Environmental and Ecological Statistics

, Volume 22, Issue 3, pp 571–600 | Cite as

Recent Bayesian approaches for spatial analysis of 2-D images with application to environmental modelling

  • M. G. Falk
  • C. L. Alston
  • C. A. McGrory
  • S. Clifford
  • E. A. Heron
  • D. Leonte
  • M. Moores
  • C. D. Walsh
  • A. N. Pettitt
  • K. L. Mengersen
Article

Abstract

From remote sensing of the environment, to brain scans in medicine, the growth in the use of image data has motivated a parallel increase in statistical techniques for analysing these images. A particular area of growth has been in Bayesian models and corresponding computational methods. Bayesian approaches have been proposed to address the gamut of supervised and unsupervised inferential aims in image analysis. In this article we provide a general review of these approaches, with a focus on unsupervised analysis of 2-D images. Four exemplar methods that canvas the broad aims of image modelling and analysis are described. An exposition of these approaches is provided by applying them to an environmental case study involving the use of satellite data to assess water quality in the Great Barrier Reef, Australia. The techniques considered in detail are hidden Markov random fields (MRF), Gaussian MRF, Poisson/gamma random fields, and Voronoi tessellations. We also consider a variety of enabling computational algorithms, including MCMC, variational Bayes and integrated nested Laplace approximations. We compare the different aims and inferential capabilities of the models and discuss the advantages and drawbacks of the corresponding computational algorithms.

Keywords

Hidden Markov random field Spatial mixture models  Poisson/Gamma random field Integrated nested Laplace approximation Variational Bayes Voronoi tessellations 

Notes

Acknowledgments

The authors would like to thank Dr. Andy Steven, CSIRO, for the data used in this paper. The authors would also like to thank two anonymous reviewers for their comments which greatly improved the manuscript. The research was funded from an Australian Research Council (ARC) Linkage Grant (LP0668185) and by the ARC Centre of Excellence for Complex Dynamic Systems and Control (CDSC).

