On structural modelling for omnifont and handwritten character recognition

  • Nadeem A. Khan
  • Hans A. Hegt
Poster Papers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1451)


A novel scheme for structural modelling of multifont and handwritten characters is presented. The focus is on constructing such structural models that can be hierarchically interpreted leading to a multistage recognition scheme. This can form a basis of a high speed but reliable classifier which leaves the task of detailed discrimination between confusion classes to the secondary stage. The proposed class descriptors or prototype shape models utilise a certain “well-thought” set of shape primitives. They are “simplified” enough to ignore the inter-class variations in font-type or writing style yet retaining enough details for discrimination between the samples of the similar classes at the secondary stage. This type of modelling when combined with a good scheme of checking the spatial interrelation of features results in a powerful character recognition system utilising minimal prototypes per class. It also proves to be robust against various distortions and degradation like touching and broken characters.


Secondary Stage Handwritten Character Prototype Class Handwritten Character Recognition Prototype Feature 
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.


  1. 1.
    Tian, Q., Omnifont Printed Character Recognition. SPIE, Vol.1606. Visual Communication and Image Processing, (1991) 260–268Google Scholar
  2. 2.
    Anisimovich, K., Using Combination of Structural, Feature andRaster Classifiers for Recognition of Handprinted Characters. Proc. Fourth International Conference on Document Analysis and Recognition, Vol. 2, (August 1997) 881–885Google Scholar
  3. 3.
    Rocha, J., Pavlidis, T.: A shape analysis model with applications to a character recognition system. IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol. 16. No. 4, (April 1994) 393–404CrossRefGoogle Scholar
  4. 4.
    Zhou, J., Pavlidis, T.: Discrimination of characters by a multi-stage recognition process. Pattern recognition, Vol. 27. No.11, (1994) 1539–1549CrossRefGoogle Scholar
  5. 5.
    Wang, L., Pavlidis, T.: Direct gray-scale extraction of features for character recognition. IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 15. No. 10, (October 1993) 1053–1067CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Nadeem A. Khan
    • 1
  • Hans A. Hegt
    • 1
  1. 1.Department of Electrical Engineering, SES-Group, EH 5.25/5.28Eindhoven University of TechnologyEindhovenThe Netherlands

Personalised recommendations