Arabian Journal for Science and Engineering

, Volume 39, Issue 1, pp 181–189 | Cite as

Using Improved ICA Method for Hyperspectral Data Classification

Research Article - Earth Sciences
First Online:
Received:
Accepted:

Abstract

The hyperspectral remote sensing data have the characteristics of high spectral resolution and vast information and high dimensionality. The traditional supervised classification method and unsupervised one have such shortcomings as the classification accuracy is low, the running time will be long, and they cannot obtain a better classification result. To solve these problems, a hyperspectral remote sensing classification algorithm of improved independent component analysis (ICA) is proposed. The core of improved ICA is that the simplification and optimization of ICA model is completed by variational approximation algorithm and then the hyperspectral remote sensing data classification realized with the aid of nonlinear mapping function in support vector machine. The proposed method is evaluated and tested on simulation experiment and a real hyperspectral imager image experiment. The experiment results show that although the running time is longer for small sample data, the number of support vectors involved in calculation process is smaller, the space complexity is relatively simpler, the overall classification accuracy is higher and the classification image quality is much better.

Keywords

Remote sensing Hyperspectral data Classification Independent component analysis (ICA) Support vector machine (SVM) FastICA 
This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Wang, J.; Chang, C.I.: Independent component analysis-based dimensionality reduction with application in hyperspectral image analysis. IEEE Trans. Geosci. Remote Sens. 44(6), 1586–1600 (2006)CrossRefGoogle Scholar
  2. 2.
    Zeng, S.G.; Wang, X.M.; Fan, R.B.; Xia, D.S.: Remote sensing classification based on independent component analysis. J. Remote Sens. 8(2), 150–157 (2004)Google Scholar
  3. 3.
    Chen, S.P.; Tong, Q.X.; Guo, H.D.: Mechanism of Remote Sensing Information. China Science Press, Beijing (1998)Google Scholar
  4. 4.
    Zhao, C.H.; Qi, B.; Zhang, Y.: Hyperspectral image classification based on variational relevance vector machine. Acta Opt. Sin. 32(8), 256–261 (2012)Google Scholar
  5. 5.
    Wu, H.: Classification methodology combined with texture feature for hyperspectral remote sensing image. Comput. Eng. Design 33(5), 1993–1996 (2012)Google Scholar
  6. 6.
    Hyvärinen, A.; Oja, E.: Independent component analysis: algorithm and application. Neural Netw. 13(4–5), 411–430 (2000)CrossRefGoogle Scholar
  7. 7.
    Schnidt, F.; Schmidt, A.; Treguier, E.; Guiheneuf, M.; Moussaoui, S.; Dobigeon, N.: Implementation strategies for hyperspectral unmixing using Bayesian source separation. IEEE Trans. Geosci. Remote Sens. 48(11), 4003–4013 (2010)Google Scholar
  8. 8.
    Ripley, B.D.: Neural networks and related methods for classification. J. R. Stat. Soc. 56(3), 409–456 (1994)MATHMathSciNetGoogle Scholar
  9. 9.
    Ishaq, A.M.: Remote sensing techniques for monitoring of pollution in coastal wasters; potential application to Saudi Arabia. Arab. J. Sci. Eng. 10(1), 15–26 (1985)Google Scholar
  10. 10.
    Cortes, C.; Vapink, V.: Support-vector networks. Math. Learn. 20(3), 273–297 (1995)MATHGoogle Scholar
  11. 11.
    Butt, M.J.; Waqas, A.; Iqbal, M.F.; Muhammad, G.; Lodhi, MAK.: Assessment of urban sprawl of Islamabad metropolitan area using multi-sensor and multi-temperoal satellite data. Arab. J. Sci. Eng. 37(1), 101–114 (2012)CrossRefGoogle Scholar
  12. 12.
    Belouchrani, A.; Meraim, K.A.; Cardoso, J.F.; Moulines, E.: A blind source separation technique based on second order statistics. IEEE Trans. Signal Process. 45(2), 434–444 (1997)CrossRefGoogle Scholar
