Fully Geometric-Constrained Progressive Endmember Finding: Growing Simplex Volume Analysis

Chapter

Abstract

Growing Simplex Volume Analysis (GSVA) has recently been developed as an alternative theory to the Simplex Volume Analysis (SVA) theory discussed in Chap.  6 and shown to be a promising approach to finding endmembers. As a matter of fact, the Simplex Growing Algorithm (SGA) developed by Chang et al. (2006) for GSVA does the same as the N-finder algorithm (N-FINDR) developed by Winter (1999a, b) for SVA. The key difference between these two is how endmembers are found by the algorithms. For SVA the number of all endmembers must be known beforehand. SVA then finds all the endmembers together through simultaneous replacement of these endmembers. By contrast, GSVA does not need to know the number of endmembers a priori. Instead, it grows simplexes successively to find endmembers one at a time. Such a growing process is terminated by a specific stopping rule designed for a particular application. Accordingly, SVA can be considered as a sequential process compared to GSVA, which is a progressive process. Nevertheless, both theories are indeed closely related one way or the other. This chapter studies GSVA and develops various algorithms to explore their relationships to SVA from a progressive perspective.

Keywords

Calcite Expense Kaolinite Berman Cuprite 

References

  1. Berman, M., H. Kiiveri, R. Lagerstrom, A. Ernst, R. Dunne, and J.F. Huntington. 2004. ICE: a statistical approach to identifying endmembers in hyperspectral images. IEEE Transaction on Goescience and Remote Sensing 42(10): 2085–2095.Google Scholar
  2. Bowles, J.P., and D.B. Gilles. 2007. An optical real-tine adaptive spectral identification system. In Hyperspectral Data Exploitation, ed. C.-I Chang, Chap. 4, 77–106.Google Scholar
  3. Chang, C.-I 2003. Hyperspectral Imaging: Techniques for Spectral detection and Classification. New York: Kluwer Academic/Plenum Publishers.Google Scholar
  4. Chang, C.-I, ed. 2007a. Hyperspectral Data Exploitation: Theory and Applications. New York: Wiley.Google Scholar
  5. Chang, C.-I 2007b. Overview, Chap. 1. In Hyperspectral Data Exploitation: Theory and Applications, ed. C.-I Chang, 1–16. New York: Wiley.Google Scholar
  6. Chang, C.-I 2007c. Information-processed matched filters for hyperspectral target detection and classification, Chap. 3. In Hyperspectral Data Exploitation: Theory and Applications, ed. C.-I Chang, 47–74. New York: Wiley.Google Scholar
  7. Chang, C.-I 2013. Hyperspectral data processing: algorithm design and analysis. New Jersey: Wiley.Google Scholar
  8. Chang, C.-I, and Q. Du. 2004. Estimation of number of spectrally distinct signal sources in hyperspectral imagery. IEEE Transaction on Geoscience and Remote Sensing 42(3): 608–619.Google Scholar
  9. Chang, C.-I, C.C. Wu, W. Liu, and Y.C. Ouyang. 2006. A growing method for simplex-based endmember extraction algorithms. IEEE Transaction on Geoscience and Remote Sensing 44(10): 2804–2819.Google Scholar
  10. Harsanyi, J.C. 1993. Detection and Classification of Subpixel Spectral Signatures in Hyperspectral Image Sequences. Doctoral dissertation. Department of Electrical Engineering, University of Maryland, Baltimore County. Baltimore, MD. Google Scholar
  11. Harsanyi, J.C., and C.-I Chang. 1994. Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach. IEEE Transaction on Geoscience and Remote Sensing 32(4): 779–785.Google Scholar
  12. Neville, R.A., K. Staenz, T. Szeredi, J. Lefebvre, and P. Hauff. 1999. Automatic endmember extraction from hyperspectral data for mineral exploration. In Proceedings of 4th International Airborne Remote Sensing Conference and Exhibition/21st Canadian Symposium on remote Sensing, 21–24, Ottawa, Ontario, Canada, June 1999.Google Scholar
  13. Plaza, A., and C.-I Chang, eds. 2007a. High Performance Computing in Remote Sensing. Boca Raton: CRC Press.Google Scholar
  14. Plaza, A., and C.-I Chang. 2007b. Specific issues about high-performance computing in remote sensing, non-literal analysis versus image-based processing, Chap. 1. In High-Performance Computing in Remote Sensing, ed. A. Plaza, and C.-I Chang. Boca Raton: CRC Press.Google Scholar
  15. Plaza, A., and C.-I Chang. 2007c. Clusters versus FPGAs for real-time processing of hyperspectral imagery. International Journal of High Performance Computing Applications 22(4): 366–385.Google Scholar
  16. Tou, J.T., and R.C. Gonzalez. 1974. Pattern Recognition Principles, 92–94. Reading, MA: Addison-Wesley.Google Scholar
  17. Winter, M.E. 1999a. Fast autonomous spectral endmember determination in hyperspectral data. In Proceedings of 13th International Conference on Applied Geologic Remote Sensing, vol. II, 337–344. B.C., Canada: Vancouver.Google Scholar
  18. Winter, M.E. 1999b. N-finder: an algorithm for fast autonomous spectral endmember determination in hyperspectral data. In Image Spectrometry V, Proceedings of SPIE 3753, 266–277.Google Scholar
  19. Xiong, W., C.-C. Wu, C.-I Chang, K. Kapalkis, and H.M. Chen. 2011. Fast algorithms to implement N-FINDR for hyperspectral endmember extraction. IEEE Journal of Selected Topic in Applied Earth Observation and Remote Sensing 4(3): 545–564.Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2016

Authors and Affiliations

  1. 1.BaltimoreUSA

Personalised recommendations