GMM Supervectors for Limited Training Data in Hyperspectral Remote Sensing Image Classification

  • AmirAbbas DavariEmail author
  • Vincent Christlein
  • Sulaiman Vesal
  • Andreas Maier
  • Christian Riess
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10425)


Severely limited training data is one of the major and most common challenges in the field of hyperspectral remote sensing image classification. Supervised learning on limited training data requires either (a) designing a highly capable classifier that can handle such information scarcity, or (b) designing a highly informative and easily separable feature set. In this paper, we adapt GMM supervectors to hyperspectral remote sensing image features. We evaluate the proposed method on two datasets. In our experiments, inclusion of GMM supervectors leads to a mean classification improvement of about \(4.6\%\).


Hyperspectral image classification Remote sensing Limited training data GMM supervector 


  1. 1.
    Bahari, M.H., Saeidi, R., Hamme, H.V., Leeuwen, D.V.: Accent recognition using i-vector, Gaussian mean supervector and Gaussian posterior probability supervector for spontaneous telephone speech. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Institute of Electrical and Electronics Engineers. IEEE, May 2013Google Scholar
  2. 2.
    Breiman, L.: Random forests. Mach. Learn. 45(1), 5–32 (2001)CrossRefzbMATHGoogle Scholar
  3. 3.
    Bruzzone, L., Chi, M., Marconcini, M.: A novel transductive svm for semisupervised classification of remote-sensing images. IEEE Trans. Geosci. Remote Sens. 44(11), 3363–3373 (2006)CrossRefGoogle Scholar
  4. 4.
    Castaings, T., Waske, B., Atli Benediktsson, J., Chanussot, J.: On the influence of feature reduction for the classification of hyperspectral images based on the extended morphological profile. Int. J. Remote Sens. 31(22), 5921–5939 (2010)CrossRefGoogle Scholar
  5. 5.
    Cerva, P., Silovsky, J., Zdansky, J.: Comparison of generative and discriminative approaches for speaker recognition with limited data. Radioengineering 18(3), 307–316 (2009)Google Scholar
  6. 6.
    Chi, M., Feng, R., Bruzzone, L.: Classification of hyperspectral remote-sensing data with primal SVM for small-sized training dataset problem. Adv. Space Res. 41(11), 1793–1799 (2008)CrossRefGoogle Scholar
  7. 7.
    Christlein, V., Bernecker, D., Hönig, F., Maier, A., Angelopoulou, E.: Writer identification using GMM supervectors and exemplar-SVMs. Pattern Recogn. 63, 258–267 (2017)CrossRefGoogle Scholar
  8. 8.
    Dalla Mura, M., Atli Benediktsson, J., Waske, B., Bruzzone, L.: Extended profiles with morphological attribute filters for the analysis of hyperspectral data. Int. J. Remote Sens. 31(22), 5975–5991 (2010)CrossRefGoogle Scholar
  9. 9.
    Dalla Mura, M., Benediktsson, J.A., Waske, B., Bruzzone, L.: Morphological attribute profiles for the analysis of very high resolution images. IEEE Trans. Geosci. Remote Sens. 48(10), 3747–3762 (2010)CrossRefGoogle Scholar
  10. 10.
    Dempster, A., Laird, N., Rubin, D.: Maximum likelihood from incomplete data via the EM algorithm. J. Roy. Stat. Soc.: Ser. B (Methodol.) 39(1), 1–38 (1977)MathSciNetzbMATHGoogle Scholar
  11. 11.
    Fukunaga, K.: Introduction to Statistical Pattern Recognition. Academic Press, San Diego (2013)zbMATHGoogle Scholar
  12. 12.
    Hoffbeck, J.P., Landgrebe, D.A.: Covariance matrix estimation and classification with limited training data. IEEE Trans. Pattern Anal. Mach. Intell. 18(7), 763–767 (1996)CrossRefGoogle Scholar
  13. 13.
    Hu, M.-K.: Visual pattern recognition by moment invariants. IRE Trans. Inf. Theory 8(2), 179–187 (1962)CrossRefzbMATHGoogle Scholar
  14. 14.
    Huang, X., Guan, X., Benediktsson, J.A., Zhang, L., Li, J., Plaza, A., Dalla Mura, M.: Multiple morphological profiles from multicomponent-base images for hyperspectral image classification. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 7(12), 4653–4669 (2014)CrossRefGoogle Scholar
  15. 15.
    Hughes, G.: On the mean accuracy of statistical pattern recognizers. IEEE Trans. Inf. Theory 14(1), 55–63 (1968)CrossRefGoogle Scholar
