Kernel Oblique Subspace Projection Approach for Target Detection in Hyperspectral Imagery

  • Liaoying Zhao
  • Yinhe Shen
  • Xiaorun Li
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6319)

Abstract

In this paper, a kernel-based nonlinear version of the oblique subspace projection (OBSP) operator is defined in terms of kernel functions. Input data are implicitly mapped into a high-dimensional kernel feature space by a nonlinear mapping, which is associated with a kernel function. The OBSP expression is then derived in the feature space, which is kernelized in terms of the kernel functions in order to avoid explicit computation in the high-dimensional feature space. The resulting kernelized OBSP algorithm is equivalent to a nonlinear OBSP in the original input space. Experimental results based on simulated hyperspectral data and real hyperspectral imagery shows that the kernel oblique subspace projection (KOBSP) outperforms the conventional OBSP.

Keywords

kernel function kernel oblique subspace projection hyperspectral image 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Reed, I.S., Yu, X.: Addaptive multiple-band cfar detection of an optical pattern with unknown spectral distribution. IEEE Trans. Acountics. Speech and Signal Process 38(10), 1760–1770 (1990)CrossRefGoogle Scholar
  2. 2.
    Harsanyi, J.C., Chang, C.-I.: Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach. IEEE Trans. on Geoscience and Remote Sensing 32(4), 779–785 (1994)CrossRefGoogle Scholar
  3. 3.
    Scharf, L.L., Friedlander, B.: Matched subspace detectors. IEEE Trans. Signal Process. 42(8), 2146–2157 (1994)CrossRefGoogle Scholar
  4. 4.
    Manolakis, D., Shaw, G., Keshava, N.: Comparative analysis of hyperspectral adaptive matched filter detectors. In: Algorithms for Multispectral, Hyperspectral, and Ultraspectral Imagery VI, Orlando,Fla, USA. Proceedings of SPIE, vol. 4049, pp. 2–17 (April 2000)Google Scholar
  5. 5.
    Kraut, S., Scharf, L.L., McWhorter, T.: Adaptive subspace detectors. IEEE Trans. Signal Process. 49(1), 1–16 (2001)CrossRefGoogle Scholar
  6. 6.
    Kraut, S., Scharf, L.L.: The CFAR adaptive subspace detectoris a scale-invariant GLRT. IEEE Transactions on SignalProcessing 47(9), 2538–2541 (1999)CrossRefGoogle Scholar
  7. 7.
    Cheng-I, C., Xiaoli, Z., et al.: Least squares subspace projection approach to mixed pixel classification for hyperspectral images. IEEE Transactions on Geoscience and Remote Sensing 36(3), 898–912 (1998)CrossRefGoogle Scholar
  8. 8.
    Tong, Q.X., Zhang, B., Zheng, L.F.: Hyperspectral remote sensing. Higher Education Press, Beijing (2006)Google Scholar
  9. 9.
    Kwon, H., Nasrabadi, N.M.: Kernel Orthogonal subspace projection for hyperspectral signal classifycation. IEEE Transactions on Geoscience and Remote Sensing 43(12), 2952–2962 (2005)CrossRefGoogle Scholar
  10. 10.
    Capobianco, L., Camps-Valls, G., Garzelli, A.: Target detection with a contextual kernel orthogo nal subspace projection. In: SPIE International Symposium Remote Sensing XIV, Cardiff, UK (September 2008)Google Scholar
  11. 11.
    Zhao, L.X., Zhang, K., Li, X.R.: Kernel Signature Space Orthogonal Projection for Target Detection in Hyperspectral Imagery. Journal of Remote Sensing (in press)Google Scholar
  12. 12.
    John, S.T., Nello, C.: Kernel Methods for Pattern Analysis. China machine press (2005)Google Scholar
  13. 13.
    Bian, Z.Q., Zhang, X.G.: Pattern Recognition. Tsinghua University Press, Beijing (2000)Google Scholar
  14. 14.
  15. 15.
    Wu, B., Zhang, L.P., Li, P.X.: Unmixing hyperspectral imagery based on support vector nonlinear approximating regression. Journal of Remote Sensing 10(3), 312–318 (2006)Google Scholar
  16. 16.
    Miao, L., Qi, H.: Endmember Extraction From Highly Mixed Data Using Minimum Volume Constrained Nonnegative Matrix Factorization. IEEE Transactions on Geoscience and Remote Sensing 45(3), 765–777 (2007)CrossRefGoogle Scholar
  17. 17.
    Sen, J.: Unsupervised hyperspectral unming theory and techniques. Ph.D Dissertation, Zhejiang University (2007)Google Scholar
  18. 18.
    Swayze, G.: The hydrothermal and structural history of the Cuprite Mining District, Southwestern Nevada: An integra-ted geological and geophysical approach, Ph.D. dissertation, University of Colorado, Boulder (1997), http://speclab.cr.usgs.gov and ftp://ftpext.cr.usgs.gov/pub/cr/co/denver/speclab/pub/cuprit/gregg.thesis.images/mineralzone.thesis.gif

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Liaoying Zhao
    • 1
  • Yinhe Shen
    • 1
  • Xiaorun Li
    • 2
  1. 1.Institute of Computer Application TechnologyHangZhou Dianzi UniversityHangzhouChina
  2. 2.College of Electrical EngineeringZhejiang UniversityHangzhouChina

Personalised recommendations