Arabic text recognition using steady state genetic algorithm

Abdullah M. Al-Mutawa, Chilukuri K Mohan

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Text recognition is a challenging task when text contains overlapping and cursive characters. This paper addresses the task of matching input Arabic text with character forms in the database described in terms of features. We introduce new features that approximate character shapes, designed to make the recognition process independent of font size. We apply steady state genetic algorithms to the partial shape-matching task, with extremely encouraging recognition rate. No separate segmentation is required in our approach.

Original languageEnglish (US)
Title of host publicationIntelligent Engineering Systems Through Artificial Neural Networks
Pages311-316
Number of pages6
Volume1998
StatePublished - 1998

Fingerprint

Genetic algorithms

Keywords

  • Character Recognition
  • Edge Detection
  • Feature Extraction
  • Genetic Algorithms

ASJC Scopus subject areas

  • Software

Cite this

Al-Mutawa, A. M., & Mohan, C. K. (1998). Arabic text recognition using steady state genetic algorithm. In Intelligent Engineering Systems Through Artificial Neural Networks (Vol. 1998, pp. 311-316)

Arabic text recognition using steady state genetic algorithm. / Al-Mutawa, Abdullah M.; Mohan, Chilukuri K.

Intelligent Engineering Systems Through Artificial Neural Networks. Vol. 1998 1998. p. 311-316.

Research output: Chapter in Book/Report/Conference proceedingChapter

Al-Mutawa, AM & Mohan, CK 1998, Arabic text recognition using steady state genetic algorithm. in Intelligent Engineering Systems Through Artificial Neural Networks. vol. 1998, pp. 311-316.
Al-Mutawa AM, Mohan CK. Arabic text recognition using steady state genetic algorithm. In Intelligent Engineering Systems Through Artificial Neural Networks. Vol. 1998. 1998. p. 311-316
Al-Mutawa, Abdullah M. ; Mohan, Chilukuri K. / Arabic text recognition using steady state genetic algorithm. Intelligent Engineering Systems Through Artificial Neural Networks. Vol. 1998 1998. pp. 311-316
@inbook{9e209325f3874671807706bf34cb7451,
title = "Arabic text recognition using steady state genetic algorithm",
abstract = "Text recognition is a challenging task when text contains overlapping and cursive characters. This paper addresses the task of matching input Arabic text with character forms in the database described in terms of features. We introduce new features that approximate character shapes, designed to make the recognition process independent of font size. We apply steady state genetic algorithms to the partial shape-matching task, with extremely encouraging recognition rate. No separate segmentation is required in our approach.",
keywords = "Character Recognition, Edge Detection, Feature Extraction, Genetic Algorithms",
author = "Al-Mutawa, {Abdullah M.} and Mohan, {Chilukuri K}",
year = "1998",
language = "English (US)",
volume = "1998",
pages = "311--316",
booktitle = "Intelligent Engineering Systems Through Artificial Neural Networks",

}

TY - CHAP

T1 - Arabic text recognition using steady state genetic algorithm

AU - Al-Mutawa, Abdullah M.

AU - Mohan, Chilukuri K

PY - 1998

Y1 - 1998

N2 - Text recognition is a challenging task when text contains overlapping and cursive characters. This paper addresses the task of matching input Arabic text with character forms in the database described in terms of features. We introduce new features that approximate character shapes, designed to make the recognition process independent of font size. We apply steady state genetic algorithms to the partial shape-matching task, with extremely encouraging recognition rate. No separate segmentation is required in our approach.

AB - Text recognition is a challenging task when text contains overlapping and cursive characters. This paper addresses the task of matching input Arabic text with character forms in the database described in terms of features. We introduce new features that approximate character shapes, designed to make the recognition process independent of font size. We apply steady state genetic algorithms to the partial shape-matching task, with extremely encouraging recognition rate. No separate segmentation is required in our approach.

KW - Character Recognition

KW - Edge Detection

KW - Feature Extraction

KW - Genetic Algorithms

UR - http://www.scopus.com/inward/record.url?scp=0032291748&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032291748&partnerID=8YFLogxK

M3 - Chapter

VL - 1998

SP - 311

EP - 316

BT - Intelligent Engineering Systems Through Artificial Neural Networks

ER -