  16. 16.
    Jackson, Q., Landgrebe, D.A.: An adaptive classifier design for high-dimensional data analysis with a limited training data set. IEEE Trans. Geosci. Remote Sens. 39(12), 2664–2679 (2001)CrossRefGoogle Scholar
  17. 17.
    Kelly, F.: Automatic recognition of ageing speakers. Ph.D. thesis, Trinity College Dublin (2014)Google Scholar
  18. 18.
    Kuo, B.-C., Landgrebe, D.A.: Nonparametric weighted feature extraction for classification. IEEE Trans. Geosci. Remote Sens. 42(5), 1096–1105 (2004)CrossRefGoogle Scholar
  19. 19.
    Landgrebe, D.A.: Signal Theory Methods in Multispectral Remote Sensing, vol. 29. Wiley, Hoboken (2005)Google Scholar
  20. 20.
    Lee, C., Landgrebe, D.A.: Feature extraction based on decision boundaries. IEEE Trans. Pattern Anal. Mach. Intell. 15(4), 388–400 (1993)CrossRefGoogle Scholar
  21. 21.
    Liu, T., Gu, Y., Jia, X., Benediktsson, J.A., Chanussot, J.: Class-specific sparse multiple kernel learning for spectral-spatial hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 54(12), 7351 (2016)CrossRefGoogle Scholar
  22. 22.
    McLachlan, G., Peel, D.: Finite Mixture Models. Wiley, Hoboken (2004)zbMATHGoogle Scholar
  23. 23.
    Oliveira-Brochado, A., Martins, F.V.: Assessing the number of components in mixture models: a review. Technical report, Universidade do Porto, Faculdade de Economia do Porto (2005)Google Scholar
  24. 24.
    Reynolds, D.A., Quatieri, T.F., Dunn, R.B.: Speaker verification using adapted Gaussian mixture models. Digit. Signal Proc. 10(1–3), 19–41 (2000)CrossRefGoogle Scholar
  25. 25.
    Salembier, P., Oliveras, A., Garrido, L.: Antiextensive connected operators for image and sequence processing. IEEE Trans. Image Process. 7(4), 555–570 (1998)CrossRefGoogle Scholar
  26. 26.
    Soille, P.: Constrained connectivity for hierarchical image partitioning and simplification. IEEE Trans. Pattern Anal. Mach. Intell. 30(7), 1132–1145 (2008)CrossRefGoogle Scholar
  27. 27.
    Srinivasan, B.V., Zotkin, D.N., Duraiswami, R.: A partial least squares framework for speaker recognition. In: 2011 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP. Institute of Electrical and Electronics Engineers (IEEE), May 2011Google Scholar
  28. 28.
    Tadjudin, S., Landgrebe, D.A.: Covariance estimation for limited training samples. In: 1998 Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, IGARSS 1998, vol. 5, pp. 2688–2690. IEEE (1998)Google Scholar
  29. 29.
    Valero, S., Salembier, P., Chanussot, J.: Hyperspectral image representation and processing with binary partition trees. IEEE Trans. Image Process. 22(4), 1430–1443 (2013)MathSciNetCrossRefGoogle Scholar
  30. 30.
    Vatsavai, R.R., Shekhar, S., Burk, T.E.: A semi-supervised learning method for remote sensing data mining. In: 2005 Proceedings of the 17th IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2005, IEEE (2005). 5 ppGoogle Scholar
  31. 31.
    Xu, M., Zhou, X., Li, Z., Dai, B., Huang, T.S.: Extended hierarchical Gaussianization for scene classification. In: 2010 17th IEEE International Conference on Image Processing (ICIP), Hong Kong, pp. 1837–1840, September 2010Google Scholar
  32. 32.
    Xu, X., Li, J., Huang, X., Dalla Mura, M., Plaza, A.: Multiple morphological component analysis based decomposition for remote sensing image classification. IEEE Trans. Geosci. Remote Sens. 54(5), 3083–3102 (2016)CrossRefGoogle Scholar
  33. 33.
    Zapata-Zapata, G.J., Arias-Londoño, J.D., Vargas-Bonilla, J.F., Orozco-Arroyave, J.R.: On-line signature verification using gaussian mixture models and small-sample learning strategies. Revista Facultad de Ingeniería Universidad de Antioquia 79, 86–97 (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • AmirAbbas Davari
    • 1
    Email author
  • Vincent Christlein
    • 1
  • Sulaiman Vesal
    • 1
  • Andreas Maier
    • 1
  • Christian Riess
    • 1
  1. 1.Pattern Recognition Lab, Computer Science DepartmentFriedrich-Alexander-University Erlangen-NurembergErlangenGermany

Personalised recommendations