Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 30753
Unsupervised Segmentation by Hidden Markov Chain with Bi-dimensional Observed Process

Authors: Abdelali Joumad, Abdelaziz Nasroallah

Abstract:

In unsupervised segmentation context, we propose a bi-dimensional hidden Markov chain model (X,Y) that we adapt to the image segmentation problem. The bi-dimensional observed process Y = (Y 1, Y 2) is such that Y 1 represents the noisy image and Y 2 represents a noisy supplementary information on the image, for example a noisy proportion of pixels of the same type in a neighborhood of the current pixel. The proposed model can be seen as a competitive alternative to the Hilbert-Peano scan. We propose a bayesian algorithm to estimate parameters of the considered model. The performance of this algorithm is globally favorable, compared to the bi-dimensional EM algorithm through numerical and visual data.

Keywords: Image Segmentation, bayesian approach, Hidden Markov chain with a bi-dimensional observed process, Peano-Hilbert scan, MCMC methods, Bi-dimensional EM algorithm

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1056130

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1209

References:


[1] A. Belmaati, L. Omari, H. Elmaroufy and S. Elfakir, Mod`ele de Markov cach'e `a processus observ'e bidimensionnel : 'etude de la transition entre les tats de l-infection par le VIH, R.E.S.P, 52: IS42-IS44, 2007.
[2] A. P. Dempster, N. M. Laird and D. B. Rubin, Maximum likelihood from incomplete data via the em algorithm (with discussion), Journal of the Royal Statistical Society, B 39 :1-38, 1977.
[3] A. Peng and W. Pieczynski, Adaptive mixture estimation and unsupervised local Bayesian image segmentation, CVGIP: Graphical Models and Image Processing, Vol. 57, No. 5, pp. 389-399, 1995.
[4] A. E. Raftery and S. Lewis, Implementing MCMC. In W. R. Gillks, S. Richardson Markov chain Monte Carlo in practice, Chapman and Hall/CRC, pp. 115-130, 1996.
[5] B. Chalmond, An iterative Gibbsian technique for reconstruction of mary images, Pattern Recognition, Vol. 22, N6, pp . 747-761, 1989.
[6] B. Benmiloud, A. Peng and W. Pieczynski, Estimation conditionnelle it'erative dans les chaˆınes de Markov cach'ees et segmentation statistique non supervis'ee d-images, Actes de Quatorzi`eme Colloque GRETSI 93, Juans-les-Pins, France, pp. 105-108, 1993.
[7] B. Benmiloud and W. Pieczynski Estimation des paramtres dans les chaˆınes de Markov cach'ees et segmentation d-images, Traitement du Signal, Vol. 12, No. 5, pp. 433-454, 1995.
[8] B. Braathen, W. Pieczynski and P. Masson, Global and local methods of unsupervised Bayesian segmentation of images, Machine Graphics and Vision, Vol. 2, N1, pp.39-52, 1993.
[9] C. P. Robert Le choix bay'esien, Springer, 2006.
[10] F. Heitz and P. Bouthemy Estimation et segmentation du mouvement : approche Bay'esienne et mod'elisation Markovienne des occlusions, Actes de la 7`eme Conf. AFCET/RFIA, Paris, pp. 1359-1368, 1989.
[11] F. Desbouvries and W. Pieczynski, Mod`eles de Markov Triplet et filtrage de Kalman-Triplet Markov Models and Kalman filtering, Comptes Rendus de l-Acad'emie des Sciences-Math'ematique, S'erie I, Vol. 336, No. 8, pp. 667-670, 2003.
[12] F. Salzenstein and W. Pieczynski, Sur le choix de m'ethode de segmentation statistique d-images, Traitement du Signal, Vol. 15, No. 2, pp. 120-127, 1998.
[13] J. Besag, On the statistical analysis of dirty pictures, Journal of the Royal Statistical Society, Series B, 48, pp . 259-302, 1986.
[14] H. Caillol, A. Hillion and W. Pieczynski, Fuzzy random fields and unsupervised image segmentation, IEEE Transaction on geoscience and remote sensing, Vol. 31, No 4, pp. 801-810, 1993.
[15] K. Yao, A representation theorem and its applications to spherically invariant random processes, IEEETrans.Inform.Theory,vol.IT-19, no.5,pp.600-608, 1973.
[16] K. Abend, T. J . Harley and L. N. Kanal, Classification of binary random patterns, IEEE Transactions on Information Theory, Vol. IT-11, N4, 1965.
