Nonparametric and Probabilistic Classification Using NN-balls with Environmental and Remote Sensing Applications



National and international policies today require environmental follow-up systems that detect, in a quality assured way, changes over time in land use and landscape indicators. Questions related to environmental health and spatial patterns call for new statistical tools. We present in this paper some new developments on the classification of land use by using multispectral and multitemporal satellite images, based on techniques of nearest neighbour balls. The probabilistic classifiers introduced are useful for measuring uncertainty at pixel level and obtaining reliable area estimates locally. Also some theoretical considerations for the reference sample plot method (today named k-NN method in natural resource applications) are presented.


Feature Vector Feature Space Linear Discriminant Analysis Multispectral Image Area Estimate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bereau M, Dubuisson B (1991) A fuzzy extended k-nearest neighbor rule. Fuzzy Sets Sys 44:17–32MATHCrossRefGoogle Scholar
  2. 2.
    Bezdek J, Keller J, Krishnapuram R, Pal N (1999) Fuzzy models and algorithms for pattern recognition and image processing. Kluwer, NorwellMATHGoogle Scholar
  3. 3.
    Choi H, Baraniuk R (1999) Image segmentation using wavelet-domain classification. SPIE Proceedings, Denver, July 1999, vol 3816, pp 306–320Google Scholar
  4. 4.
    Ekström M (2008) Subsampling variance estimation for nonstationary spatial lattice data. Scand J Stat 35:38–63MATHCrossRefGoogle Scholar
  5. 5.
    Ekström M, Sjöstedt-de Luna S (2004) Subsampling methods to estimate the variance of sample means based on nonstationary spatial data with varying expected values. J Am Stat Assoc 99:82–95MATHCrossRefGoogle Scholar
  6. 6.
    Fix E, Hodges J (1951) Discriminatory analysis – nonparametric discrimination: consistency properties. Report 4, US Air Force School of Aviation Medicine, Random Field, TexasGoogle Scholar
  7. 7.
    Jimenez R, Yukich J (2002) Asymptotics for statistical distances based on voronoi tessellation. J Theor Prob 15:503–541MATHCrossRefMathSciNetGoogle Scholar
  8. 8.
    Katila M, Tomppo E (2002) Stratification by ancillary data in multisource forest inventories employing k nearest neighbour estimation. Can J For Res 3:1548–1561CrossRefGoogle Scholar
  9. 9.
    Kilkki P, Päivinen R (1987) Reference sample plots to combine field measurements and satellite data in forest inventory. In: Remote sensing-aided forest inventory, pp 209–212. Seminars organised by SNS and Taksaattoriklubi, Hyytil, FinlandGoogle Scholar
  10. 10.
    Kissiov V, Hadjitodorov S (1992) A fuzzy version of the k-nn method. Fuzzy Sets Sys 49: 323–329MATHCrossRefMathSciNetGoogle Scholar
  11. 11.
    Kuncheva L (2000) Fuzzy classifier design. In: Studies in fuzzness and soft computing, vol 49. Springer, HeidelbergGoogle Scholar
  12. 12.
    Loftsgaarden D, Quesenberry C (1965) A nonparametric estimate of a multivariate density function. Ann Math Stat 36:1049–1051MATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Rajski C (1964) On the normed information rate of discrete random variables. Translation of the third Prague conference, pp 583–585Google Scholar
  14. 14.
    Ranneby B, Jammalamadaka S, Teterukovskiy A (2005) The maximum spacing estimate for multivariate observations. J Stat Plan Inference 129:427–446MATHCrossRefMathSciNetGoogle Scholar
  15. 15.
    Ranneby B, Yu J (2003) Remote sensing classification of environmental indicators. In: Proceedings of 30th international symposium on remote sensing of environment – Information for risk management and sustainable development, HawaiiGoogle Scholar
  16. 16.
    Ripley B (1996) Pattern recognition and neural networks. Cambridge University Press, CambridgeMATHGoogle Scholar
  17. 17.
    Sherman M (1996) Variance estimation for statistics computed from spatial lattice data. J R Stat Soc Series B 58:509–523MATHMathSciNetGoogle Scholar
  18. 18.
    Taxt T, Lid Hjort N, Eikvil L (1991) Statistical classification using a linear mixture of two multinormal probability densities. Pattern Recognit Lett 12:731–737CrossRefGoogle Scholar
  19. 19.
    Teterukovsky A, Ranneby B (2003) The maximum spacing estimation based on dirichlet tessellation. PhD thesis, Swedish University of Agricultural Sciences. In Doctoral thesis Silvestria 277, Swedish University of Agricultural SciencesGoogle Scholar
  20. 20.
    Teterukovsky A, Yu J (2002) Contextual reclassification of multispectral images: A markov random field approach. Inf Processes 2:12–21Google Scholar
  21. 21.
    Tomppo E (1991) Satellite imagery-based national forest inventory of Finland. Int Arch Photogramm Remote Sens 28:419–424Google Scholar
  22. 22.
    Yamaguchi Y, Nagai T, Yamada H (1995) JERS-1 SAR image analysis by wavelet transform. IEICE Trans Commun E78-B(12):1617–1621Google Scholar
  23. 23.
    Yu J, Ekström M (2003) Multispectral image classification using wavelets: A simulation study. Pattern Recognit 36(4):889–898CrossRefGoogle Scholar
  24. 24.
    Yu J, Ekström M, Nilsson M (2000) Image classification using wavelets with application to forestry. In: Proceedings of the 4th international conference on methodological issues in official statistics, StockholmGoogle Scholar
  25. 25.
    Yu J, Ranneby B (2007) Nonparametric and probabilistic classification of agricultural crops using multitemporal images. J Remote Sens 11(5):748–755Google Scholar
  26. 26.
    Yu J, Ranneby B (2007) Nonparametric classification and probabilistic classifiers with environmental and remote sensing applications. In: Lai T, Shao Q, Qian L (eds) Asymptotic theory in probability and statistics with applications, Advances lectures in mathematics, vol 2. International Press, CambridgeGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  1. 1.Centre of BiostochasticsSwedish University of Agricultural SciencesUmeåSweden

Personalised recommendations