Difficult Cases in Handwritten Numeral Recognition

  • Raymond Legault
  • Ching Y. Suen
  • Christine Nadal

Abstract

One avenue to improve the performance of computer systems for the recognition of handwritten data is to develop a better grasp of human expertise in this area. Through an experiment, we investigate the clues and cues used by humans to recognize the most confusing samples of our database. Four algorithms are also applied, individually and in a combined manner, to the same difficult data set. Machine and human performances are compared. Such knowledge should be used to refine the existing algorithms which are proved to be clearly inferior to the human experts in recognition of unconstrained handwritten numerals.

Keywords

Hull 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [Beu73]
    M. Beun. A flexible method for automatic reading of handwritten numerals. Philips Tech. Rev., 33: 89–101, 130–137, 1973.Google Scholar
  2. [HSC+88]
    J. J. Hull, S. N. Srihari, E. Cohen, C. L. Kuan, P. Cullen, and P. Palumbo. A blackboard-based approach to handwritten zip code recognition. In Proc. U.S. Postal Service Advanced Technology Conf. pages 1018–1032, 1988.Google Scholar
  3. [KDS90]
    A. Krzyzak, W. Dai, and C. Y. Suen. Unconstrained handwritten character classification using modified backpropagation model. In Proc. Int. Workshop on Frontiers in Handwriting Recognition, pages 155–166, Concordia University, Montreal, April 1990.Google Scholar
  4. [KS88]
    C. L. Kuan and S. N. Srihari. A stroke-based approach to handwritten numeral recognition. In Proc. U.S. Postal Service Advanced Technology Conf., pages 1033–1041, 1988.Google Scholar
  5. [LBD+90]
    Yan Le Cun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, L. D. Jackel, and H. S. Baird. Constrained neural network for unconstrained handwritten digit recognition. In Proc. Int. Workshop on Frontiers in Handwriting Recognition, pages 145–154, Concordia University, Montreal, April 1990.Google Scholar
  6. [LS88]
    L. Lam and C. Y. Suen. Structural classification and relaxation matching of totally unconstrained handwritten zip-code numbers. Pattern Recognition, 21 (1): 19–31, 1988.CrossRefGoogle Scholar
  7. [MS90]
    T. Mai and C. Y. Suen. A generalized knowledge-based system for the recognition of unconstrained handwritten numerals. IEEE Trans. Syst. Cybern., 20 (4): 835–848, 1990.CrossRefGoogle Scholar
  8. [NW60]
    U. Neisser and P. Weene. A note on human recognition of handprinted characters. Information and Control, 3: 191–196, 1960.CrossRefGoogle Scholar
  9. [SNM+90]
    C. Y. Suen, C. Nadal, T. A. Mai, R. Legault, and L. Lam. Recognition of handwritten numerals based on the concept of multiple experts. In Proc. Int. Workshop on Frontiers in Handwriting Recognition, pages 131–144, Concordia University, Montreal, April 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Raymond Legault
    • 1
  • Ching Y. Suen
    • 1
  • Christine Nadal
    • 1
  1. 1.Centre for Pattern Recognition and Machine IntelligenceConcordia UniversityMontrealCanada

Personalised recommendations