References

  1. Alston C, Mengersen K (2010) Allowing for the effect of data binning in a Bayesian normal mixture model. Comput Stat Data Anal 54(4):916–923CrossRefGoogle Scholar
  2. Alston CL, Mengersen KL, Thompson JM, Littlefield PJ, Perry D, Ball AJ (2005) Extending the Bayesian mixture model to incorporate spatial information in analysing sheep CAT scan images. Aust J Agric Res 56(4):373–388CrossRefGoogle Scholar
  3. Antoniak C (1974) Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems. Ann Stat 2(6):1152–1174CrossRefGoogle Scholar
  4. Banerjee S, Carlin B, Gelfand A (2004) Hierarchical modeling and analysis for spatial data. Chapman & Hall/CRC Press, Boca RatonGoogle Scholar
  5. Beal MJ, Ghahramani Z (2003) The variational Bayesian EM algorithm for incomplete data: with application to scoring graphical model structures. In: Proceedings of the seventh Valencia international meeting, volume 7 of Bayesian statistics. Oxford University Press, London, pp 453–464Google Scholar
  6. Besag J (1974) Spatial interaction and statistical analysis of lattice systems (with discussion). J R Stat Soc Ser B Stat Methodol 36(2):192–236Google Scholar
  7. Besag J (1975) Statistical analysis of non-lattice data. J R Stat Soc D 24(3):179–195Google Scholar
  8. Besag J (1986) On the statistical analysis of dirty pictures. JRSS B 48(3):259–302Google Scholar
  9. Best NG, Ickstadt K, Wolpert RL (2000) Spatial Poisson regression for health and exposure data measured at disparate resolutions. J Am Stat Assoc 95(452):1076–1088CrossRefGoogle Scholar
  10. Blondeau-Patissier D, Brando VE, Oubelkheir K, Dekker AG, Clementson LA, Daniel P (2009) Bio-optical variability of the absorption and scattering properties of the Queensland inshore and reef waters, Australia. J Geophys Res 114:C05003. doi:10.1029/2008JC005039 Google Scholar
  11. Brando VE, Dekker AG, Schroeder T, Park YJ, Clementson LA, Steven A, Blondeau-Patissier D (2008) Satellite retrieval of chlorophyll CDOM and NAP in optically complex waters using a semi-analytical inversion based on specific inherent optical properties. A case study for Great Barrier Reef coastal waters. In: Proceedings of the XIX ocean optics conferenceGoogle Scholar
  12. Chan KS (1993) Asymptotic behavior of the Gibbs sampler. J Am Stat Assoc 88(421):320–326Google Scholar
  13. Cox DR (1955) Some statistical methods connected with series of events. JRSS B 17(2):129–164Google Scholar
  14. Cressie N, Wikle C (2011) Statistics for spatio-temporal data. Wiley, New YorkGoogle Scholar
  15. Cucala J, Marin JM, Robert CP, Titterington DM (2009) A Bayesian reassessment of nearest-neighbour classification. J Am Stat Assoc 104(485):263–273CrossRefGoogle Scholar
  16. de Berg M, van Kreveld M, Overmars M, Schwarzkopf O (2000) Computational geometry: algorithms and applications, 2nd edn. Springer, BerlinCrossRefGoogle Scholar
  17. Dempster A, Laird N, Rubin D (1977) Maximum likelihood from incomplete data via the EM algorithm. JRSS B 39(1):1–38Google Scholar
  18. Denham R, Falk M, Mengersen K (2011) The Bayesian conditional independence model for measurement error: applications in ecology. Environ Ecol Stat 18:239–255CrossRefGoogle Scholar
  19. Denison DGT, Adams NM, Holmes CC, Hand DJ (2002) Bayesian partition modelling. Comput Stat Data Anal 38(4):475–485CrossRefGoogle Scholar
  20. Denison DGT, Holmes CC (2001) Bayesian partitioning for estimating disease risk. Biometrics 57(1):143–149CrossRefPubMedGoogle Scholar
  21. Engen S (1978) Stochastic abundance models, with emphasis on biological communities and species diversity., Monographs on applied probability and statisticsChapman and Hall, LondonCrossRefGoogle Scholar
  22. Ferguson T (1973) A Bayesian analysis of some nonparametric problems. Ann Stat 1(2):209–230CrossRefGoogle Scholar
  23. Friel N, Pettitt AN (2004) Likelihood estimation and inference for the autologistic model. J Comput Graph Stat 13(1):232–246CrossRefGoogle Scholar
  24. Gelfand A (1996) Model determination using sampling based methods. In: Gilks W, Richardson S, Spiegehalter D (eds) Markov Chain Monte Carlo in Practice, chap 9. Chapman and Hall, Boca Raton, FLGoogle Scholar
  25. Gelfand AE, Smith AFM (1990) Sampling-based approaches to calculating marginal densities. J Am Stat Assoc 85(410):398–409Google Scholar
  26. Geman S, Geman D (1984) Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images. IEEE Trans PAMI 6:721–741CrossRefGoogle Scholar
  27. Gilks W, Richardson S, Spiegelhalter D (eds) (1996) Markov chain Monte Carlo in practice., Interdisciplinary statistics seriesChapman & Hall/CRC Press, Boca RatonGoogle Scholar
