Advertisement

A text recognition system based on a Neural Network and on a Deformed System

  • Javier Echanobe
  • José R. González de Mendívil
  • José R. Garitagoitia
Neural Networks for Perception
Part of the Lecture Notes in Computer Science book series (LNCS, volume 930)

Abstract

This paper shows a text recognition system based on a Neural Network which is used as Isolated Character Classifier (ICC), and on a Deformed System that incorporates the contextual knowledge defined by a dictionary. The Neural Network provides for every input character a fuzzy character built up with the ouput unit values. The fuzzy characters are the inputs for the Deformed System which is defined as an automaton representing the dictionary and whose behaviour is fuzzily constrained by fuzzy inputs. Therefore the classification and contextual processes are computed together. Experimental results show good performance for the system.

Keywords

Neural Network Output Unit Composition Function Change Error Fuzzy Input 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    J. J. Hull, S.N. Srihari and R. Choudhari, “An Integrated Algorithm for Text Recognition: Comparison with a Cascaded Algorithm”, IEEE Trans. Pattern Analysis Mach. Intell., vol. PAMI-5, no. 4, pp. 384–395, 1983.Google Scholar
  2. [2]
    D. G. Elliman and I.T. Lancaster, “A Review of Segmentation and Contextual Analysis Techniques for Text Recognition”, Pattern Recognition, vol. 23, no. 3/4, pp. 337–346, 1990.Google Scholar
  3. [3]
    R. Shinghal and G.T. Toussaint, “Experiments in Text Recognition with the Modified Viterbi Algorithm”, IEEE Trans. Pattern Analysis Mach. Intell., vol. PAMI-1, no. 2, pp. 184–193, 1979.Google Scholar
  4. [4]
    G.M. Landau, “Fast string matching with k differences”, J. Comput. Syst. Sci., vol. 37, pp. 63–78, 1988.Google Scholar
  5. [5]
    R. Shinghal, “A Hybrid Algorithm for Contextual Text Recognition”, Pattern Recognition, vol. 16, no. 2, pp. 184–193, 1983.Google Scholar
  6. [6]
    C.V. Negoita, D.A. Ralescu, Application of Fuzzy Sets to System Analysis, Birkaeuser, Basilea 1975.Google Scholar
  7. [7]
    R. Reina, José R. González de Mendívil, José R. Garitagoitia, “Improved Character Recognition System based on a Neural Network incorporating the context via Fuzzy Automata” 2nd International Conference on Fuzzy Logic & Neural Networks, Izuka (Japan), July 17–22, 1992.Google Scholar
  8. [8]
    J. Echanove, R. Reina, J.R. Garitagoitia and J.R. González de Mendívil, “Deformed Systems in Text Recognition”, International Conference On Artificial Neural Networks., Sorrento (Italy), May 26–29, 1994.Google Scholar
  9. [9]
    T.M. Cover, P.E. Hart, “Nearest Neighbor Pattern Classification”, IEEE Transactions on Information Theory, IT-13, January 1967, 21–27.Google Scholar
  10. [10]
    R. O. Duda, P.E. Hart, “Pattern Classification and Scene Analysis”, Addison-Wesley, New York, 1973.Google Scholar
  11. [11]
    K.Fukushima, S. Miyake, T. Ito, “Neocognitron: A Neural network for a Mechanism of Visual Pattern Recognition” IEEE Trans. Sys. Man and Cyber., SMC-13, no. 5, pp. 826–834, 1983.Google Scholar
  12. [12]
    J. Hopcroft, J. Ullman, Introduction to Automata Theory, Languages and Computation, Addison-Wesley Publishing Company, Reading Massachusetts, 1979.Google Scholar
  13. [13]
    E. Vidal, F. Casacuberta, E. Sanchís, J. M. Benedí, “A General Fuzzy-Parsing Scheme for Speech Recognition”, NATO ASI series, vol. F16, pp. 427–446, 1985.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Javier Echanobe
    • 1
  • José R. González de Mendívil
    • 1
  • José R. Garitagoitia
    • 1
  1. 1.Department of Electricity and ElectronicsUniversity of the Basque CountryBilbaoSpain

Personalised recommendations