Regularities and Singularities in Line Pictures

  • Jean-Claude Simon
  • Olivier Baret


The main difficulty of Pattern Recognition is the interpretation of the iconic picture, the pixels, into primitive features. The interest of a feature classification as “regular” and “singular” is recalled and justified. These ideas are applied to line images, and particularly to handwritten words. On the graph representation of a word, a regular feature, the “axis,” is found; it allows to segment the word representation and to obtain a descriptive chain. Examples of such symbolic descriptions are given, with their interpretation as a word among a list of 25 words. The results are of the order of 87% success, 3% substitution on one handwriting. Preliminary results show that it may be extended to large categories of handwriting.


Black Pixel Line Image Regular Feature Binary File Candidate Word 
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]
    J. C. Simon, Pattern and Operators. MacGraw-Hill in North America, North Oxford Acad, in Europe, (1986).Google Scholar
  2. [2]
    M. R. Garey and D. S. Johnson, Computers and Intractability. W. H. Freeman, San Francisco (1979).Google Scholar
  3. [3]
    J. C. Simon (ed.), From Pixels to Features, North-Holland, Amsterdam (1989).MATHGoogle Scholar
  4. [4]
    T. Pavlidis, “Residual Analysis for feature extraction,” pp. 219–227, in [3] (1989).Google Scholar
  5. [5]
    J. C. Simon, “Invariance in P. R., Application to line images,” Image and Vision Comp. Jr. 4, 1, pp. 11–23 (1986).Google Scholar
  6. [6]
    J. C. Simon, “A complemental approach to feature detection,” pp. 229–236 I n [3] (1989).Google Scholar
  7. [7]
    J. Serra, Image Analysis and Mathematical Morphology, Vol. II, Academic Press, Chaps. 11–13 (1988).Google Scholar
  8. [8]
    K. L. Heng, “Analyse d’une image de ligne: des pixels aux primitives.” Thèse Univ P. et M. Curie, May 1986.Google Scholar
  9. [9]
    J. M. Lery, “Reconnaissance de lécriture manuscrite monoscripteur: description symbolique des mots par un graphe de primitives.” Thèse Univ. P. et M. Curie, June 1989.Google Scholar
  10. [10]
    A. Leturcq, “La reconnaissance des caractères manuscrits sans apprentissage.” Thèse Univ. P et M. Curie, March 1990.Google Scholar
  11. [11]
    J. C. Simon and O. Baret, “Formes régulières, formes singulières, application à la reconnaissance de l’ éricriture manuscrite.”C. R. Acad. Sc. Paris, T.309, série II, 19, pp. 1901–1906, December 1989.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Jean-Claude Simon
    • 1
  • Olivier Baret
    • 1
  1. 1.Laforia / CNRS and ENPCParisFrance

Personalised recommendations