Advertisement

Ensemble Strategies for Classifying Hyperspectral Remote Sensing Data

  • Xavier Ceamanos
  • Björn Waske
  • Jón Atli Benediktsson
  • Jocelyn Chanussot
  • Johannes R. Sveinsson
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5519)

Abstract

The classification of hyperspectral imagery, using multiple classifier systems is discussed and an SVM-based ensemble is introduced. The data set is separated into separate feature subsets using the correlation between the different spectral bands as a criterion. Afterwards, each source is classified separately by an SVM classifier. Finally, the different outputs are used as inputs for final decision fusion that is based on an additional SVM classifier. The results using the proposed strategy are compared to classification results achieved by a single SVM and other well known classifier ensembles, such as random forests, boosting and bagging.

Keywords

hyperspectral land cover classification support vector machines multiple classifier systems classifier ensmeble 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Hughes, G.F.: On the mean accuracy of statistical pattern recognizers. IEEE Trans. on Information Theory 14, 55–63 (1968)CrossRefGoogle Scholar
  2. 2.
    Melgani, F., Bruzzone, L.: Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. and Remote Sens. 42, 1778–1790 (2004)CrossRefGoogle Scholar
  3. 3.
    Pal, M., Mather, P.M.: Some issues in the classification of DAIS hyperspectral data. Int. J. Remote Sens. 27, 2896–2916 (2006)Google Scholar
  4. 4.
    Vapnik, V.N.: Statistical Learning Theory. Wiley, New York (1998)zbMATHGoogle Scholar
  5. 5.
    Benediktsson, J.A., Chanussot, J., Fauvel, M.: Multiple Classifier Systems in Remote Sensing: From Basics to Recent Developments. In: Haindl, M., Kittler, J., Roli, F. (eds.) MCS 2007. LNCS, vol. 4472, pp. 501–512. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  6. 6.
    Ham, J., Chen, Y.C., Crawford, M.M., Ghosh, J.: Investigation of the random forest framework for classification of hyperspectral data. IEEE Trans. on Geosci. and Remote Sens. 43, 492–501 (2005)CrossRefGoogle Scholar
  7. 7.
    Joelsson, S.R., Benediktsson, J.A., Sveinsson, J.R.: Random forest classification of remote sensing data. In: Chen, C.H. (ed.) Signal and Image Processing for Remote Sensing, pp. 327–344. CRC Press, Boca Raton (2007)Google Scholar
  8. 8.
    Cheung-Wai Chan, J., Paelinckx, D.: Evaluation of Random Forest and Adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery. Remote Sens. of Env. 112, 2999–3011 (2008)CrossRefGoogle Scholar
  9. 9.
    Waske, B., Benediktsson, J.A.: Fusion of Support Vector Machines for Classification of Multisensor Data. IEEE Trans. Geosci. and Remote Sens. 45, 3858–3866 (2007)CrossRefGoogle Scholar
  10. 10.
    Benediktsson, J.A., Kanellopoulos, I.: Classification of Multisource and Hyperspectral Data Based on Decision Fusion. IEEE Trans. Geosci. and Remote Sens. 37, 1367–1377 (1999)CrossRefGoogle Scholar
  11. 11.
    Liao, Y., Moody, J.: Constructing heterogeneous committees using input feature grouping. In: Solla, S.A., Leen, T.K., Muller, K.-R. (eds.) Advances in Neural Information Processing Systems (NIPS) Conference, 1999, Denver, USA, vol. 12. MIT Press, Cambridge (2000)Google Scholar
  12. 12.
    Burges, C.J.C.: A Tutorial on Support Vector Machines for Pattern Recognition. Data Mining and Knowledge Discovery 2, 121–167 (1998)CrossRefGoogle Scholar
  13. 13.
    Polikar, R.: Ensemble Based Systems in Decision Making. IEEE Circuits and Systems Magazine 6, 21–45 (2006)CrossRefGoogle Scholar
  14. 14.
    Briem, G.J., Benediktsson, J.A., Sveinsson, J.R.: Multiple Classifiers Applied to Multisource Remote Sensing Data. IEEE Trans. Geosci. Remote Sens. 40, 2291–2299 (2002)CrossRefGoogle Scholar
  15. 15.
    Waske, B., van der Linden, S.: Classifying multilevel imagery from SAR and optical sensors by decision fusion. IEEE Trans. on Geosci. and Remote Sens. 46, 1457–1466 (2008)CrossRefGoogle Scholar
  16. 16.
    Freund, Y., Schapire, R.E.: Experiments with a new boosting algorithm. In: 13th International Conference of Machine Learning, Bari, Italy (1996)Google Scholar
  17. 17.
    Breiman, L.: Bagging predictors. Mach. Learning 24, 123–140 (1996)zbMATHGoogle Scholar
  18. 18.
    Breiman, L.: Random forests. Mach. Learning 45, 5–32 (2001)CrossRefzbMATHGoogle Scholar
  19. 19.
    Witten, I.H., Eibe, F.: Data Mining: Practical machine learning tools and techniques, 2nd edn. Morgan Kaufmann, San Francisco (2005)zbMATHGoogle Scholar
  20. 20.
    Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines (2001), http://www.csie.ntu.edu.tw/~cjlin/libsvm

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Xavier Ceamanos
    • 1
  • Björn Waske
    • 2
  • Jón Atli Benediktsson
    • 2
  • Jocelyn Chanussot
    • 1
  • Johannes R. Sveinsson
    • 2
  1. 1.GIPSA-LAB, Signal & Image DepartmentGrenoble Institute of Technology, INPGSaint Martin d’HèresFrance
  2. 2.Faculty of Electrical and Computer EngineeringUniversity of IcelandReykjavikIceland

Personalised recommendations