Advertisement

Handwritten Digit Recognition Using SVM Binary Classifiers and Unbalanced Decision Trees

  • Adriano Mendes Gil
  • Cícero Ferreira Fernandes Costa FilhoEmail author
  • Marly Guimarães Fernandes Costa
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8814)

Abstract

In this work, we use SVM binary classifiers coupled with a binary classifier architecture, an unbalanced decision tree, for handwritten digit recognition. According to input variables, two classifiers were trained and tested. One using digit characteristics and the other using the whole image as input variables. Developed recently, the unbalanced decision tree architecture provides a simple structure for a multiclass classifier using binary classifiers. In this work, using the whole image as input, 100% handwritten digit recognition accuracy was obtained in the MNIST database. These are the best results published in the literature for the MNIST database.

Keywords

Handwritten digit recognition MNIST database Support vector machine Unbalanced decision tree Binary classifiers 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Cheriet, M., Kharma, N., Liu, C.L., Suen, C.: Character Recognition Systems. Wiley, New Jersey (2007)CrossRefzbMATHGoogle Scholar
  2. 2.
    Decoste, D., Schölkopf, B.: Training Invariant Support Vector Machines. Kluwer Academic Publishers, The Netherlands (2002)Google Scholar
  3. 3.
    Ciresan, D.C., Meier, U., Gambardella, L.M., Schmidhuber, J.: Convolutional Neural Network Committees for Handwritten Character Classification. In: International Conference on Document Analysis and Recognition (ICDAR), pp. 1135 –1139 (2011)Google Scholar
  4. 4.
    Ciresan, D., Meier, U., Schmidhuber, J.: Multi-column Deep Neural Networks for Image Classification. Dalle Molle Institute for Artificial Intelligence. IDSIA/USI-SUPSI, Manno, Switzerland (2012)Google Scholar
  5. 5.
    Chung, Y.Y., Wong, M.T.: Handwritten character recognition by Fourier descriptors and neural network. In: IEEE Region 10 Annual Conference on Speech and Image Technologies for Computing and Telecommunications, vol. 1, pp. 391–394 (2007)Google Scholar
  6. 6.
    Poon, J.C., Man, G.M.: An enhanced approach to character recognition by Fourier descriptor. In: ICCS/ISITA 1992, vol. 2, pp. 558–562 (1992)Google Scholar
  7. 7.
    Kussul, E., Baidyk, T.: Improved method of handwritten digit recognition tested on MNIST database. In: 15th International Conference on Vision Interface, vol. 22, pp. 971–981 (2004)Google Scholar
  8. 8.
    Masmoudi, M., Samet, M., Taktak, F., Alimi, A.M.: A hardware implementation of neural network for the recognition of printed numerals. In: The Eleventh International Conference on Microelectronics, pp. 113–116 (1999)Google Scholar
  9. 9.
    Mandalia, A.D., Pandya, A.S., Sudhakar, R.: A hybrid approach to recognize handwritten alphanumeric characters. In: International Conference on System, Man and Cybernetics, vol. 1, pp. 723–726 (1992)Google Scholar
  10. 10.
    Travieso, C.M., Alonso, J., Ferrer, M.A.: Combining different off-line handwritten character recognizers. In: 15th International Conference on Intelligent Engineering Systems, Propad, pp. 315–318 (2011)Google Scholar
  11. 11.
    Oh, I.-S., Suen, C.Y.: A class-modular feedforward neural network for handwriting recognition. Pattern Recognition 35(1), 229–244 (2002)CrossRefzbMATHGoogle Scholar
  12. 12.
    LeCun, Y., Cortes, C., Burges, C.J.: The MNIST database of Handwritten Digits, http://yann.lecun.com/exdb/mnist/ (accessed in January 02, 2014)
  13. 13.
    Deng, L.: The MNIST Database of Handwritten Digit Images for Machine Learning Research. IEEE Signal Processing Magazine, 141–142 (2012)Google Scholar
  14. 14.
    Hassan, A., Damper, R.I.: Classification of emotional speech using 3DEC hierarchical classifier. Speech Communication 54, 903–916 (2012)CrossRefGoogle Scholar
  15. 15.
    Ramanan, A., Suppharangsan, S., Niranjan, M.: Unbalanced Decision Trees for Multi-class Classification. In: International Conference on Industrial and Information Systems, Sri Lanka, pp. 291–294 (2007)Google Scholar
  16. 16.
    Theodoridis, S., Koutroumbas, K.: Pattern Recognition. Elsevier Academic Press, San Diego (2006)zbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Adriano Mendes Gil
    • 1
  • Cícero Ferreira Fernandes Costa Filho
    • 2
    Email author
  • Marly Guimarães Fernandes Costa
    • 2
  1. 1.Instituto Nokia de TecnologiaManausBrazil
  2. 2.Universidade Federal do Amazonas/Centro de Pesquisa e Desenvolvimento em Tecnologia Eletrônica e da Informação – UFAM/CETELIManausBrazil

Personalised recommendations