Skip to main content
SpringerLink
Log in

Handwritten Chinese character recognition by rule-embedded Neocognitron

  • Articles
  • Published:
Neural Computing & Applications Aims and scope Submit manuscript

Abstract

This paper presents an attempt to integrate neural computation with a domain knowledge technique to resolve the problem of the wide variety in handwritten Chinese characters. Despite their complexity, Chinese characters can be seen as structured patterns. Therefore, we propose a symbolic representation to describe these structural formations. In particular, we consider the Fuzzy Attributed Production Rule (FAPR) as a possible symbolic representation. On the neural computational side, we study Fukushima's Neocognitron model, which has been successfully demonstrated to recognize handwritten alphanumerics. Despite its power and tolerance capabilities, the supervised training scheme used by Fukushima is impractical for a large character set such as Chinese characters. We thus propose a ruleembedded Neocognitron network which can be readily mapped with structure-knowledge of Chinese characters as represented in FAPRs. In this paper, we demonstrate how 50 Chinese characters are mapped onto the network, and that the system performance in tolerating character structure deviations is satisfactory.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Umeda M. Recognition of multi-font printed Chinese characters. In: Proc 6th Int Conf Pattern Recognition 1982; 2: 793–796

    Google Scholar 

  2. Huang JS, Chung ML. Separating similar complex Chinese characters by Walsh transform. Pattern Recognition 1987; 20: 425–428

    Article  Google Scholar 

  3. Stallings W. The morphology of Chinese characters: a survey of models and applications. Computers and the Humanities 1975; 9: 13–24

    Google Scholar 

  4. Tsui HT. Guided stroke structures extraction for the recognition of handprinted Chinese characters. In: Proc 6th Int Conf Pattern Recognition 1982; 2: 786–788

    Google Scholar 

  5. Chan KP, Cheung YS. Fuzzy-attribute graph and its application to Chinese character recognition. Computer Processing of Chinese & Oriental Languages 1989; 4(2&3): 85–98

    Google Scholar 

  6. Chen LH, Lieh JR. Handwritten character recognition using a 2-layer random graph model by relaxation matching. Pattern Recognition 1990; 23: 1189–1205

    Article  Google Scholar 

  7. Lu SW, Ren Y, Suen C. Hierarchical Attributed Graph Representation and Recognition of Handwritten Chinese Characters. Pattern Recognition 1991; 24: 617–632

    Article  Google Scholar 

  8. Yong Y. A neural network model of CAAM and its application to handprinted Chinese character. In: Proc 1st Int Conf Neural Networks 1987; 3: 309–316

    Google Scholar 

  9. Fukushima K. Character recognition with neural networks. Neurocomputing 1992; 4: 221–233

    Article  Google Scholar 

  10. Tai JW, Liu YJ, Zhang LQ. A new approach for feature extraction and feature selection of handwritten Chinese character recognition. In: Impedovo S, Simon JC (eds). From Pixels to Features III: Frontiers in Handwriting Recognition. North-Holland, 1992, pp 479–491

  11. Fukushima K. Neocognitron: a hierarchical neural network capable of visual pattern recognition. Neural Networks 1988; 1: 119–130

    Article  Google Scholar 

  12. Fukushima K, Wake N. Neocognitron: a hierarchical neural network model for pattern recognition. In: Proc Int Conf Fuzzy Logic & Neural Networks, IIZUKA, 1990; pp 387–390

  13. Kim EJ, Lee Y. Handwritten Hangul recognition using a modified neocognitron. Neural Networks 1991; 4: 743–750

    Article  Google Scholar 

  14. Cihai (‘Sea of Words’, a large dictionary). Beijing, Shanghai: Shangwu Yinshuguan (The Commercial Press), 1979

  15. Liu YJ, Tai JW. An on-line Chinese character recognition system for handwritten in Chinese calligraphy. In: Impedovo S, Simon JC (eds). From Pixels to Features III: Frontiers in Handwriting Recognition. North-Holland, 1992, pp 87–99

  16. Yeung DS, Chan HY, Cheung KF. Synthesizing a neocognitron network without training. In: Proc Int Conf Fuzzy Logic & Neural Networks, IIZUKA 1992; 2: 1163–1166

    Google Scholar 

  17. Yeung DS, Fong HS, Cheung KF. Development of a neural network system for Chinese character recognition. In: Proc Int Conf Fuzzy Logic & Neural Networks, IIZUKA 1992; 2: 1159–1162.

    Google Scholar 

  18. Yeung DS, Fong HS, Cheung KF. A neocognitron-based Chinese character recognition system. In: Proc Int Joint Conf Neural Networks, Beijing, 1992; 3: 617–622

    Google Scholar 

  19. Yeung DS, Fong HS. A knowledge matrix representation for a rule-mapped neural network. International Journal of Neurocomputing 1993 (accepted)

Download references

Author information

Authors and Affiliations

  1. Department of Computing, Hong Kong Polytechnic, Hung Horn, Kowloon, Hong Kong

    Daniel S. Yeung & H. S. Fong

Authors
  1. Daniel S. Yeung
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. H. S. Fong
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Yeung, D.S., Fong, H.S. Handwritten Chinese character recognition by rule-embedded Neocognitron. Neural Comput & Applic 2, 216–226 (1994). https://doi.org/10.1007/BF01414810

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01414810

Keywords

Search

Navigation