Advertisement

Computing Efficiently Spectral-Spatial Classification of Hyperspectral Images on Commodity GPUs

  • Pablo Quesada-Barriuso
  • Francisco Argüello
  • Dora B. Heras
Chapter
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 234)

Abstract

The high computational cost of the techniques for segmentation and classification of hyperspectral images makes them good candidates for parallel processing, in particular, for computing on Graphics Processing Units (GPUs). In this paper an efficient projection on the GPUs for the spectral–spatial classification of hyperspectral images using the Compute Unified Device Architecture (CUDA) for NVIDIA devices is presented. A watershed transform is applied after reducing the hyperspectral image to one band through the calculation of a morphological gradient, while the spectral classification is carried out by Support Vector Machine (SVMs). The results are combined with an adaptive majority vote. The different computational stages are concatenated in a pipeline that minimizes the data transfer between the main memory of the host computer and the global memory of the graphics device to maximize the computational throughput. The memory hierarchy and the thousands of threads available in this architecture are efficiently exploited. It is possible to study different data partitioning strategies and thread block arrangements in order to promote concurrent execution of a large number of threads. The objective is to efficiently exploit commodity hardware with the aim of achieving real-time execution for on-board processing.

Keywords

Hyperspectral images Watershed Classification  CUDA 

Notes

Acknowledgments

This work was supported in part by the Ministry of Science and Innovation, Government of Spain, cofounded by the FEDER funds of European Union, under contract TIN 2010-17541, and by Xunta de Galicia, Program for Consolidation of Competitive Research Groups ref. 2010/28. Pablo acknowledges financial support from the Ministry of Science and Innovation, Government of Spain, under a MICINN-FPI grant.

