Multi-level arabic handwritten words recognition

  • H. Miled
  • M. Cheriet
  • C. Olivier
Poster Papers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1451)


In this paper, we present a strategy of Arabic words recognition by combining two levels which are based on global and analytical approaches according to the topological properties of Arabic handwriting. In the first level (global), we consider the visual indices which can be generated by: diacritics and strokes (denoted tracing) that form the main shapes of the word. Each word is described as a sequence of visual indices which is treated by a “global” classifier based on Hidden Markov Model (HMM). In the second level, the word is segmented into graphemes, then each grapheme is transformed into a HMM observation by a vector quantization phase. An analytical HMM is developed in order to manage the observation sequences. At this level the diacritics are not taken in consideration which allows to reduce the number of estimated character models. Finally we combine the two approaches to decide on the class of an unknown word. In fact, the global model serves as a filter. It produces a set of hypotheses to the analytical model, which in turns, defines and outputs the final decision.

Key words

Arabic handwriting recognition character and cursive scripts recognition visual indices HMM modeling classifier combination hybrid approach 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Fayol, J. E. Gombert, P. Lecocq, L. Sprenger-Charolles, D. Zager, “Psychologie Cognitive de la Lecture”, Presses Universitaires de France, 1992.Google Scholar
  2. [2]
    A. Amin, “Off Line Arabic Character Recognition-A survey”, In Proc. of ICDAR'97, Ulm, Germany, pp 596–599, Aug. 1997.Google Scholar
  3. [3]
    B. El-Badr, S. A. Mahmoud, “A Survey and Bibliographies of Arabic optical text recognition”, Signal Processing, vol. 41, pp. 49–76, 1995.CrossRefGoogle Scholar
  4. [4]
    T. K. Ho, J. J. Hull, S. N. Srihari, “Decision Combination in Multiple Classifier Systems”, IEEE PAMI, vol. 16, no. l, Jan. 1994.Google Scholar
  5. [5]
    L. Xu, A. Krzyzak, C. Y. Suen, “Methods of Combining Multiple Classifiers and Their Applications to Handwriting Recognition”, IEEE Transactions on Systems, Man, and Cybernetics, vol. 22, no. 3, pp 418–435, 1992.Google Scholar
  6. [6]
    R. K. Pawalka, N. Sherkat, R. J. Whitsow, “Recognizer characterisation for combining handwriting recognition results at word level”, In Proc. of ICDAR'95, Montréal, Canada, pp 68–73, Aug. 1995.Google Scholar
  7. [7]
    B. Plessis, A. Sicsu, L. Heutte, E. Menu, E. Lecolinet, O. Debon, J. V. Moreau, “A Multi-class ifier combination strategy for recognition of handwritten cursive words”, In Proc. of ICDAR'93, Tsukuba Science City, Japan, pp 642–645, Aug. 1997.Google Scholar
  8. [8]
    A. Ameur, K. Romeo, H. Miled, M. Cheriet, “Approche Globale pour la Reconnaissance des Mots Manuscrits Arabes”, Proc. of CNED'94, Rouen, France, pp 151–157, Juillet 1994.Google Scholar
  9. [9]
    J. C. Simon, 0. Baret, “Handwriting recognition as an application of regularities and singularities in line pictures”, Proc. of IWFHR, Montréal, Canada, pp 23–36, 1990.Google Scholar
  10. [10]
    M. Cheriet, H. Miled, C. Olivier: “Visual Aspect of Cursive Arabic Handwriting Recognition”, Visual Interface, Vancouver, Canada, Aug. 1998.Google Scholar
  11. [11]
    L. R. Rabiner, “A tutorial on hidden Markov models and selected applications in speech recognition”, Proc. of IEEE, vol. 77, no. 2, pp 257–285, 1989.CrossRefGoogle Scholar
  12. [12]
    H. Bunke, M. Roth and E. G. Schukat-Talamazzini, ”Off-line cursive handwriting recognition using Hidden Markov Models”, Pattern Recognition, vol. 28, no. 9, pp 1399–1413, 1995.CrossRefGoogle Scholar
  13. [13]
    N. Ben Amara, A. Belaid, ”Printed PAW Recognition Based on Planar Hidden Markov Models”, Proc. of ICPR'96, Vienna, Autria, vol. 2, pp 220–224, Aug. 1996.Google Scholar
  14. [14]
    C. Olivier, T. Paquet, M. Avila, Y. Lecourtier, ”Optimal Order of Markov Models applied to Bankchecks”, Inter. Journal of Pattern Recognition and Artificial Intelligence, vol. 11, no. 5, pp 789–800, 1997.CrossRefGoogle Scholar
  15. [15]
    C. Olivier, H. Miled, K. Romeo, Y. Lecourtier, ”Segmentation and Coding of Arabic Handwritten Words”, Proc. of ICPR'96, Vienna, Austria, vol. 3, pp 264–268, Aug. 1996.Google Scholar
  16. [16]
    H. Miled, C. Olivier, M. Cheriet, Y. Lecourtier, ”Coupling Observation/Letter for a Markovian Modelisation Applied to the Recognition of the Arabic Handwriting”, Proc. of lCDAR'97, Ulm, Germany, pp. 580–583, Aug. 1997.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • H. Miled
    • 1
    • 2
  • M. Cheriet
    • 2
  • C. Olivier
    • 1
    • 3
  1. 1.PSI-La3i, UFR des SciencesUniversité de RouenFrance
  2. 2.LIVIA, École de Technologie SupérieureMontrealCanada
  3. 3.SIC-IRCOM, UMR CNRS 6615 Université de PoitiersFrance

Personalised recommendations