  13. 13.
    Bell, A.J.; Sejnowski, T.J.: An information-maximization approach to blind separation and blind deconvolution. Neural Comput. 7(6), 1129–1159 (1995)CrossRefGoogle Scholar
  14. 14.
    Fan, T.; Li, Z.N.; Xiao, Y.X.: Blind separation of mechanical source signals based on the estimation of source numbers. J. Mech. Strength 33(1), 15–19 (2011)MATHGoogle Scholar
  15. 15.
    Benaroya, L.; Bimbot, E.; Gribonval, R.: Audio source separation with a signal sensor. IEEE Trans. Audio Speech Lang. Process 14(1), 191–199 (2006)CrossRefGoogle Scholar
  16. 16.
    Hegazi, E.; Hamdan, A.; Mastromarino, J.: Remote fingerprinting of crude oil using time-resolved fluorescence spectra. Arab. J. Sci. Eng. 30(1B), 3–12 (2005)Google Scholar
  17. 17.
    Dubar, O.; Hart, P.E.; Stork, D.G.: Pattern Classification. Wiley, New York (2001)Google Scholar
  18. 18.
    Chang, C.L.: Hyperspectral Imaging: Techniques for Spectral Detection and Classification. Kluwer Academic/Plenum Publishers, New York (2003)CrossRefGoogle Scholar
  19. 19.
    Descour, M.R.; Shen, S.S.: Imaging Spectrometry V. SPIE Engineering Press, San Diego (1999)Google Scholar
  20. 20.
    Dai, Q.; Ma, J.W.; Li, Q.Q.; Chen, X.; Feng, C.: The study on remote sensing data classification using Bayesian network. J. Electro. Info. Technol. 27(11), 1782–1785 (2005)Google Scholar
  21. 21.
    Luo, W.F.; Zhong, L.; Zhang, B.; Gao, L.R.: Independent component analysis for spectral unmixing in hyperspectral remote sensing image. Spectrosc. Spect. Anal. 30(6), 1628–1633 (2010)Google Scholar
  22. 22.
    Li, Z.N.; Fan, T.; Liu, L.Z.; Lu, J.F.: Blind separation of the mechanical fault sources based on variational baysian theory. J. Vib. Shock 28(6), 12–16 (2009)Google Scholar
  23. 23.
    Fung, T.; Ledrew, E.: Application of principal components analysis change detection. Photogramm Eng. Remote Sens. 53(5), 1649–1658 (1987)Google Scholar
  24. 24.
    Chang, C.I.; Chiang, S.S.; Smith, J.A.; Ginsberg, I.W.: Linear spectral random mixture analysis for hyperspectral imagery. IEEE Trans. Geosci. Remote Sens. 40(2), 375–392 (2002)CrossRefGoogle Scholar
  25. 25.
    Dalponte, M.; Bruzzone, L.; Gianelle, D.: Fusion of hyperspectral and LIDAR remote sensing data for the classification of complex forest areas. IEEE Trans. Geosci. Remote Sens. 46(5), 1416–1427(2008)CrossRefGoogle Scholar
  26. 26.
    Zhao, J.W.; Ouyang, Q.; Chen, Q.S.; Wang, J.H.: Detection of bruise on pear by hyperspectral imaging sensor with different classification algorithms. Sensor Lett. 8(4), 570–576 (2010)CrossRefGoogle Scholar
  27. 27.
    Zhong, J.Q.; Wang, R.S.: Multitemporal remote sensing images change detection based on ICA. J. Electro. Inf. Technol. 28(6), 994–998 (2006)Google Scholar
  28. 28.
    Li, C.F.; Yin, J.Y.: A multispectral remote sensing data spectral unmixing algorithm based on variational Bayesian ICA. J. Indian Soc. Remote Sens. 41(2), 259–268 (2013)CrossRefGoogle Scholar
  29. 29.
    Chang, C.I.; Plaza, A.: Fast iterative algorithm for implementation of pixel purity index. IEEE Geosci. Remote Sens. Lett. 3(1), 63–67 (2006)CrossRefGoogle Scholar
  30. 30.
    Lee, T.W.; Girolami, M.; Sejnowski, T.J.: Independent component analysis using an extended infomax algorithm for mixed aubguassian and supergaussian sources. Neural Comput. 11(2), 417–441 (1999)CrossRefGoogle Scholar
  31. 31.
    Wang, X.M.; Zeng, S.G.; Xia, D.S.: Remote image classification based on loose modified fast independent component analysis algorithm. Comput. Eng. Appl. 41(7), 84–86 (2005)Google Scholar

Copyright information

© King Fahd University of Petroleum and Minerals 2013

Authors and Affiliations

  1. 1.School of Computer Engineering and ScienceShanghai UniversityShanghaiPeople’s Republic of China

Personalised recommendations