Classifiers Combination for Arabic Words Recognition: Application to Handwritten Algerian City Names

  • Soulef Nemouchi
  • Labiba Souici Meslati
  • Nadir Farah
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7340)


In this paper, we present a global recognition system for Arabic handwritten words; we focus on the two phases of feature extraction and classification. In our system, we have retained three feature sets. The Zernike moments and the structural features of the word are extracted from the binary image, the Freeman code is established from the contour image of the word and the zoning is given from the skeleton image. These features, representing the words, are extracted to be used as input, in an individual or combined way, of the four classifiers used in our system: the Fuzzy C-Means algorithm (FCM), the K-Means algorithm, the K Nearest Neighbor algorithm (KNN) and a Probabilistic Neural Network (PNN). The system architecture is a parallel one where each expert (classifier) gives his point of view and we combine the results to make a final decision. The classifier results are combined using two methods: the simple vote and the weighted sum.


Arabic handwriting recognition Fuzzy C-Means (FCM) K Nearest Neighbor algorithm (KNN) K-Means algorithm Probabilistic Neural Network (PNN) zernike moments zoning Freeman chain code 


  1. 1.
    Bunke, H.: Recognition of Cursive Roman Handwriting - Past, Present and Future. In: International Conference on Document Analysis and Recognition, ICDAR, Edinburgh, Scotland (2003)Google Scholar
  2. 2.
    Steinherz, T., Rivlin, E., Intrator, N.: Offline cursive script recognition: a survey. IJDAR, International Journal on Document Analysis and Recognition 2, 90–110 (1999)CrossRefGoogle Scholar
  3. 3.
    Essoukhri Ben Amara, N., Bouslama, F.: Classification of Arabic script using multiple sources of information: state of the art and perspectives. IJDAR, International Journal on Document Analysis and Recognition 5, 195–212 (2003)CrossRefGoogle Scholar
  4. 4.
    Lorigo, L.M., Govindaraju, V.: Offline Arabic Handwriting Recognition: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 28(5), 712–724 (2006)CrossRefGoogle Scholar
  5. 5.
    Souici, L., Aoun, A., Sellami, M.: Global recognition system for Arabic literal amounts. In: ICCTA 1999, Alexandria, Egypt (1999)Google Scholar
  6. 6.
    Souici-Meslati, L., Sellami, M.: A hybrid approach for Arabic literal amounts recognition. AJSE 29(2B), 177–194 (2004)Google Scholar
  7. 7.
    Farah, N., Ennaji, A., Khadir, T., Sellami, M.: Benefit of multiclassifier systems for Arabic handwritten words recognition. In: International Conference on Document Analysis and Recognition, ICDAR, Seoul, Korea, vol. 1, pp. 222–226 (2005)Google Scholar
  8. 8.
    Farah, N., Souici, L., Sellami, M.: Classifiers combination and syntax analysis for Arabic literal amount recognition. Engineering Applications of Artificial Intelligence 19(1), 29–39 (2006)CrossRefGoogle Scholar
  9. 9.
    Vinciarelli, A.: A survey one off-line cursive Word recognition. Pattern Recognition 35(7), 1433–1446 (2002)zbMATHCrossRefGoogle Scholar
  10. 10.
    Arrivault, D.: Apport des graphes dans la reconnaissance non-contrainte de caractères manuscrits anciens. PhD Thesis, University of Poitiers, France (2002) (in French)Google Scholar
  11. 11.
    Al-Ohali, Y.: Handwritten Word Recognition – Application to Arabic Cheque Processing. PhD Thesis, Concordia University, Montreal, Canada (2002)Google Scholar
  12. 12.
    Azizi, N., Farah, N., Sellami, M.: Off-line handwriting word recognition using ensemble of classifier selection and features fusion. JATIT, Journal of Theoretical and Applied Information Technology 14(2), 141–150 (2010)Google Scholar
  13. 13.
    Zouari, H., Heutte, L., Lecourtier, Y., Alimi, A.: Building Diverse Classifier Outputs to Evaluate the Behavior of Combination Methods: The Case of Two Classifiers. In: Roli, F., Kittler, J., Windeatt, T. (eds.) MCS 2004. LNCS, vol. 3077, pp. 273–282. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  14. 14.
    Souici-Meslati, L., Sellami, M.: Toward a generalization of neuro-symbolic recognition: an application to arabic words. KES, International Journal of Knowledge-Based and Intelligent Engineering Systems 10(5), 347–361 (2006)Google Scholar
  15. 15.
    Prema, K.V., Subba Reddy, N.V.: Two-tier architecture for unconstrained handwritten character recognition. Sadhana 27, Part 5, 585–594 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Soulef Nemouchi
    • 1
  • Labiba Souici Meslati
    • 2
  • Nadir Farah
    • 3
  1. 1.Commerciales et Sciences de GestionEPSECG, Ecole Préparatoire des Sciences EconomiquesAnnabaAlgeria
  2. 2.LRI LaboratoryBadji Mokhtar - Annaba UniversityAlgeria
  3. 3.LabGED LaboratoryBadji Mokhtar - Annaba UniversityAlgeria

Personalised recommendations