networks (DBNs) were evaluated. The work proposed is divided into three stages, namely preprocessing, feature extraction and classification. Preprocessing includes baseline estimation and normalization as well as segmentation. In the second stage, features are extracted from each of the normalized words, where a set of new features for handwritten Arabic words is proposed, based on a sliding window approach moving across the mirrored word image. The third stage is for classification and recognition, where machine learning is applied using HMMs and DBNs. In order to validate the techniques, extensive experiments were conducted using the IFN/ENIT database which contains 32,492 Arabic words. Experimental results and quantitative evaluations showed that HMM outperforms DBN in terms of higher recognition rate and lower complexity. 

KW - off-line handwritten recognition

KW - hidden markov model (HMM)

KW - dynamic bayesian network (DBN)

KW - performance evaluation

KW - IFN/ENIT database

UR - http://www.scopus.com/inward/record.url?scp=79955006396&partnerID=8YFLogxK

U2 - 10.1016/j.knosys.2011.02.008

DO - 10.1016/j.knosys.2011.02.008

M3 - Article

VL - 24

SP - 680

EP - 688

JO - Knowledge Based Systems

T2 - Knowledge Based Systems

JF - Knowledge Based Systems

SN - 0950-7051

IS - 5

ER -