Les modélisations par chaı̂nes de Markov cachées (CMC), dont la possibilité de calcul explicite de la loi Markovienne du processus caché conditionnelle aux observations est le principal intérêt, trouvent de très nombreuses applications dans les domaines les plus divers. Ce modèle a été récemment généralisé aux chaı̂nes de Markov « Couple », présentant les mêmes avantages que les CMC au niveau des traitements et proposant un pouvoir modélisant plus important. L'objet de cette Note est préciser comment les Arbres de Markov cachés, qui sont des extensions des CMC, peuvent également être généralisés aux modèles originaux appelés « Arbres de Markov couple ».
The Hidden Markov Chain (HMC) models are widely applied in various problems. This succes is mainly due to the fact that the hidden model distribution conditional on observations remains a Markov chain distribution, and thus different processings, like Bayesian restorations, are handleable. These models have been recetly generalized to “Pairwise” Markov chains, which admit the same processing power and a better modeling one. The aim of this Note is to show that the Hidden Markov trees, which can be seen as extensions of the HMC models, can also be generalized to “Pairwise” Markov trees, which present the same processing advantages and better modelling power.
Révisé le :
Publié le :
@article{CRMATH_2002__335_1_79_0, author = {Pieczynski, Wojciech}, title = {Arbres de {Markov} couple}, journal = {Comptes Rendus. Math\'ematique}, pages = {79--82}, publisher = {Elsevier}, volume = {335}, number = {1}, year = {2002}, doi = {10.1016/S1631-073X(02)02430-5}, language = {fr}, url = {http://www.numdam.org/articles/10.1016/S1631-073X(02)02430-5/} }
Pieczynski, Wojciech. Arbres de Markov couple. Comptes Rendus. Mathématique, Tome 335 (2002) no. 1, pp. 79-82. doi : 10.1016/S1631-073X(02)02430-5. http://www.numdam.org/articles/10.1016/S1631-073X(02)02430-5/
[1] A maximization technique occuring in the statistical analysis of probabilistic functions of Markov chains, Ann. Math. Statist., Volume 41 (1970), pp. 164-171
[2] Multiscale autoregressive models and wavelets, IEEE Trans. Inform. Theory, Volume 45 (1999) no. 3, pp. 828-845
[3] Modelling and segmentation of noisy and textured images using Gibbs random fields, IEEE Trans. PAMI, Volume 9 (1987) no. 1, pp. 39-55
[4] Discrete Markov image modeling and inference on the quadtree, IEEE Trans. Image Processing, Volume 9 (2000) no. 3, pp. 390-404
[5] Efficient multiscale regularization with applications to the computation of optical flow, IEEE Trans. SP, Volume 41 (1993) no. 12, pp. 3377-3396
[6] Pairwise Markov random fields and segmentation of textured images, Machine Graphics & Vision, Volume 9 (2000) no. 3, pp. 705-718
[7] Pairwise Markov chains and Bayesian unsupervised fusion, Proceedings of 3rd ICIF, Vol. 1, FUSION 2000, July 10–13, Paris, France, 2000, p. MoD4-24-MoD4-31
[8] W. Pieczynski, Pairwise Markov Chains, IEEE Trans. PAMI, submitted
[9] Graphical Models in Applied Multivariate Statistics, Wiley Ser. Probab. Math. Statist., 1996
Cité par Sources :