Electronic Journal of Statistics

Plugin procedure in segmentation and application to hyperspectral image segmentation

Robin Girard

Full-text: Open access

Abstract

In this work we give our contribution to the problem of segmentation with plug-in procedures. We propose general sufficient conditions under which plug in procedure are efficient. We also give an algorithm that satisfy these conditions. We apply this algorithm to hyperspectral images segmentation. Hyperspectral images are images that have both spatial and spectral coherence with thousands of spectral bands on each pixel. In the proposed procedure we combine a reduction dimension technique and a spatial regularization technique. This regularization is based on the mixlet modeling of Kolaczyck et al. [10].

Article information

Source
Electron. J. Statist., Volume 4 (2010), 655-676.

Dates
First available in Project Euclid: 2 August 2010

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1280758357

Digital Object Identifier
doi:10.1214/10-EJS567

Mathematical Reviews number (MathSciNet)
MR2678966

Zentralblatt MATH identifier
1329.62280

Subjects
Primary: 60K35: Interacting random processes; statistical mechanics type models; percolation theory [See also 82B43, 82C43]

Keywords
Segmentation mixture model penalized maximum likelihood estimation dimensionality reduction

Citation

Girard, Robin. Plugin procedure in segmentation and application to hyperspectral image segmentation. Electron. J. Statist. 4 (2010), 655--676. doi:10.1214/10-EJS567. https://projecteuclid.org/euclid.ejs/1280758357


Export citation

References

  • [1] A. Antoniadis, J. Bigot, and R. von Sachs. A multiscale approach for statistical characterization of functional images., Journal of Computational and Graphical Statistics, 18(1):216–237, 2009.
  • [2] A. Barron, L. Birgé, and P. Massart. Risk bound for model selection via penalization., probability theory and related field, 113:301–413, 1999.
  • [3] L. Birgé. Model selection via testing: an alternative to (penalized) maximum likelihood estimators., Annales de l’I.H.P. Probabilités et statistiques, 42(3):273–325, 2006.
  • [4] V. I. Bogachev., Gaussian Measures. AMS, 1998.
  • [5] O. Bousquet, S. Boucheron, and G. Lugosi. Theory of classification: a survey of recent advances., ESAIM: Probability and Statistics, 2004.
  • [6] L. Devroye, L. Gyorfi, and G. Lugosi., A probabilistic theory of pattern recognition. Springer-Verlag, 1996.
  • [7] D. Donoho. Wedgelets: Nearly-minimax estimation of edges., Annals of Statistics, pages 859–897, 1999.
  • [8] D. Donoho and I. Johnstone. Ideal spatial adaptation by wavelet shrinkage., Biometrica, 81(3):425–455, 1994.
  • [9] S. Geman and D. Nowak. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images, IEEE Trans. Pattern Anal. Mach. Intell, 6:721–741, 1984.
  • [10] E. Kolaczyk, J. Junchang, and S. Gopal. Multiscale, multigranular statistical image segmentation., JASA, 100(472) :1358, December 2005.
  • [11] E. Kolaczyk and R. Nowak. Multiscale likelihood analysis and complexity penalized estimation., Annals of Statistics, 32(2):500–527, 2004.
  • [12] Korostelev and A. Tsybacov., Minimax Theory of Image Reconstruction, volume 82 of Lecture Notes In Statistics. Springer-Verlag, 1993.
  • [13] Q. J. Li., Estimation of mixture Models. PhD thesis, Yale university, 1999.
  • [14] F. Schmidt., Classification de la surface de Mars par imagerie hyperspectrale OMEGA. Suivi spatio-temporel et études des dépôts saisonniers de CO2 et H2O. PhD thesis, UJF, 2007.