Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon openaccess Non-linear unmixing of hyperspectral images using multiple-kernel self-organising maps

This is an open access article published by the IET under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Received 28/02/2018, Accepted 22/03/2019, Revised 27/11/2018, Published 01/08/2019

The spatial pixel resolution of common multispectral and hyperspectral sensors is generally not sufficient to avoid that multiple elementary materials contribute to the observed spectrum of a single pixel. To alleviate this limitation, spectral unmixing is a by-pass procedure which consists in decomposing the observed spectra associated with these mixed pixels into a set of component spectra, or endmembers, and a set of corresponding proportions, or abundances, that represent the proportion of each endmember in these pixels. In this study, a spectral unmixing technique is proposed to handle the challenging scenario of non-linear mixtures. This algorithm relies on a dedicated implementation of multiple-kernel learning using self-organising map proposed as a solver for the non-linear unmixing problem. Based on a priori knowledge of the endmember spectra, it aims at estimating their relative abundances without specifying the non-linear model under consideration. It is compared to state-of-the-art algorithms using synthetic yet realistic and real hyperspectral images. Results obtained from experiments conducted on synthetic and real hyperspectral images assess the potential and the effectiveness of this unmixing strategy. Finally, the relevance and potential parallel implementation of the proposed method is demonstrated.

Inspec keywords: geophysical image processing; self-organising feature maps; image processing; hyperspectral imaging; spectral analysis; remote sensing; geophysical signal processing; learning (artificial intelligence)

Other keywords: relative abundances; nonlinear unmixing problem; endmember spectra; multiple-kernel learning; hyperspectral images; by-pass procedure; unmixing strategy; mixed pixels; nonlinear model; spatial pixel resolution; component spectra; spectral unmixing technique; corresponding proportions; multiple elementary materials; nonlinear mixtures; self-organising map; hyperspectral sensors; common multispectral sensors; single pixel; multiple-kernel self-organising maps; observed spectrum

Subjects: Knowledge engineering techniques; Instrumentation and techniques for geophysical, hydrospheric and lower atmosphere research; Optical, image and video signal processing; Geophysics computing; Other topics in statistics; Geophysical techniques and equipment; Computer vision and image processing techniques