  28. Goldsmith J, Wand MP, Crainceanu C (2011) Functional regression via variational Bayes. Electron J Stat 5:572–602PubMedCentralCrossRefPubMedGoogle Scholar
  29. Green P, Richardson S (2002) Hidden Markov models and disease mapping. J Am Stat Assoc 97(460):1055–1070CrossRefGoogle Scholar
  30. Green PJ (1995) Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 82(4):711–732CrossRefGoogle Scholar
  31. Grenander U (1983) Tutorial in Pattern Theory. Technical report, Division of Applied Mathematics. Brown UniversityGoogle Scholar
  32. Grenander U, Miller M (2007) Pattern theory: from representation to inference. Oxford University Press, OxfordGoogle Scholar
  33. Hastings WK (1970) Monte Carlo sampling methods using Markov chains and their applications. Biometrika 57(1):97–109CrossRefGoogle Scholar
  34. Heron EA, Walsh CD (2008) A continuous latent spatial model for crack initiation in bone cement. J R Stat Soc Ser C Appl Stat 57(1):25–42CrossRefGoogle Scholar
  35. Ickstadt K, Wolpert R (1999) Spatial regression for marked point processes (with discussion). In: Bernado J, Berger J, Dawid A, Smith A (eds) Bayesian statistics 6. Oxford University Press, Oxford, pp 323–341Google Scholar
  36. Illian JB, Sørbye SH, Rue H (2012) A toolbox for fitting complex spatial point process models using integrated nested Laplace approximation (INLA). Ann Appl Stat 6(4):1499–1530Google Scholar
  37. Jordan M, Ghahramani Z, Jaakkola T, Saul L (1999) An introduction to variational methods for graphical models. Mach Learn 37:183–233CrossRefGoogle Scholar
  38. Lang S, Brezger A (2004) Bayesian P-Splines. J Comput Graph Stat 13(1):183–212CrossRefGoogle Scholar
  39. Leonte D, Nott DJ (2006) Bayesian spatial modelling of gamma ray count data. Math Geol 38(2):135–154CrossRefGoogle Scholar
  40. Liang S, Banerjee S, Carlin BP (2009) Bayesian wombling for spatial point processes. Biometrics 65(4):1243–1253PubMedCentralCrossRefPubMedGoogle Scholar
  41. Lindgren F, Rue H, Lindström J (2011) An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach. J R Stat Soc B 73:423–498Google Scholar
  42. Lu H, Carlin BP (2005) Bayesian areal wombling for geographical boundary analysis. Geogr Anal 37(3):265–285CrossRefGoogle Scholar
  43. Ma H, Carlin BP (2007) Bayesian multivariate areal wombling for multiple disease boundary analysis. Bayesian Anal 2(2):281–302CrossRefGoogle Scholar
  44. Marin J-M, Robert C (2007) Bayesian core: a practical approach to computational Bayesian statistics., Springer Texts in StatisticsSpringer, BerlinGoogle Scholar
  45. Martin A, Quinn K, Park J (2011) MCMCpack: Markov chain Monte Carlo in R. J Stat Softw 42(9):1–21Google Scholar
  46. Matérn B (1960) Spatial variation. stochastic models and their application to some problems in forest surveys and other sampling investigations. Meddelanden fran statens Skogsforskningsinstitut 49:1–144Google Scholar
  47. McGrory CA, Titterington DM, Reeves R, Pettitt AN (2009) Variational Bayes for estimating the parameters of a hidden Potts model. Stat Comput 19(3):329–340CrossRefGoogle Scholar
  48. Mengersen KL, Tweedie RL (1996) Rates of convergence of the Hastings and Metropolis algorithms. Ann Stat 24(1):101–121CrossRefGoogle Scholar
  49. Metropolis N, Rosenbluth AW, Rosenbluth MN, Teller AH, Teller E (1953) Equation of state calculations by fast computing machines. J Chem Phys 21(6):1087–1092CrossRefGoogle Scholar
  50. Møller J, Pettitt AN, Reeves R, Berthelsen KK (2006) An efficient Markov chain monte carlo method for distributions with intractable normalising constants. Biometrika 93(2):451–458CrossRefGoogle Scholar
  51. Møller J, Syversveen A, Waagepetersen R (1998) Log Gaussian Cox processes. Scand J Stat 25(3):451–482CrossRefGoogle Scholar
  52. Møller J, Waagepetersen R (2004) Statistical inference and simulation for spatial point processes. Chapman & Hall/CRC Press, Boca RatonGoogle Scholar
  53. Neal RM (2003) Slice sampling. Ann Stat 31(3):705–741CrossRefGoogle Scholar
  54. Nicholls GK (1998) Bayesian image analysis with Markov chain Monte Carlo and coloured continuum triangulation models. JRSS B 60(3):643–659CrossRefGoogle Scholar
  55. Okabe A, Boots B, Sugihara K, Chiu SN (2000) Spatial tessellations: concepts and applications of Voronoi diagrams, 2nd edn. Wiley, ChichesterCrossRefGoogle Scholar
  56. Orbanz P, Buhmann, J (2006) Smooth image segmentation by nonparametric Bayesian inference. In: Proceedings of the ECCV 2006, volume 3951 of Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, pp 444–457Google Scholar
