Spatial modelling for mixed-state observations. (English) Zbl 1135.62043

Hardouin, Cécile; Yao, Jian-Feng

Electron. J. Stat. 2, 213-233 (2008).

Summary: In several application fields like daily pluviometry data modelling, or motion analysis from image sequences, observations contain two components of different nature. A first part is made with discrete values accounting for some symbolic information and a second part records a continuous (real-valued) measurement. We call such type of observations “mixed-state observations”. This paper introduces spatial models suited for the analysis of these kinds of data. We consider multi-parameter auto-models whose local conditional distributions belong to a mixed state exponential family. Specific examples with exponential distributions are detailed, and we present some experimental results for modelling motion measurements from video sequences.

MSC:

62H11	Directional data; spatial statistics
62H10	Multivariate distribution of statistics
62P99	Applications of statistics

Keywords:

mixed state variables; auto-models; spatial cooperation; Markov random fields

References:

[1]	Pierre Ailliot, Craig Thompson, and Peter Thompson. Space time modeling of precipitation using a hidden markov model and censored gaussian distributions. Technical report, Victoria University of Wellington, http://pagesperso.univ-brest.fr/, ailliot/doc/rain a illiot.pdf, 2006.
[2]	David J. Allcroft and Chris A. Glasbey. A latent Gaussian Markov random-field model for spatiotemporal rainfall disaggregation., J. Roy. Statist. Soc. Ser. C , 52(4):487-498, 2003. ISSN 0035-9254. · Zbl 1111.62362
[3]	Barry C. Arnold, Enrique Castillo, and José María Sarabia., Conditional specification of statistical models . Springer Series in Statistics. Springer-Verlag, New York, 1999. ISBN 0-387-98761-4.
[4]	Julian Besag. Spatial interaction and the statistical analysis of lattice systems., J. Roy. Statist. Soc. Ser. B , 36:192-236, 1974. ISSN 0035-9246. With discussion by D. R. Cox, A. G. Hawkes, P. Clifford, P. Whittle, K. Ord, R. Mead, J. M. Hammersley, and M. S. Bartlett and with a reply by the author. · Zbl 0327.60067
[5]	Patrick Bouthemy, Cécile Hardouin, Gwënaelle Piriou, and Jian-Feng Yao. Mixed-state auto-models and motion texture modeling., Journal of Mathematical Imaging and Vision , 25(3):387-402, 2006. · Zbl 1478.62184
[6]	Noel Cressie and Subhash Lele. New models for Markov random fields., J. Appl. Probab. , 29(4):877-884, 1992. ISSN 0021-9002. · Zbl 0764.60050
[7]	Ronan Fablet and Patrick Bouthemy. Motion recognition using non parametric image motion models estimated from temporal and multiscale cooccurrence statistics., IEEE Trans. on Pattern Analysis and Machine Intelligence , 25(12) :1619-1624, December 2003.
[8]	Xavier Guyon., Random fields on a network: Modeling, Statistics and Applications . Probability and its Applications (New York). Springer-Verlag, New York, 1995. ISBN 0-387-94428-1. · Zbl 0839.60003
[9]	Cécile Hardouin and Jian-Feng Yao. Multi-parameter auto-models and their application. Technical report, IRMAR/Université de Rennes 1, http://hal.archives-ouvertes.fr/hal -00154382/fr/, 2006. Forthcoming in, Biometrika . · Zbl 05198206
[10]	Mark S. Kaiser and Noel Cressie. The construction of multivariate distributions from Markov random fields., J. Multivariate Anal. , 73(2):199-220, 2000. ISSN 0047-259X. · Zbl 1065.62520
[11]	Mark S. Kaiser, Noel Cressie, and Jaehyung Lee. Spatial mixture models based on exponential family conditional distributions., Statist. Sinica , 12(2):449-474, 2002. ISSN 1017-0405. · Zbl 0998.62079
[12]	Fabien Salzenstein and Wojciech Pieczynski. Parameter estimation in hidden fuzzy markov random fields and image segmentation., CVGIP: Graph. Models Image Process. , 59:205-220, 1997.