Multiple Classifier Combination for Character Recognition: Revisiting the Majority Voting System and Its Variations

  • A. F. R. Rahman
  • H. Alam
  • M. C. Fairhurst
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2423)


In recent years, strategies based on combination of multiple classifiers have created great interest in the character recognition research community. A huge number of complex and sophisticated decision combination strategies have been explored by researchers. However, it has been realized recently that the comparatively simple Majority Voting System and its variations can achieve very robust and often comparable, if not better, performance than many of these complex systems. In this paper, a review of various Majority Voting Systems and their variations are discussed, and a comparative study of some of these methods is presented for a typical character recognition task.


Multiple classifier combination majority voting character recognition 


  1. 1.
    T. K. Ho, J. J. Hull, and S. N. Srihari. Decision combination in multiple classifier systems. IEEE Trans. Pattern Analysis and Machine Intelligence, 16(1):66–75, January 1994.CrossRefGoogle Scholar
  2. 2.
    Y. S. Huang, K. Liu, and C. Y. Suen. The combination of multiple classifiers by a neural-network approach. Int. Jour. of Pattern Recognition and Artificial Intelligence, 9(3):579–597, 1995.CrossRefGoogle Scholar
  3. 3.
    A. F. R. Rahman and M. C. Fairhurst. Serial combination of multiple experts: A unified evaluation. Pattern Analysis and Applications, 2:292–311, 1999.CrossRefGoogle Scholar
  4. 4.
    A. F. R. Rahman and M. C. Fairhurst. Enhancing multiple expert decision combination strategies through exploitation of a priori information sources. IEE Proc. on Vision, Image and Signal Processing, 146(1):1–10, 1999.Google Scholar
  5. 5.
    A. F. R. Rahman and M. C. Fairhurst. Machine-printed character recognition revisited: Re-application of recent advances in handwritten character recognition research. Special Issue on Document Image Processing and Multimedia Environments, Image & Vision Computing, 16(12–13):819–842, 1998.Google Scholar
  6. 6.
    L. Lam, Y. S. Huang, and C. Y. Suen. Combination of multiple classifier decisions for optical character recognition in Handbook of Character Recognition and Document Image Analysis, pages 79–101. World Scientific Publishing Company, 1997. H. Bunke and P. S. P. WangEds..Google Scholar
  7. 7.
    J. Kittler, M. Hatef, R.P.W. Duin, and J. Matas. On combining classifiers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(3):226–239, 1998.CrossRefGoogle Scholar
  8. 8.
    L. Lam and C. Y. Suen. Application of majority voting to pattern recognition: An analysis of its behavior and performance. IEEE Trans. Pattern Analysis and Machine Intelligence, 27(5):553–568, 1997.Google Scholar
  9. 9.
    L. Lam and C. Y. Suen. A theoretical-analysis of the application of majority voting to pattern-recognition. In Proc. 12th IAPR Int. Conf. on Pattern Recognition, Conf. B: Pattern Recognition and Neural Networks, volume 2, pages 418–420, Jerusalem,Israel, 1994.Google Scholar
  10. 10.
    I. Todhunter. A History of the Mathematical Theory of Probability from the Time of Pascal to that of Laplace. Macmillan, Cambridge, UK, 1865.Google Scholar
  11. 11.
    N. C. de Condorcet. Essai sur l’Application de l’Analyze à la Probabilité des Décisions Rendues à la Pluralité des Voix. Imprimŕie Royale, Paris, France, 1785.Google Scholar
  12. 12.
    C. Y. Suen, C. Nadal, T. A. Mai, R. Legault, and L. Lam. Recognition of totally unconstrained handwritten numerals based on the concept of multiple experts. In Proc. IWFHR, pages 131–143, Montréal, Canada, 1990.Google Scholar
  13. 13.
    G. S. Ng and H. Singh. Democracy in pattern classifications: Combinations of votes from various pattern classifiers. Artificial Intelligence in Engineering, 12(3):189–204, 1998.CrossRefGoogle Scholar
  14. 14.
    A. Stajniak, J. Szostakowski, and S. Skoneczny. Mixed neural-traditional classifier for character recognition. In Proc. Int. Conf. on Imaging Sciences and Display Technologies: International Society for Optical Engineering (SPIE), volume 2949, pages 102–110, Berlin,Germany, 1997.Google Scholar
  15. 15.
    A. Belaid and J. C. Anigbogu. Use of many classifiers for multifont text recognition. Traitement du Signal, 11(1):57–75, 1994.Google Scholar
  16. 16.
    J. R. Parker. Voting methods for multiple autonomous agents. In Proc. 3rd Australian and New Zealand Conf. on Intelligent Information Systems, pages 128–133, Perth, WA, Australia, 1995.Google Scholar
  17. 17.
    C. Ji and S. Ma. Combination of weak classfiers. IEEE Trans. on Neural Networks, 8(1):32–42, 1997.CrossRefGoogle Scholar