  57. Ormerod JT, Wand MP (2010) Explaining variational approximations. Am Stat 64(2):140–153CrossRefGoogle Scholar
  58. O’Rourke J (1998) Computational geometry in C, 2nd edn. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  59. Pitman J, Yor M (1997) The two-parameter Poisson-Dirichlet distribution derived from a stable subordinator. Ann Probab 25(2):855–900CrossRefGoogle Scholar
  60. Plummer M (2003) JAGS: a program for analysis of Bayesian graphical models using Gibbs sampling. In Hornik K, Leisch F, Zeileis A (eds) Proceedings of the 3rd international workshop on distributed statistical computingGoogle Scholar
  61. Potts RB (1952) Some generalized order-disorder transitions. Proc Camb Philos Soc 48:106–109CrossRefGoogle Scholar
  62. Robert C, Casella G (2004) Monte Carlo statistical methods, 2nd edn., Springer Texts in StatisticsSpringer, New YorkCrossRefGoogle Scholar
  63. Rue H (2001) Fast sampling of Gaussian Markov random fields. JRSS B 63(2):325–338CrossRefGoogle Scholar
  64. Rue H, Held L (2005) Gaussian Markov random fields: theory and applications, volume 104 of Monographs on statistics and applied probability. Chapman & Hall/CRC Press, Boca RatonCrossRefGoogle Scholar
  65. Rue H, Martino S (December, 2010) INLA: functions which allow to perform a full Bayesian analysis of structured additive models using Integrated Nested Laplace Approximaxion (0.0 ed.)Google Scholar
  66. Rue H, Martino S, Chopin N (2009) Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations. J R Stat Soc B 71:319–392CrossRefGoogle Scholar
  67. Rue H, Tjelmeland H (2002) Fitting Gaussian Markov random fields to Gaussian fields. Scand J Stat 29(1):31–49CrossRefGoogle Scholar
  68. Rydén T, Titterington D (1998) Computational Bayesian analysis of hidden Markov models. J Comput Graph Stat 7(2):194–211Google Scholar
  69. Shyr A, Darrell T, Jordan M, Urtasun R (2011) Supervised hierarchical Pitman-Yor process for natural scene segmentation. In: Proceedings of the IEEE CVPR, pp 2281–2288Google Scholar
  70. Smith M, Kohn R (1996) Nonparametric regression using Bayesian variable selection. J Econom 75(2):317–343CrossRefGoogle Scholar
  71. Strickland C, Denham R, Alston C, Mengersen K (2012) A Python package for Bayesian estimation using Markov chain Monte Carlo. In: Alston C, Mengersen K, Pettitt A (eds) Case atudies in Bayesian statistical modelling and analysis. Wiley, New York, pp 421–460CrossRefGoogle Scholar
  72. Sudderth E, Jordan M (2009) Shared segmentation of natural scenes using dependent Pitman-Yor processes. Proc NIPS 21:1585–1592Google Scholar
  73. Tanner MA, Wong WH (1987) The calculation of posterior distributions by data augmentation. J Am Stat Assoc 82(398):528–540Google Scholar
  74. Teh Y, Jordan M, Beal M, Blei D (2006) Hierarchical Dirichlet processes. J Am Stat Assoc 101(476):1566–1581Google Scholar
  75. Thomas A, Best N, Lunn D, Spiegelhalter D (2012) The BUGS book: a practical introduction to Bayesian analysis, volume 98 of Chapman & Hall/CRC Texts in Statistical Science. Taylor & Francis, LondonGoogle Scholar
  76. Tierney L, Kadane JB (1986) Accurate approximations for posterior moments and marginal densities. J Am Stat Assoc 81(393):82–86Google Scholar
  77. Winkler G (2003) Image analysis, random fields and Markov Chain Monte Carlo methods: a mathematical introduction 2nd edn. Springer, BerlinCrossRefGoogle Scholar
  78. Wolpert RL, Ickstadt K (1998a) Poisson/gamma random field models for spatial statistics. Biometrika 85(2):251–267CrossRefGoogle Scholar
  79. Wolpert RL, Ickstadt K (1998b) Simulation of Lévy Random Fields. In: Müller P, Dey D, Sinha D (eds) Practical nonparametric and semiparametric Bayesian statistics. Springer, Berlin, pp 227–242CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • M. G. Falk
    • 1
  • C. L. Alston
    • 1
  • C. A. McGrory
    • 1
  • S. Clifford
    • 1
  • E. A. Heron
    • 1
    • 2
  • D. Leonte
    • 1
    • 3
  • M. Moores
    • 1
    • 4
  • C. D. Walsh
    • 1
    • 5
  • A. N. Pettitt
    • 1
  • K. L. Mengersen
    • 1
  1. 1.Mathematical Sciences, Science and EngineeringQueensland University of TechnologyBrisbaneAustralia
  2. 2.Department of PsychiatryTrinity College DublinDublin 2Ireland
  3. 3.National Industry Chemicals Notification and Assessment SchemeSydneyAustralia
  4. 4.Department of StatisticsUniversity of WarwickCoventryUK
  5. 5.Department of StatisticsTrinity College DublinDublin 2Ireland

Personalised recommendations