Abstract
We consider the problem of simulating a multigaussian random vector observed through non linear and/or corrupted sensors. To sample the conditional distribution, linear methods such as conditional kriging or sequential Gaussian simulation can be applied only to the case of linear observations. But more complex observations can be integrated into stochastic imaging via the Gibbs sampler: the conditional distribution is sampled as the stationary distribution of a Markov chain. This is illustrated on three examples: inequality constraints, Poisson noise and 1D convolution.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
F. Alabert. Stochastic imaging of spatial distributions using hard and soft data. Msc. thesis, Stanford U, 1987.
D. Allard. Simulation of a geological lithofacies with respect of connectivity information using the truncated gaussian model. In M. Armstrong, editor, Workshop on geostatistical simulations, volume 1, pages 1–2, Dordrecht Holland, 1994. Kluwer Academic Press.
J. Besag. On statistical analysis of dirty pictures. J. R. Statist. Soc. B, 48:259–302, 1986.
J. Besag, J. York, and A. Mollié. Bayesian image restoration, with two applications in spatial statistics. Ann. Inst. Statis. Math., 43(1):1–59, 1991.
L.J. Bortoli, F. Alabert, A. Haas, and A. Journel. Constraining stochastic images to seismic data. In A. Soares, editor, Geostatistics TRÓIA’92, volume 1, pages 325–337, Dordrecht Holland, 1993. Kluwer Academic Press.
C. Daly. Applications de la Géostatistique à quelques problèmes de filtrage. thèse de docteur en géostatistique, ENSMP, Fontainebleau, France, 1991.
C. Daly. Filtering non-linear functions. In A. Soares, editor, Geostatistics TRÓIA’92, volume 1, pages 61–72, Dordrecht Holland, 1993. Kluwer Academic Press.
M. W. Davis. Production of conditional simulation via lu decomposition of the covariance matrix. Math. Geol., 19:91–98, 1987.
X. Freulon. Conditionnement du modèle gaussien par des inégalités ou des randomisées. thèse de docteur en géostatistique, ENSMP, Fontainebleau, France, 1992.
X. Freulon and Ch. de Fouquet. Conditioniing a gaussian model with inequalities. In A. Soares, editor, Geostatistics TRÓIA’92, volume 1, pages 201–212, Dordrecht Holland, 1993. Kluwer Academic Press.
S. Geman and D. Geman. Stochastic relaxation, gibbs distribution and the bayesian restoration of images. I.E.E.E. transactions: Pattern Analysis and Machine Intelligence, 6:721–741, 1984.
A. G. Journel. Simulations conditionnelles de gisements miniers. thèse de docteur-ingénieur, Université de Nancy-I, Nancy, France, 1977.
Ch. Lantuéjoul. Non conditional simulation of a gaussian random function by the turning band method. In M. Armstrong, editor, Workshop on geostatistical simulations, volume 1, pages 1–2, Dordrecht Holland, 1994. Kluwer Academic Press.
G. Matheron et al. Conditional simulation of the geometry of fluviodeltaic reservoirs. SPE, 1987. 62nd Annual Conference, Dallas, Texas.
L. Younes. Probleme d’estimation parametrique pour les champs de Gibbs Markoviens — Application au traitement d’images. Thèse, Université d’Orsay, 1988.
Z. Zhang and A. Galli. Getting better quality seismic sections. In A. Soares, editor, Geostatistics TRÓIA’92, volume 1, pages 285–297, Dordrecht Holland, 1993. Kluwer Academic Press.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1994 Springer Science+Business Media Dordrecht
About this paper
Cite this paper
Freulon, X. (1994). Conditional simulation of a Gaussian random vector with non linear and/or noisy observations. In: Armstrong, M., Dowd, P.A. (eds) Geostatistical Simulations. Quantitative Geology and Geostatistics, vol 7. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-8267-4_5
Download citation
DOI: https://doi.org/10.1007/978-94-015-8267-4_5
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4372-6
Online ISBN: 978-94-015-8267-4
eBook Packages: Springer Book Archive