  18. 18.
    E. Alpaydin. Improved classification accuracy by training multiple models and taking a vote. In Sixth Italian Workshop. Neural Nets Wirn Vietri-93, pages 180–185, 1994.Google Scholar
  19. 19.
    J. Kittler, A. Hojjatoleslami, and T. Windeatt. Weighting factors in multiple expert fusion. In Proc. British Machine Vision Conference, pages 41–50, 1997.Google Scholar
  20. 20.
    S. R. Gang, T. K. Woo, and I. C. Sung. Recognition of printed and handwritten numerals using multiple features and modularized neural networks. Journal of the Korean Institute of Telematics, 32B(10):101–1111, 1995.Google Scholar
  21. 21.
    A. F. R. Rahman and M. C. Fairhurst. Exploiting second order information to design a novel multiple expert decision combination platform for pattern classification. Electronics Letters, 33(6):476–477, 1997.CrossRefGoogle Scholar
  22. 22.
    A. F. R. Rahman and M. C. Fairhurst. A New Multiple Expert Framework for Decision Fusion. In Proc. 9th Int. Graphonomics Society Conference (IGS’99), pages 161–166, Singapore, 1999.Google Scholar
  23. 23.
    R. Rovatti, R. Ragazzoni, Z. M. Kovacs, and R. Guerrieri. Voting rules for k-nearest neighbors classifiers. Neural Computation, 7(3):594–605, 1995.CrossRefGoogle Scholar
  24. 24.
    M. C. Fairhurst and A. F. R. Rahman. Enhancing Consensus in Multiple Expert Decision Fusion. IEE Proc. on Vision, Image and Signal Processing, 147(1):39–46, 2000.Google Scholar
  25. 25.
    T. K. Ho, J. J. Hull, and S. N. Srihari. Combination of Decisions by Multiple Classifiers in Structured Document Image Analysis, pages 188–202. S-V, 1992. H. S. Baird, H. Bunke and K. YamamotoEds..Google Scholar
  26. 26.
    T. K. Ho, J. J. Hull, and S. N. Srihari. On multiple classifier systems for pattern recognition. In Proc. 11th ICPR, pages 84–87,The Hague, Netherlands, 1992.Google Scholar
  27. 27.
    Q. P. Duong. The combination of forecasts: A ranking and subset selection approach. Mathematical and Computer Modelling, 12(9):1131–1143, 1989.zbMATHCrossRefMathSciNetGoogle Scholar
  28. 28.
    V. D. Mazurov, A. I. Krivonogov, and V. L. Kazantsev. Solving of optimisation and identification problems by the committee methods. Pattern Recognition, 20(4):371–378, 1987.zbMATHCrossRefMathSciNetGoogle Scholar
  29. 29.
    Y. H. Hu, J. M. Park, and T. Knoblock. Committee pattern classifiers. In Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, pages 3389–3392, Munich, Germany, 1997.Google Scholar
  30. 30.
    F. Kimura and M. Shridhar. Handwritten numeral recognition based on multiple algorithms. Pattern Recognition, 24(10):969–983, 1991.CrossRefGoogle Scholar
  31. 31.
    T. K. Ho, J. J. Hull, and S. N. Srihari. A regression approach to combination of decisions by multiple character recognition algorithms. In Machine Vision Applications in Character Recognition and Industrial Inspection, Proc. SPIE 1661, pages 137–145, 1992.Google Scholar
  32. 32.
    T. K. Ho. Adaptive coordination of multiple classifiers. In Proc. of Int. Workshop on Document Analysis Systems, pages 338–351, Malvern, Pennsylvalia, USA, 1996.Google Scholar
  33. 34.
    A. F. R. Rahman and M. C. Fairhurst. An evaluation of multi-expert configurations for recognition of handwritten numerals. Pattern Recognition, 31(9):1255–1273, 1998.CrossRefGoogle Scholar
  34. 35.
    A. F. R. Rahman and M. C. Fairhurst. Towards the design of a generic multiple expert image classification system capable of automatic self-configuration. In Proc. Int. Conf. on Quality Control by Artificial Vision (QCAV’99), pages 259–264, Québec, Canada, 1999.Google Scholar
  35. 36.
    M. C. Fairhurst and A. F. R. Rahman. A generalised approach to the recognition of structurally similar handwritten characters. IEE Proc. on Vision, Image and Signal Processing, 144(1):15–22, 1997.Google Scholar
  36. 37.
    J. Paik, S. Jung, and Y. Lee. Multiple combined recognition system for automatic processing of credit card slip applications. In Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), pages 520–523, 1993.Google Scholar
  37. 38.
    Y. Anzai, H. Mori, M. Ito, and Y. Hayashi. A serial-parallel integrated informationprocessing model for complex human problem solving. In Cognitive Engineering in the Design of Human-Computer Interaction and Expert Systems. Proceedings of the Second International Conference on Human-Computer Interaction. Vol.II, pages 175–182, 1987.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • A. F. R. Rahman
    • 1
  • H. Alam
    • 1
  • M. C. Fairhurst
    • 2
  1. 1.Document Analysis and Recognition Team (DART)BCL Technologies Inc.Santa ClaraUSA
  2. 2.Department of ElectronicsUniversity of KentKentUK

Personalised recommendations