Advanced SearchSearch Tips
Hyper-Parameter in Hidden Markov Random Field
facebook(new window)  Pirnt(new window) E-mail(new window) Excel Download
 Title & Authors
Hyper-Parameter in Hidden Markov Random Field
Lim, Jo-Han; Yu, Dong-Hyeon; Pyu, Kyung-Suk;
  PDF(new window)
Hidden Markov random eld(HMRF) is one of the most common model for image segmentation which is an important preprocessing in many imaging devices. The HMRF has unknown hyper-parameters on Markov random field to be estimated in segmenting testing images. However, in practice, due to computational complexity, it is often assumed to be a fixed constant. In this paper, we numerically show that the segmentation results very depending on the fixed hyper-parameter, and, if the parameter is misspecified, they further depend on the choice of the class-labelling algorithm. In contrast, the HMRF with estimated hyper-parameter provides consistent segmentation results regardless of the choice of class labelling and the estimation method. Thus, we recommend practitioners estimate the hyper-parameter even though it is computationally complex.
Hidden Markov random field;hyper-parameter;image segmentation;
 Cited by
Aiyer, A., Pyun, K., Huang, Y., OBrien, D. B. and Gray, R. M. (2005). Lloyd clustering of Gauss mixture models for image compression and classication, Signal Processing: Image Communication, 20, 459-485. crossref(new window)

Besag, J. (1986). On the statistical analysis of dirty pictures, Journal of the Royal Statistical Society-B, 48, 259-302.

Celeux, G. and Diebolt, J. (1985). The SEM algorithm: A probabilisitic teacher algorithm derived from the EM algorithm for the mixture, Computational Statistics Quarterly, 2, 73-82.

Diebolt, J. and Ip, E. H. S. (1996). Stochastic EM: Method and application, In: Gilks W. R., Richardson S. T. and Spiegelhalter D. J., Editors, Markov Chain Monte Carlo in Practice, Chapman & Hall, London.

Geman, S. and Geman, D. (1984). Stochastic relaxation, Gibbs distribution, and the Bayesian restoration of images, IEEE Transaction on Pattern Analysis and Machine Intelligence, 11, 689-691.

Lavielle, M. and Moulines, E. (1997). A simulated annealing version of the EM algorithm for non-Gaussian deconvolution, Statistics and Computing, 7, 229-236. crossref(new window)

Li, J. and Gray, R. M. (1999). Image classication based on a multi-resolution two dimensional hidden Markov model, Proceedings International Conference, 1, 348-352.

Li, J., Najmi, A. and Gray, R. M. (1999). Image classication by a two dimensional hidden Markov model, Acoustics, Speech, and Signal Processing, IEEE International Conference, 6, 3313-3316.

Li, J., Najmi, A. and Gray, R. M. (2000). Image classication by a two dimensional hidden Markov model, IEEE-Transaction on Signal Processing, 48, 517-533. crossref(new window)

Liu, J. S. (2001). Monte Carlo Strategies in Scientic Computing, Springer-Verlag, Germany.

Pyun, K. S., Lim, J. and Gray, R. M. (2009). A robust hidden Markov Gauss mixture vector quantizer for a noisy source, IEEE-Transaction on Image Processing, 18, 1385-1394. crossref(new window)

Pyun, K. S., Lim, J., Won, C. S. and Gray, R. M. (2007). Image segmentation using hidden Markov Gauss mixture model, IEEE-Transaction on Image Processing, 16, 1902-1911. crossref(new window)

Swendsen, R. H. and Wang, J. (1987). Nonuniversal critical dynamics in Monte Carlo simulations, Physics Review Letters, 58, 86-88. crossref(new window)

Ueda, N. and Nakano, R. (1995). Deterministic annealing variant of the EM algorithm, Neural Information Processing Systems, 7, 545-552.

Wei, G. C. G. and Tanner, M. A. (1991). Applications of multiple imputation to the analysis of censored regression data, Biometrics, 47, 1297-1309. crossref(new window)

Won, C. S. and Gray, R. M. (2004). Stochastic Image Processing, Springer, New York.

Zhou, Z., Leahy, R. M. and Qi, J. (1997). Approximate maximum likelihood hyperparameter estimation for Gibbs priors, IEEE Transaction on Image Processing, 6, 844-861. crossref(new window)