Volume 12, Number 1, September 2016 - DOI: http://dx.doi.org/10.21700/ijcis.2016.273


Computing and Information Sciences is a peer reviewed journal that is committed to timely publication of original research, surveying and tutorial contributions on the analysis and development of computing and information science. The journal is designed mainly to serve researchers and developers, dealing with information and computing. Papers that can provide both theoretical analysis, along with carefully designed computational experiments, are particularly welcome. The journal is published 2-3 times per year with distribution to librarians, universities, research centers, researchers in computing, mathematics, and information science. The journal maintains strict refereeing procedures through its editorial policies in order to publish papers of only the highest quality. The refereeing is done by anonymous Reviewers. Often, reviews take four months to six months to obtain, occasionally longer, and it takes an additional several months for the publication process.

DOI: http://dx.doi.org/10.21700/ijcis.2016.103

Word Extraction and Recognition in Arabic Handwritten Text

Nabil Aouadi
Afef Kacem Echi* email: afef.kacem@esstt.rnu.tn 

LaTICE Laboratory, University of Tunis Avenue Taha Hussein Montfleury, 1008 Tunis, Tunisia

*Corresponding author.

Received: 10 June 2016
Revised: 25 June 2016
Accepted: 25 August 2016
Published: 26 September 2016

Abstract: Segmenting arabic manuscripts into text-lines and words is an important step to make recognition systems more efficient and accurate. The major problem making this task crucial is the word extraction process: first, words are often a succession of sub-words where the space value between these sub-words do not respect any rules. Second, the presence of connections even between non adjacent sub-words in the same text-line, makes word's parts identification and the entire word extraction difficult. This work proposes an automatic system for arabic handwritten word extraction and recognition based on 1) localizing and segmenting touching characters, 2) extracting real sub-words and structural features from word images and 3) recognizing them by a Markovian classifier. The performance of the proposed system is tested using samples extracted from historical handwritten documents. The obtained results are encouraging. We achieved an average rate of recognition of 87%.

Keywords: Arabic handwriting recognition; Touching letters; Text-line Segmentation; Word Segmentation; Structural Feature Extraction; Word Recognition; Hidden Markov Model.

  • PDF (557 KB)
  • ZIP (535 KB)



    Prof. Jihad Mohamad Alja'am 
    Email: editor.ijcis@gmail.com 

    The Journal Secretary
    Eng. Dana Bandok
    Ontario, Canada 
    Email: sec.ijcis@gmail.com 

    Home Page »