[17] L. Younes, Parametric inference for imperfectly observed Gibbsian fields, Probability Theory and Related Fields, 82, pp . 625-645, 1989.
[18] L. E. Baum, T. Petrie, G. Soules and N. Weiss, A maximization technique occuring in the statstical analysis of probabilistic functions of Markov chains , Ann, Math. Statistic, 41, pp. 164-171, 1970.
[19] M. Emsalem, H. Caillol, P. Obvi'e, G. Carnat and W. Pieczynski, Fast unsupervised statistical image segmentation, IEEE International Geoscience and Remote Sensing Symposium (IGARSS 92), Houston, Texas, 1992.
[20] M. Vieira, Estimation bay'esienne par des m'ethodes MCMC-Application `a la surveillance des moteurs asynchrones, Th`ese, Universit'e de Nice- Sophia-Antiplois, 1999.
[21] M. Rangaswamy, D. Weiner and A. Ozturk, Non-Gaussian random vector identification using spherically invariant random process, IEEE Trans. on Aerospace and Electronic Systems , 29(1) :111-124, 1993.
[22] N. Brunel, J. Lapuyade-Lahorgue and W. Pieczynski, Modeling and unsupervised classification of multivariate hidden Markov chains with copulas, IEEE Trans. on Automatic Control, Vol. 55, No. 2, pp. 338-349, 2010.
[23] N. Giordana and W. Pieczynski, Estimation of generalized multisensor hidden Markov chains and unsupervised image segmentation, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 5, pp. 465-475, 1997.
[24] N. Brunel, W. Pieczynski and S. Derrode, Copulas in vectorial hidden Markov chains for multicomponent image segmentation, in InProc.Int.Conf.Acoust.,Speech Signal Processing(ICASSP-05), pp.717- 720, 2005.
[25] O. Capp'e, E. Moulines and T. Ryd'en, Inference hidden Markov models, Springer, 2005.
[26] P. A. Devijver and M. Dekesel, Champs al'eatoires de Pickard et mod'elisation d-images digitales, Traitement du Signal, Vol. 5, N5, pp. 131-150, 1988.
[27] P. Lalande, D'etection du mouvement dans une s'equence d-images selon une approche markovienne; Application `a la robotique sous-marine, Th`ese, Universit'e de Rennes I, 1990.
[28] P. Bouthemy and P. Lalande, Motion detection in an image sequence usin Gibbs distribution, Actes d-ICASSP-89, Glasgow, pp. 1651-1654, 1989.
[29] P. Masson and W. Pieczynski, SEM Algorithm and unsupervised statistical segmentation and satellite images, IEEE Transaction on geoscience and remote sensing, Vol. 31, No 3, pp. 618-633, 1993.
[30] S. Geman and D. Geman, Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images, IEEE Transactions on PAMI, Vol. PAMI-6, N6, pp. 721-741, 1984.
[31] S. Derrode, G. Mercier and W. Pieczynski, Unsupervised multicomponent image segmentation combining a vectorial HMC model and ICA, International Conference on Image Processing (ICIP 2003), Barcelona, Spain, September 14-17, 2003.
[32] S. Rafi, M. Castella and W. Pieczynski, Pairwise Markov model applied to unsupervised image separation, IASTED International Conference on Signal Processing, Pattern Recognition, and Applications (SPPRA), 16-18 February, Innsbruck, Austria, 2011.
[33] T. J.BarnardandF.Khan, Statistical normalization of spherically invariant non-Gaussian clutter, IEEEJ.Oceanic Eng,vol.29,no.2,pp.303-309, 2004.
[34] Y. Delignon and W. Pieczynski, Modeling non-Rayleigh speckle distribution in SAR images, IEEETrans.Geosci.Remote Sensing, vol.40,no.6,pp.1430-1435, 2002.
[35] W. Pieczynski, Champs de Markov cach'es et estimation conditionnelle it'erative, Traitement du Signal, Vol. 11, No. 2, pp. 141-153, 1994.
[36] W. Pieczynski, Mod`eles de Markov en traitement d-images , Traitement du Signal, Vol. 20, No. 3, pp. 255-278, 2003.
[37] W. Pieczynski, EM and ICE in hidden and triplet Markov models, Stochastic Modeling Techniques and Data Analysis international conference (SMTDA -10), Chania, Greece, 2010.
[38] W. R. Gillks, A. Thomas and D. J. Spiegelhalter, Software for the Gibbs Sampler, Comp. Sci. Statist, 24: 439-448, 1992.
[39] W. R. Gillks and S. Richardson, Markov chain Monte Carlo in practice, Chapman and Hall/CRC, 1996.
[40] X. Guyon, Champs al'eatoires sur un r'eseau, Collection Techniques Stochastiques, Masson, Paris, 1993.