Chinese Character Recognition-Comparison of Classification Methodologies

  • Sameer Singh
  • Adnan Amin
  • K. C. Sum
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2412)


In this paper we propose the use of dominant point method for Chinese character recognition. We compare the performance of three classifiers on the same inputs; a statistical linear classifier, a machine learning C4.5 classifier, and a fuzzy nearest neighbour method of classification. Such a comparison highlights the degree of advantage of correct recognition method for our problem.


Chinese Character Character Recognition Fuzzy Classifier Dominant Point Chinese Character Recognition 
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.
    Kanal, J., Liu, Y. and Nagy, G.: An OCR-Oriented overview of idographic writing systems, Handbook of Character Recognition and Document Image Analysis, Ed. H. Bunke and P. S. P Wang, World Scientific, pp. 285–304, 1997.Google Scholar
  2. 2.
    Lee, H.J.: Chinese character recognition in Taiwan, Handbook of Character Recognition and Document Image Analysis, Ed. H. Bunke and P. Wang, W. Scientific, pp. 331–355, 1997.Google Scholar
  3. 3.
    Cheng, H.D. and Wang, J.F.: Preclassification for handwritten Chinese character recognition by a peripheral shape coding method, Pattern Recognition 26, pp. 711–719, 1993.CrossRefGoogle Scholar
  4. 4.
    Cheng, R.H. Lee, C.W. and Chen, Z.: Preclassification of handwritten Chinese character recognition based on basic stroke substructures, Proc. 4th International workshop on Frontiers in Handwritten Recognition, Taiwan, pp. 176–184, 1994.Google Scholar
  5. 5.
    Kumamoto, T., Toraichi, K., Horiuchiz, T., Yamamoto, K. and Yamada, H.: On Speeding candidate selection in handprinted Chinese character recognition, Pattern Recognition 24, pp. 793–799, 1991.CrossRefGoogle Scholar
  6. 6.
    Tung, C.H., Lee, H.J. and Tsai, J.Y.: Multistage pre-candidate selection in handwritten Chinese character recognition systems, Pattern Recognition 27, pp. 1093–1102, 1994.CrossRefGoogle Scholar
  7. 7.
    Lin, T.Z. and Fan, K.C.: Coarse classification of on-line Chinese character via structure feature-based method, Pattern Recognition 27, pp. 1365–1378, 1994.CrossRefGoogle Scholar
  8. 8.
    Huang, J.S. and Chung, M.L.: Separating similar complex Chinese characters by Walsh transform, Pattern Recognition, pp. 425–428, 1987.Google Scholar
  9. 9.
    Tu, T.Y. and Ma, Y.L.: Character recognition by stochastic sectionalgram approach, Pattern Recognition, 21, pp. 593–601, 1988.Google Scholar
  10. 10.
    Li, T.F., and Yu, S.S.: Hand printed Chinese character recognition using the probability distribution feature, IJPRAI Journal, 8, pp. 1241–1258, 1994.Google Scholar
  11. 11.
    Fu, K.S.: Syntactic (linguistic) pattern recognition in digital pattern recognition, Ed. K. S. Fu, New York Springer-Verlag, 1980Google Scholar
  12. 12.
    Lee, H.J. and Chen, B.: Recognition of handwritten Chinese characters via short line segments, Pattern Recognition 25, pp. 543–552, 1992.CrossRefGoogle Scholar
  13. 13.
    Cheng, F.H., Hsu, W.H. and Kuo, M.C.: Recognition of handprinted Chinese character via stroke relaxation, Pattern Recognition 26, pp. 579–593, 1993.CrossRefGoogle Scholar
  14. 14.
    Hsieh, A.J., Fan, K.C. and Fan, T.I., Bipertite weighted matching for on-line handwritten Chinese character recognition, Pattern Recognition 28, pp. 143–151, 1995.CrossRefGoogle Scholar
  15. 15.
    Chou, S.L. and Tsai, W.H.: Recognition of handwritten Chinese character by stroke-segment matching using an iteration scheme, IJPRAI Journal, 5, pp. 175–197, 1991.Google Scholar
  16. 16.
    Lin, C.K., Fan, K.C. and Lee, F.T.P.: On-line recognition by deviation-expansion model and dynamic programming matching, Pattern Recognition, 26, pp. 259–268, 1993.CrossRefGoogle Scholar
  17. 17.