References

  1. 1.
    Plaza, A., Benediktsson, J.A., Boardman, J.W., Brazile, J., Bruzzone, L., Camps-Valls, G., Chanussot, J., Fauvel, M., Gamba, P., Gualtieri, A., Marconcini, M., Tilton, J.C., Trianni, G.: Recent advances in techniques for hyperspectral image processing. Remote Sens. Environ. 113, S110–S122 (2009)CrossRefGoogle Scholar
  2. 2.
    van der Meer, F.D., van der Werff, H.M., van Ruitenbeek, F.J., Hecker, C.A., Bakker, W.H., Noomen, M.F., van der Meijde, M., Carranza, E.J.M., de Smeth, J.B., Woldai, T.: Multi- and hyperspectral geologic remote sensing: A review. Int. J. Appl. Earth Observ. Geoinform. 14(11), 112–128 (2012)CrossRefGoogle Scholar
  3. 3.
    Gualtieri, J.A., Cromp, R.F.: Support vector machines for hyperspectral remote sensing classification. Proc. SPIE 3584, 221–232 (1998)CrossRefGoogle Scholar
  4. 4.
    Jia, X., Richards, J.A., Ricken, D.E.: Remote Sensing Digital Image Analysis: An Introduction. Springer Verlag, Berlin (1999)Google Scholar
  5. 5.
    Varshney, P. K., Arora, M. K. (eds.).: Advanced Image Processing Techniques for Remotely Sensed Hyperspectral Data. Springer Verlag, Berlin (2004)Google Scholar
  6. 6.
    Fauvel, M., Chanussot, J., Benediktsson, J.A., Sveinsson, J.R.: Spectral and spatial classification of hyperspectral data using SVMs and morphological profiles. IEEE Trans. Geosci. Remote Sens. 46(10), 3804–3814 (2008)CrossRefGoogle Scholar
  7. 7.
    Tarabalka, Y., Benediktsson, J.A. and Chanussot, J.: Spectral-spatial classification of hyperspectral imagery based on partitional clustering techniques. Geosci. Remote Sens. IEEE Trans. 47(8) , 2973–2987 (2009)Google Scholar
  8. 8.
    Tarabalka, Y., Chanussot, J., Benediktsson, J.A.: Segmentation and classification of hyperspectral images using watershed transformation. Pattern Recogn. 43(7), 2367–2379 (2010)CrossRefMATHGoogle Scholar
  9. 9.
    Fauvel, M.: Spectral and spatial methods for the classification of urban remote sensing data, Ph.D. Dissertation, Grenoble Institute of Technology, Grenoble (2007)Google Scholar
  10. 10.
    Farag, A.A., Mohamed, R.M., El-Baz, A.: A unified framework for MAP estimation in remote sensing image segmentation. Geosci. Remote Sens. IEEE Trans. 43(7), 1617–1634 (2005)CrossRefGoogle Scholar
  11. 11.
    Couprie, C., Grady, L., Najman, L., Talbot, H.: Power watershed: a unifying graph-based optimization framework. Pattern Anal. Mach. Intell. IEEE Trans. 33(7), 1384–1399 (2011)CrossRefGoogle Scholar
  12. 12.
    Tarabalka, Y., Benediktsson, J.A. and Chanussot, J.: Classification of hyperspectral data using support vector machines and adaptive neighborhoods. In: Proceedings of 6th EARSeL SIG IS, Workshop (2009)Google Scholar
  13. 13.
    Bernard, K., Tarabalka, Y., Angulo, J., Chanussot, J., Benediktsson, J.A.: Spectral spatial classification of hyperspectral data based on a stochastic minimum spanning forest approach. Image Process. IEEE Trans. 21(4), 2008–2021 (2012)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Vincent, L., Soille, P.: Watersheds in digital spaces: an eficient algorithm based on immersion simulations. IEEE Trans. Pattern Anal. Mach. Intell. 13, 583–598 (1991)CrossRefGoogle Scholar
  15. 15.
    Plaza, A., Plaza, J., Vegas, H.: Improving the performance of hyperspectral image and signal processing algorithms using parallel, distributed speciallized hardware-based systems. J. Signal Proces. Syst. 61(3), 293–315 (2010)CrossRefGoogle Scholar
  16. 16.
    González, C., Sánchez, S., Paz, A., Resano, J., Mozos, D., Plaza, A.: Use of FPGA or GPU-based architectures for remotely sensed hyperspectral image processing. Integr. VLSI J. 46(2), 89–103 (2013)CrossRefGoogle Scholar
  17. 17.
    Tarabalka, Y., Haavardsholm, T.V., Kåsen, I., Skauli, T.: Real-time anomaly detection in hyperspectral images using multivariate normal mixture models and GPU processing. J. Real Time Image Process. 4(3), 287–300 (2009)CrossRefGoogle Scholar
  18. 18.
    Heras, D. B., Argüello, F., Gómez, J. L., Becerra, J. A., Duro, R. J.: Towards real-time hyperspectral image processing, a GP-GPU implementation of target identification. In: 2011 IEEE 6th Internatioonal Conference on Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), vol. 1 pp. 316–321 (2011)Google Scholar
  19. 19.
    Priego, B., Souto, D., Bellas, F., Duro, R.J.: Unsupervised segmentation of hyperspectral images through evolved cellular automata. Adv. Knowl. Based Intell. Inform. Eng. Syst. 243, 2160–2169 (2012)Google Scholar
  20. 20.
    Quesada-Barriuso, P., Argüello, F., Heras, D.B.: Efficient segmentation of hyperspectral images on commodity GPUs. Adv. Knowl. Based Intell. Inform. Eng. Syst. 243, 2130–2139 (2012)Google Scholar
  21. 21.
    Christophe, E., Michel, J., Inglada, J.: Remote sensing processing: from multicore to GPU. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 4(3), 643–652 (2011)CrossRefGoogle Scholar
  22. 22.
    Bernabé, S., Plaza, A., Reddy Marpu, P., Benediktsson, J.A.: A new parallel tool for classification of remotely sensed imagery. Comput. Geosci. 46, 208–218 (2012)CrossRefGoogle Scholar
