Skeletonization of Arabic characters using clustering based skeletonization algorithm (CBSA)

https://doi.org/10.1016/0031-3203(91)90058-DGet rights and content

Abstract

Character skeletonization is an essential step in many character recognition techniques. In this paper, skeletonization of Arabic characters is addressed. While other techniques employ thinning algorithms, in this paper clustering of Arabic characters is used. The use of clustering technique (an expensive step) is justified by the properties of the generated skeleton which has the advantages of other thinning techniques and is robust. The presented technique may be used in the modeling and training stages to reduce the processing time of the recognition system.

References (21)

  • B. Parhami et al.

    Automatic recognition of printed Farsi texts

    Pattern Recognition

    (1981)
  • K.J. Udupa et al.

    Some new concepts for encoding line patterns

    Pattern Recognition

    (1975)
  • L. Zadeh

    Fuzzy sets

    Inf. Control

    (1965)
  • A. Rajavelu et al.

    A neural network approach to character recognition

    Neural Networks

    (1989)
  • Pepi Siy et al.

    Fuzzy logic for handwritten numeral character recognition

    IEEE Trans. Syst. Man Cybern.

    (1974)
  • W.J.M. Kickert et al.

    Application of fuzzy set theory to syntactic pattern recognition of handwritten capitals

    IEEE Trans. Syst. Man Cybern.

    (1976)
  • Ching Y. Suen et al.

    Automatic recognition of handprinted characters—the state of the art

  • G. Wolberg

    A syntactic omni-font character recognition

  • M. Beun

    A flexible method for automatic reading of handwritten numerals

    Phillips tech. Rev.

    (1973)
  • H. Almuallim et al.

    A method of recognition of Arabic cursive handwriting

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1987)
There are more references available in the full text version of this article.

Cited by (33)

  • Manifold learning for the shape-based recognition of historical arabic documents

    2013, Handbook of Statistics
    Citation Excerpt :

    Usually, a fraction of the information available is extracted and used as features. Skeleton images, in contrast, contain much less noisy information (Mahmoud et al., 1991; Steinherz et al., 2000). However, it is much more difficult to compare them at the image level, and so it is the high-level features that are usually extracted from them (Zhu, 2007).

  • High-performance Arabic character recognition

    1998, Journal of Systems and Software
  • Appendix: Digital topology - A brief introduction and bibliography

    1996, Machine Intelligence and Pattern Recognition
  • A neural network based dedicated thinning method

    1995, Pattern Recognition Letters
  • An overview of character recognition focused on off-line handwriting

    2001, IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews
View all citing articles on Scopus
View full text