    Cheng, F.H., Hsu, W.H. and Chen, M.Y.: Recognition of handwritten Chinese characters by modified Hough transform, IEEE Trans. PAMI-6, pp. 386–405, 1989.Google Scholar
  18. 18.
    Cheng, F.H., Hsu, W.H., and Chen, C.A.: Fuzzy approach to solve the recognition problem of handwritten Chinese characters, Pattern Recognition 33, pp. 475–484, 1990.Google Scholar
  19. 19.
    Wang, A.B., Huang, J.S. and Fan, K.C.: Recognition of handwritten Chinese characters by modified relaxation methods, Image and Vision Computing 12, pp. 509–522, 1994.CrossRefGoogle Scholar
  20. 20.
    Chen, L.H. and Leih, J.R.: Handwritten character recognition using a 2-layer randomgraph model by relation matching, Pattern Recognition 23, pp. 1189–1205, 1990.CrossRefGoogle Scholar
  21. 21.
    Hsieh, C.C. and Lee, H.J.: A probabilistic stroke-based Viterbi algorithm for handwritten Chinese characters recognition, IJPRAI Journal, 7, pp. 329–352, 1993.Google Scholar
  22. 22.
    Tsay, Y.T. and Tsai, W.H.: Attributed string matching by split-and-merge for on-line Chinese character recognition, IEEE Trans. PAMI-15, pp. 180–185, 1993.Google Scholar
  23. 23.
    Cheng, F.H. and Hsu, W.H.: Radical extraction from handwritten Chinese characters by background thinning methods, IEEE Trans. IECE E71, pp. 88–98, 1988.Google Scholar
  24. 24.
    Liao, C.W. and Huang, J.S.: A transformed invariant matching algorithm for handwritten Chinese character recognition, Pattern Recognition 23, pp. 1167–1188, 1990.CrossRefGoogle Scholar
  25. 25.
    Cheng, F.H. and Hsu, W.H.: Radical extraction by background thinning method for handwritten Chinese characters, Proc. Int. Conf. on Chinese Computing, pp. 175–182, 1987.Google Scholar
  26. 26.
    Amin, A. and Al-Sadoun, H.B.: Handprinted Arabic character recognition system using an artificial neural network, Pattern Recognition, 28(4), 663–675, 1996.CrossRefGoogle Scholar
  27. 27.
    Jang, B.K. and Chin, R.T.: One-pass parallel thinning: analysis, properties, and quantitative evaluation, IEEE Trans. PAMI-14, 1129–1140, 1992.Google Scholar
  28. 28.
    Liao, C.W. and Huang, J.S.: Stroke segmentation by Bernstein-Bezier curve fitting, Pattern Recognition 23, pp. 475–484, 1990.CrossRefGoogle Scholar
  29. 29.
    Freeman, H.: On the encoding of arbitrary geometric configurations, IEEE Trans. Electronic Computers EC-10, 260–268, 1968.MathSciNetCrossRefGoogle Scholar
  30. 30.
    Cheng, F.H. and Hse, W.H.: Three stroke extraction methods for recognition of handwritten Chinese characters, Proc. Int. Conf. Chinese Computing, Singapore, 191–195, 1986.Google Scholar
  31. 31.
    The, C.H. and Chin, R.T. On the Detection of Dominant Points in Digital Curves, IEEE PAMI-11, 859–872, 1989.Google Scholar
  32. 32.
    Rosenfeld, A. and Johnston, E.: Angle detection on digital curves, IEEE Trans. Computers C-22, 875–878, 1973.CrossRefGoogle Scholar
  33. 33.
    Huberty, C.J.: Applied discriminant analysis, Wiley-Statistics, 1994.Google Scholar
  34. 34.
    Stone, M.: Cross-validatory choice and assessment of statistical predictions, Journal of the Royal Statistical Society. 36: 111–147, 1974.zbMATHGoogle Scholar
  35. 35.
    Quinlan, J.R.: C4.5; Programs for machine learning, Morgan Kauffman, 1993.Google Scholar
  36. 36.
    Quinlan, J.R.: Discovering rules for a large collection of examples, Edinburgh, 1979.Google Scholar
  37. 37.
    Pal, S.K. and Majumder, D.D.: Fuzzy mathematical approach to pattern recognition, John Wiley, 1986.Google Scholar
  38. 38.
    Amin, A. and Singh, S.: Recognition of hand-printed Chinese characters using decision trees/machine learning C4.5 system, Pattern Analysis and Applications, 1(2): 130–141, 1998.zbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Sameer Singh
    • 1
  • Adnan Amin
    • 2
  • K. C. Sum
    • 2
  1. 1.Department of Computer ScienceUniversity of ExeterExeterUK
  2. 2.School of Computer Science and EngineeringUNSWSydneyAustralia

Personalised recommendations