Geometrical-matrix feature extraction for on-line handwritten characters recognition

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Most of Arabic handwriting recognition in the literature has focused only on recognizing offline script, and few of research take online case. So it's still remains as an active area of research. however there is a lack of studies in terms of recognizing Arab characters, especially on the online cases. The process of handwriting recognition faces a lot of challenges; feature extraction is the most important problem in character recognition. The main theme of this paper is new feature extraction method employed in online Arabic character recognition. An Arabic character recognition handwritten system cannot be successful, without using suitable feature extraction methods. In this work we have proposed the hybrid Edge Direction Matrixes and geometrical feature extraction method for on-line handwritten Arabic character recognition system. In addition, horizontal and vertical projection profile, and Laplacian filter have been used in the preprocessing phase. The training and testing of the online handwriting recognition system was conducted using our dataset; we have used 840characters from different writers, 504 characters for training, and 336 characters for testing. The evaluation was conducted on state of the art methods in the classification phase. The results have revealed that the proposed method gives best recognition rate for character category.

Original languageEnglish
Pages (from-to)86-93
Number of pages8
JournalJournal of Theoretical and Applied Information Technology
Volume49
Issue number1
Publication statusPublished - 2013

Fingerprint

Character recognition
Character Recognition
Feature Extraction
Feature extraction
Handwriting Recognition
Testing
Face recognition
Preprocessing
Horizontal
Vertical
Projection
Filter
Character
Evaluation

Keywords

  • Arabic character
  • Classification
  • Edge direction matrixes
  • Geometrical feature
  • Online recognition

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

@article{bd174a02df864da4b610e04b2aa4e5ab,
title = "Geometrical-matrix feature extraction for on-line handwritten characters recognition",
abstract = "Most of Arabic handwriting recognition in the literature has focused only on recognizing offline script, and few of research take online case. So it's still remains as an active area of research. however there is a lack of studies in terms of recognizing Arab characters, especially on the online cases. The process of handwriting recognition faces a lot of challenges; feature extraction is the most important problem in character recognition. The main theme of this paper is new feature extraction method employed in online Arabic character recognition. An Arabic character recognition handwritten system cannot be successful, without using suitable feature extraction methods. In this work we have proposed the hybrid Edge Direction Matrixes and geometrical feature extraction method for on-line handwritten Arabic character recognition system. In addition, horizontal and vertical projection profile, and Laplacian filter have been used in the preprocessing phase. The training and testing of the online handwriting recognition system was conducted using our dataset; we have used 840characters from different writers, 504 characters for training, and 336 characters for testing. The evaluation was conducted on state of the art methods in the classification phase. The results have revealed that the proposed method gives best recognition rate for character category.",
keywords = "Arabic character, Classification, Edge direction matrixes, Geometrical feature, Online recognition",
author = "Ismail, {Saad M.} and {Sheikh Abdullah}, {Siti Norul Huda}",
year = "2013",
language = "English",
volume = "49",
pages = "86--93",
journal = "Journal of Theoretical and Applied Information Technology",
issn = "1992-8645",
publisher = "Asian Research Publishing Network (ARPN)",
number = "1",

}

TY - JOUR

T1 - Geometrical-matrix feature extraction for on-line handwritten characters recognition

AU - Ismail, Saad M.

AU - Sheikh Abdullah, Siti Norul Huda

PY - 2013

Y1 - 2013

N2 - Most of Arabic handwriting recognition in the literature has focused only on recognizing offline script, and few of research take online case. So it's still remains as an active area of research. however there is a lack of studies in terms of recognizing Arab characters, especially on the online cases. The process of handwriting recognition faces a lot of challenges; feature extraction is the most important problem in character recognition. The main theme of this paper is new feature extraction method employed in online Arabic character recognition. An Arabic character recognition handwritten system cannot be successful, without using suitable feature extraction methods. In this work we have proposed the hybrid Edge Direction Matrixes and geometrical feature extraction method for on-line handwritten Arabic character recognition system. In addition, horizontal and vertical projection profile, and Laplacian filter have been used in the preprocessing phase. The training and testing of the online handwriting recognition system was conducted using our dataset; we have used 840characters from different writers, 504 characters for training, and 336 characters for testing. The evaluation was conducted on state of the art methods in the classification phase. The results have revealed that the proposed method gives best recognition rate for character category.

AB - Most of Arabic handwriting recognition in the literature has focused only on recognizing offline script, and few of research take online case. So it's still remains as an active area of research. however there is a lack of studies in terms of recognizing Arab characters, especially on the online cases. The process of handwriting recognition faces a lot of challenges; feature extraction is the most important problem in character recognition. The main theme of this paper is new feature extraction method employed in online Arabic character recognition. An Arabic character recognition handwritten system cannot be successful, without using suitable feature extraction methods. In this work we have proposed the hybrid Edge Direction Matrixes and geometrical feature extraction method for on-line handwritten Arabic character recognition system. In addition, horizontal and vertical projection profile, and Laplacian filter have been used in the preprocessing phase. The training and testing of the online handwriting recognition system was conducted using our dataset; we have used 840characters from different writers, 504 characters for training, and 336 characters for testing. The evaluation was conducted on state of the art methods in the classification phase. The results have revealed that the proposed method gives best recognition rate for character category.

KW - Arabic character

KW - Classification

KW - Edge direction matrixes

KW - Geometrical feature

KW - Online recognition

UR - http://www.scopus.com/inward/record.url?scp=84876069878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876069878&partnerID=8YFLogxK

M3 - Article

VL - 49

SP - 86

EP - 93

JO - Journal of Theoretical and Applied Information Technology

JF - Journal of Theoretical and Applied Information Technology

SN - 1992-8645

IS - 1

ER -