Segmental K-Means Learning with Mixture Distribution for HMM Based Handwriting Recognition

  • Tapan Kumar Bhowmik
  • Jean-Paul van Oosten
  • Lambert Schomaker
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6744)


This paper investigates the performance of hidden Markov models (HMMs) for handwriting recognition. The Segmental K-Means algorithm is used for updating the transition and observation probabilities, instead of the Baum-Welch algorithm. Observation probabilities are modelled as multi-variate Gaussian mixture distributions. A deterministic clustering technique is used to estimate the initial parameters of an HMM. Bayesian information criterion (BIC) is used to select the topology of the model. The wavelet transform is used to extract features from a grey-scale image, and avoids binarization of the image.


Hide Markov Model Bayesian Information Criterion Mixture Component Mixture Distribution Handwriting 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.


  1. 1.
    Rabiner, L.: A tutorial on hidden markov models and selected applications in speech recognition. Trans. on. IEEE 77(2), 257–286 (1989)Google Scholar
  2. 2.
    Juang, B.H., Rabiner, L.: The segmental k-means algorithm for estimating parameters of hidden markov models. IEEE Trans. on ASSP 38(9), 1639–1641 (1990)CrossRefzbMATHGoogle Scholar
  3. 3.
    Wu, C.F.J.: On the convergence properties of the EM algorithm. The Annals of Statistics 11(1), 95–103 (1983)MathSciNetCrossRefzbMATHGoogle Scholar
  4. 4.
    van der Zant, T., Schomaker, L., Haak, K.: Handwritten-word spotting using biologically inspired features. IEEE Trans. on PAMI 30(11), 1945–1957 (2008)CrossRefGoogle Scholar
  5. 5.
    Bhowmik, T.K., Ghanty, P., Roy, A., Parui, S.K.: Svm-based hierarchical architectures for handwritten bangla character recognition. International Journal on Document Analysis and Recognition(IJDAR) 12(2), 97–108 (2009)CrossRefGoogle Scholar
  6. 6.
    Su, T., Dy, J.: A deterministic method for initializing k-means clustering. In: Proc. of The 16th IEEE Int. Conf. on Tools with Artificial Intelligence, pp. 784–786 (2004)Google Scholar
  7. 7.
    Rahimi, A.: An erratum for a tutorial on hidden markov models and selected applications in speech recognition. In: Online article (2000)Google Scholar
  8. 8.
    Celeux, G., Durand, J.B.: Selecting hidden markov model state number with cross-validated likelihood. Computational Statistics 23(4), 541–564 (2008)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Biem, A.: A model selection criterion for classification: Application to hmm topology optimization. In: Proceedings of the 7th ICDAR, pp. 104–108 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Tapan Kumar Bhowmik
    • 1
  • Jean-Paul van Oosten
    • 1
  • Lambert Schomaker
    • 1
  1. 1.Faculty of Mathematics and Natural SciencesUniversity of GroningenNetherlands

Personalised recommendations