  23. 23.
    NVIDIA Corporation: NVIDIA CUDA C Programming Guide 4.2, Santa Clara (2011)Google Scholar
  24. 24.
    NVIDIA Corporation: NVIDIA’s Next Generation CUDA Compute Architecture: Kepler GK110 Whitepaper (2012)Google Scholar
  25. 25.
    NVIDIA Corporation: CUDA C Best Practices Guide (2012)Google Scholar
  26. 26.
    Gege, P., Beran, D., Mooshuber, W., Schulz, J. and Van Der Piepen, H.: System analysis and performance of the new version of the imaging spectrometer rosis, in Proceedings of the First EARSeL Workshop on Imaging Spectroscopy. University of Zurich Remote Sensing Laboratories, pp. 29–35 (1998)Google Scholar
  27. 27.
    Green, R.O., Eastwood, M.L., Sarture, C.M., Chrien, T.G., Aronsson, M., Chippendale, B.J., Faust, J.A., Pavri, B.E., Chovit, C.J., Solis, M.S., Olah, M.R., Williams, O.: Imaging spectroscopy and the airborne visible/infrared imaging spectrometer (AVIRIS). Remote Sens. Environ. 65(3), 227–248 (1998)Google Scholar
  28. 28.
    Camps-Valls, G., Bruzzone, L.: Kernel-based methods for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 43(6), 1351–1362 (2005)Google Scholar
  29. 29.
    Evans, A. and Liu, X.: A morphological gradient approach to color edge detection. Image Process IEEE Trans. 15(6), 1454–1463 (2006)Google Scholar
  30. 30.
    Roerdink, J.B.T.M., Meijster, A.: The watershed transform: definitions, algorithms and parallelization strategies. Fund. Inform. 41(1–2), 187–228 (2000)MathSciNetMATHGoogle Scholar
  31. 31.
    Meyer, F.: Topographic distance and watershed lines. Math. Morphol. Appl. Signal Process. 38(1), 113–125 (1994)MATHGoogle Scholar
  32. 32.
    Galilée, B., Mamalet, F., Renaudin, M., Coulon, P.Y.: Parallel asynchronous watershed algorithm-architecture. IEEE Trans. Paral. Distrib. Syst. 18(1), 44–56 (2007)Google Scholar
  33. 33.
    Nehaniv, C.L.: Evolution in asynchronous cellular automata. In: Proceedings of the 8th International Conference on Artificial life, MIT Press, pp. 65–73 (2003)Google Scholar
  34. 34.
    Quesada-Barriuso, P., Heras, D.B. and Argüello, F.: Efficient GPU asynchronous implementation of a watershed algorithm based on cellular automata. In: Proceedings of IEEE International Symposium on Parallel and Distributed Processing with Applications, pp. 79–86 (2012)Google Scholar
  35. 35.
    Boyer, R.S., Moore, J.S.: MJRTY - a fast majority vote algorithm, In: Boyer, R. S. (ed.) Automated Reasoning 1 Automated Reasoning Series, Springer, Netherlands, pp. 105–117 (1991)Google Scholar
  36. 36.
    Santos, A., Araújo, A., Menotti, D.: Combining multiple approaches for accuracy improvement in remote sensed hyperspectral images classification.In: Workshop of Thesis and Dissertations - XXV Conference on Graphics, Patterns and Images, pp. 54–59 (2012)Google Scholar
  37. 37.
    Balasalle, J., López, M.A., Rutherford, M.J.: Optimizing memory access patterns for cellular automata on GPUs. In: Hwu W. W. (ed.) GPU Computing Gems, Jade Edition. Morgan Kaufmann Publishers Inc., San Francisco, pp. 67–75 (2012)Google Scholar
  38. 38.
    Athanasopoulos, A., Dimou, A., Mezaris, V. and Kompatsiaris, I.: GPU acceleration for support vector machines, In: Proceedings of 12th International Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS 2011), Delft (2011)Google Scholar
  39. 39.
    Carpenter, A.: CUSVM: a CUDA implementation of support vector classification and regression, patternsonscreen (2009)Google Scholar
  40. 40.
    Catanzaro, B., Sundaram, N. and Keutzer, K.: Fast support vector machine training and classification on graphics processors. In: Proceedings of the 25th International Conference on Machine learning - ICML’08, pp. 104–111 (2008)Google Scholar
  41. 41.
    Herrero-López, S., Williams, J. R. and Sanchez, A.: Parallel multiclass classification using SVMs on GPUs. In: Proceedings of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units, ACM, pp. 2–11 (2010)Google Scholar
  42. 42.
    Li, Q., Salman, R., Test, E., Strack, R., Kecman, V.: GPUSVM: a comprehensive CUDA based support vector machine package. Central Eur. J. Comput. Sci. 1(4), 387–405 (2011)CrossRefMATHGoogle Scholar
  43. 43.
    Harris, M.: Optimizing parallel reduction in CUDA. NVIDIA Developer Technology, (2007)Google Scholar
  44. 44.
    Hawick, K.A., Leist, A., Playne, D.P.: Parallel graph component labelling with GPUs and CUDA. Parallel Comput. 36(12), 655–678 (2010)CrossRefMATHGoogle Scholar
  45. 45.
    Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. 2(3), 27 (2011)Google Scholar
  46. 46.
    Hsu, C.-W., Lin, C.-J.: A comparison of methods for multiclass support vector machines. IEEE Trans. Neural Netw. 13(2), 415–425 (2002)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Pablo Quesada-Barriuso
    • 1
  • Francisco Argüello
    • 1
  • Dora B. Heras
    • 1
  1. 1.Centro de Investigación en Tecnoloxías da Información (CITIUS)University of Santiago de CompostelaSantiago de CompostelaSpain

Personalised recommendations