References

    1. 1)
      • 4. Meganem, I., Déliot, P., Briottet, X., et al: ‘Linear-quadratic mixing model for reflectances in urban environments’, IEEE Trans. Geosci. Remote Sens., 2014, 52, pp. 544–558.
    2. 2)
      • 6. Heylen, R., Parente, M., Gader, P.: ‘A review of nonlinear hyperspectral unmixing methods’, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 2014, 7, pp. 1844–1868.
    3. 3)
      • 8. Dobigeon, N., Altmann, Y., Brun, N., et al: ‘Linear and nonlinear unmixing in hyperspectral imaging’, in Ruckebusch, C. (Ed.): ‘Resolving spectral mixtures – with application from ultrafast time-resolved spectroscopy to superresolution imaging’, Data Handling in Science and Technology, vol. 30, (Elsevier, Oxford, UK, 2016), pp. 185–224.
    4. 4)
      • 39. Vesanto, J., Alhoniemi, E.: ‘Clustering of the self-organizing map’, IEEE Trans. Neural Netw., 2000, 11, (3), pp. 586–600.
    5. 5)
      • 3. Dobigeon, N., Tits, L., Somers, B., et al: ‘A comparison of nonlinear mixing models for vegetated areas using simulated and real hyperspectral data’, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 2014, 7, pp. 1869–1878.
    6. 6)
      • 27. Kohonen, T., Somervuo, P.: ‘Self-organizing maps of symbol strings’, Neurocomputing, 1998, 21, (1), pp. 19–30.
    7. 7)
      • 31. Villa-Vialaneix, N., Rossi, F.: ‘A comparison between dissimilarity SOM and kernel SOM for clustering the vertices of a graph’. Proc. Workshop on Self-Organizing Maps (WSOM), Bielefeld, Germany, 2007, p. WeP–1.
    8. 8)
      • 2. Bioucas-Dias, J.M., Plaza, A., Dobigeon, N., et al: ‘Hyperspectral unmixing overview: geometrical, statistical, and sparse regression-based approaches’, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 2012, 5, pp. 354–379.
    9. 9)
      • 15. Altmann, Y., Dobigeon, N., Tourneret, J.-Y.: ‘Bilinear models for nonlinear unmixing of hyperspectral images’. Proc. IEEE GRSS Workshop Hyperspectral Image Signal Processing: Evolution in Remote Sensing (WHISPERS), Lisbon, Portugal, June 2011, pp. 1–4.
    10. 10)
      • 29. Olteanu, M., Villa-Vialaneix, N., Cierco-Ayrolles, C.: ‘Multiple kernel selforganizing maps’. European Symp. Artificial Neural Networks, Computational Intelligence Machine Learning, Bruges, Belgium, 2013, p. 83.
    11. 11)
      • 40. Foody, G.M.: ‘Relating the land-cover composition of mixed pixels to artificial neural network classification output’, Photogramm. Eng. Remote Sens., 1996, 62, (5), pp. 491–498.
    12. 12)
      • 19. Halimi, A., Bioucas-Dias, J., Dobigeon, N., et al: ‘Fast hyperspectral unmixing in presence of nonlinearity or mismodelling effects’, IEEE Trans. Comput. Imaging, 2017, 3, pp. 146–159.
    13. 13)
      • 16. Altmann, Y., Halimi, A., Dobigeon, N., et al: ‘Supervised nonlinear spectral unmixing using a postnonlinear mixing model for hyperspectral imagery’, IEEE Trans. Image Process., 2012, 21, (6), pp. 3017–3025.
    14. 14)
      • 46. ‘Hyperspectral remote sensing scenes’. Available at http://www.ehu.eus/ccwintco/index.php?title=Hyperspectral Remote Sensing Scenes, accessed 1 February 2016.
    15. 15)
      • 9. Winter, M.E.: ‘N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data’. Proc. SPIE Imaging Spectrometry V, Denver, USA, 1999, vol. 3753, pp. 266–275.
    16. 16)
      • 47. Plaza, J., Plaza, A., Martínez, P., et al: ‘H-COMP: A tool for quantitative and comparative analysis of endmember identification algorithms’. Proc. IEEE Int. Conf. Geoscience Remote Sensing (IGARSS), Toulouse, France, 2003, vol. 1, pp. 291–293.
    17. 17)
      • 44. Martín, G., Plaza, A.: ‘Region-based spatial preprocessing for endmember extraction and spectral unmixing’, IEEE Geosci. Remote Sens. Lett., 2011, 8, (4), pp. 745–749.
    18. 18)
      • 11. Boardman, J.W.: ‘Automating spectral unmixing of AVIRIS data using convex geometry concepts’. Summaries 4th Annual JPL Airborne Geoscience Workshop, Washington, D.C., 1993, vol. 1, pp. 11–14.
    19. 19)
      • 45. Fan, W., Hu, B., Miller, J., et al: ‘Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated-forest hyperspectral data’, Int. J. Remote Sens., 2009, 30, (11), pp. 2951–2962.
    20. 20)
      • 34. Altmann, Y., Dobigeon, N., McLaughlin, S., et al: ‘Nonlinear unmixing of hyperspectral images using radial basis functions and orthogonal least squares’. Proc. IEEE Int. Conf. Geoscience Remote Sensing (IGARSS), Vancouver, Canada, July 2011, pp. 1151–1154.
    21. 21)
      • 18. Févotte, C., Dobigeon, N.: ‘Nonlinear hyperspectral unmixing with robust nonnegative matrix factorization’, IEEE Trans. Image Process., 2015, 24, pp. 4810–4819.
    22. 22)
      • 23. Sonnenburg, S., Rätsch, G., Schäfer, C., et al: ‘Large scale multiple kernel learning’, J. Mach. Learn. Res., 2006, 7, pp. 1531–1565.
    23. 23)
      • 22. Gönen, M., Alpaydın, E.: ‘Multiple kernel learning algorithms’, J. Mach. Learn. Res., 2011, 12, pp. 2211–2268.
    24. 24)
