Rule Based Neural Networks Construction for Handwritten Arabic City-Names Recognition

  • Labiba Souici
  • Nadir Farah
  • Toufik Sari
  • Mokhtar Sellami
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3192)


A recent innovation in artificial intelligence research has been the integration of multiple techniques into hybrid systems. These systems seek to overcome the deficiencies of traditional artificial techniques by combining techniques with complementary capabilities. At the crossroads of symbolic and neural processing, researchers have been actively investigating the synergies that might be obtained from combining the strengths of these two paradigms. In this article, we deal with a knowledge based artificial neural network for handwritten Arabic city-names recognition. We start with words perceptual features analysis in order to construct a hierarchical knowledge base reflecting words description. A translation algorithm then converts the symbolic representation into a neural network, which is empirically trained to overcome the handwriting variability.


Hide Markov Model Postal Address Artificial Intelligence Research Translation Algorithm Arabic Script 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Amin, A.: Off-line arabic character recognition: The state of the art. Pattern Recognition 31(5), 517–530 (1998)CrossRefMathSciNetGoogle Scholar
  2. 2.
    Bennasri, A., Zahour, A., Taconet, B.: Arabic script preprocessing and application to postal addresses. In: ACIDCA 2000, Vision & Pattern Recognition, Monastir, Tunisia, pp. 74–79 (2000)Google Scholar
  3. 3.
    Bertille, J.M., Gilloux, M., El Yacoubi, A.: Localisation et reconnaissance conjointes de noms de voies dans les lignes distribution des adresses postales. SRTP/RD/Traitement automatique ligne distribution, Transition (7), 16–25 (1994)Google Scholar
  4. 4.
    Bishop, C.M.: Neural networks for pattern recognition. Clarendon Press, Oxford (1995)Google Scholar
  5. 5.
    Côté, M., Lecolinet, E., Suen, C.Y.: Automatic reading of cursive scripts using a reading model and perceptual concepts: The percepto system. IJDAR: International Journal on Document Analysis and Recognition 1, 3–17 (1998)CrossRefGoogle Scholar
  6. 6.
    Dowton, A.C., Leedham, C.G.: Preprocessing and presorting of envelope images for automatic sorting using OCR. Pattern Recognition 23(3/4), 347–362 (1990)CrossRefGoogle Scholar
  7. 7.
    Hilario, M.: An overview of strategies for neurosymbolic integration. In: Sun, R., Alexandre, F. (eds.) Connectionist-Symbolic Integration: From Unified to Hybrid Approaches, ch. 2, Lawrence Erlbaum Associates, Mahwah (1997)Google Scholar
  8. 8.
    Madhvanath, S., Govindaraju, V.: The role of holistic paradigms in handwritten word recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 23(2) (February 2001)Google Scholar
  9. 9.
    Mahadevan, U., Srihari, S.N.: Parsing and recognition of city, state and ZIPcodes in handwritten addresses. In: ICDAR 1999, pp. 325–328 (1999)Google Scholar
  10. 10.
    Otsu, N.: A threshold selection method from gray-level histogram. IEEE Transaction on image processing, 62–66 (1979)Google Scholar
  11. 11.
    Pavlidis, T.: Algorithms for Graphic and Image Processing. Computer science press, Rockville (1982)Google Scholar
  12. 12.
    Plamondon, R., Srihari, S.N.: On-line and off-line handwritten recognition: a comprehensive survey. IEEE Trans. on PAMI 22(22), 63–84 (2000)Google Scholar
  13. 13.
    Sari, T., Souici, L., Sellami, M.: Off-line Handwritten Arabic Character Segmentation and Recognition System: ACSA-RECAM. In: IWFHR 2002, 8th International Workshop on Frontiers in Handwriting Recognition, Niagara-on-the-Lake, Ontario, Canada (August 2002)Google Scholar
  14. 14.
    Souici, L., Aoun, A., Sellami, M.: Global recognition system for Arabic literal amounts. International Conference on Computer Technologies and Applications. In: ICCTA 1999, Alexandria, Egypt (August 1999)Google Scholar
  15. 15.
    Souici-Meslati, L., Rahim, H., Zemehri, M.C., Sellami, M.: Système Connexionniste à Représentation Locale pour la Reconnaissance de Montants Littéraux Arabes. In: CIFED 2002, Conférence Internationale Francophone sur l’Ecrit et le Document, Hammamet, Tunisie (October 2002)Google Scholar
  16. 16.
    Srihari, S.N.: Recognition of handwritten and machine printed text for postal address interpretation. Pattern Recognition Letters: Special issue on postal processing and character recognition 14(4), 291–302 (1993)Google Scholar
  17. 17.
    Srihari, S.: Handwritten address interpretation: a task of many pattern recognition problems. Int. Journal of Pattern Recognition and Artificial Intelligence 14, 663–674 (2000)CrossRefGoogle Scholar
  18. 18.
    Towell, G.G., Shavlic, J.W.: Knowledge-based artificial neural networks. Artificial Intelligence 70, 119–165 (1994)zbMATHCrossRefGoogle Scholar
  19. 19.
    Towell, G.G.: Symbolic knowledge and neural networks: insertion, refinement and extraction. PhD thesis, University of Wisconsin, Madison, WI (1991)Google Scholar
  20. 20.
    Wermter, S., Sun, R.: Hybrid neural systems. Springer, Heidelberg (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Labiba Souici
    • 1
  • Nadir Farah
    • 1
  • Toufik Sari
    • 1
  • Mokhtar Sellami
    • 1
  1. 1.Departement d’informatique, Laboratoire LRIUniversity Badji MokhtarAnnabaALGERIA

Personalised recommendations