      • 1. Bioucas-Dias, J.M., Plaza, A., Camps-Valls, G., et al: ‘Hyperspectral remote sensing data analysis and future challenges’, IEEE Geosci. Remote Sens. Mag., 2013, 1, (2), pp. 6–36.
    25. 25)
      • 37. Wang, J., Chang, C.-I.: ‘Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis’, IEEE Trans. Geosci. Remote Sens., 2006, 44, (6), pp. 1586–1600.
    26. 26)
      • 14. Bioucas-Dias, J.M., Figueiredo, M.A.: ‘Alternating direction algorithms for constrained sparse regression: application to hyperspectral unmixing’. Proc. IEEE GRSS Workshop Hyperspectral Image Signal Processing: Evolution in Remote Sensing (WHISPERS), Reykjavik, Iceland, 2010, pp. 1–4.
    27. 27)
      • 28. Martínez, P., Gualtieri, J., Aguilar, P., et al: ‘Hyperspectral image classification using a self-organizing map’. Summaries of the X JPL Airborne Earth Science Workshop, Pasadena, USA, 2001.
    28. 28)
      • 32. Blayo, F.: ‘Kohonen self-organizing maps: is the normalization necessary?’, Complex Syst., 1992, 6, (6), pp. 105–123.
    29. 29)
      • 30. MacDonald, D., Fyfe, C.: ‘The kernel self-organising map’. Proc. 4th Int. Conf. on Knowledge-Based Intelligent Engineering Systems and Allied Technologies, Salt Lake City, USA, 2000, vol. 1, pp. 317–320.
    30. 30)
      • 17. Altmann, Y., Dobigeon, N., Tourneret, J.-Y.: ‘Unsupervised post-nonlinear unmixing of hyperspectral images using a Hamiltonian Monte Carlo algorithm’, IEEE Trans. Image Process., 2014, 23, pp. 2663–2675.
    31. 31)
      • 43. ‘HyperMIX: ground truth for fractal images’. Available at http://hypercomp.es/hypermix/node/585, accessed 17 June 2016.
    32. 32)
      • 33. Guilfoyle, K.J., Althouse, M.L., Chang, C.-I.: ‘A quantitative and comparative analysis of linear and nonlinear spectral mixture models using radial basis function neural networks’, IEEE Trans. Geosci. Remote Sens., 2001, 39, pp. 2314–2318.
    33. 33)
      • 41. Chen, J., Richard, C., Honeine, P.: ‘Nonlinear unmixing of hyperspectral data based on a linear-mixture/nonlinear-fluctuation model’, IEEE Trans. Signal Process., 2013, 61, (2), pp. 480–492.
    34. 34)
      • 36. Wang, J., Chang, C.-I.: ‘Applications of independent component analysis in endmember extraction and abundance quantification for hyperspectral imagery’, IEEE Trans. Geosci. Remote Sens., 2006, 44, (9), pp. 2601–2616.
    35. 35)
      • 42. ‘USGS digital spectral library’. Available at http://speclab.cr.usgs.gov/spectral-lib.html, accessed 17 June 2016.
    36. 36)
      • 21. Ayerdi, B., Graña, M.: ‘Hyperspectral image nonlinear unmixing and reconstruction by elm regression ensemble’, Neurocomputing, 2016, 174, pp. 299–309.
    37. 37)
      • 10. Chang, C.-I., Wu, C.-C., min Liu, W., et al: ‘A new growing method for simplex-based endmember extraction algorithm’, IEEE Trans. Geosci. Remote Sens., 2006, 44, (10), pp. 2804–2819.
    38. 38)
      • 13. Heinz, D.C., Chang, C.-I.: ‘Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery’, IEEE Trans. Geosci. Remote Sens., 2001, 39, (3), pp. 529–545.
    39. 39)
      • 38. Plaza, A., Chang, C.-I.: ‘Impact of initialization on design of endmember extraction algorithms’, IEEE Trans. Geosci. Remote Sens., 2006, 44, (11), pp. 3397–3407.
    40. 40)
      • 7. Dobigeon, N., Tourneret, J.-Y., Richard, C., et al: ‘Nonlinear unmixing of hyperspectral images: models and algorithms’, IEEE Signal Process. Mag., 2014, 31, pp. 82–94.
    41. 41)
      • 26. Gu, Y., Wang, S., Jia, X.: ‘Spectral unmixing in multiple-kernel Hilbert space for hyperspectral imagery’, IEEE Trans. Geosci. Remote Sens., 2013, 51, (7), pp. 3968–3981.
    42. 42)
      • 25. Chen, J., Richard, C., Honeine, P.: ‘Nonlinear unmixing of hyperspectral images based on multi-kernel learning’. Proc. IEEE GRSS Workshop Hyperspectral Image Signal Processing: Evolution in Remote Sensing (WHISPERS), Shanghai, China, 2012.
    43. 43)
      • 5. Hapke, B.: ‘Theory of reflectance and emittance spectroscopy’ (Cambridge Univ. Press, Cambridge, UK, 1993).
    44. 44)
      • 12. Nascimento, J.M., Dias, J.M.B.: ‘Vertex component analysis: a fast algorithm to unmix hyperspectral data’, IEEE Trans. Geosci. Remote Sens., 2005, 43, (4), pp. 898–910.
    45. 45)
      • 24. Liu, K.-H., Lin, Y.-Y., Chen, C.-S.: ‘Linear spectral mixture analysis via multiple-kernel learning for hyperspectral image classification’, IEEE Trans. Geosci. Remote Sens., 2015, 53, (4), pp. 2254–2269.
    46. 46)
      • 20. Licciardi, G.A., Frate, F.D.: ‘Pixel unmixing in hyperspectral data by means of neural networks’, IEEE Trans. Geosci. Remote Sens., 2011, 49, (11), pp. 4163–4172.
    47. 47)
      • 35. Chang, C.-I., Du, Q.: ‘Estimation of number of spectrally distinct signal sources in hyperspectral imagery’, IEEE Trans. Geosci. Remote Sens., 2004, 42, (3), pp. 608–619.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2018.5094
Loading

Related content

content/journals/10.1049/iet-ipr.2018.5094
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Print this page
This is a required field
Please enter a valid email address