Advertisement

Integrating Spectral and Spatial features for Hyperspectral Image Classification with a Modified Composite Kernel Framework

  • Shashi Ranjan
  • Jignesh N. SarvaiyaEmail author
  • Jigisha N. Patel
Original Article
  • 34 Downloads

Abstract

This paper presents two modified composite kernel frameworks for the hyperspectral image classification. The first proposed framework is a simple extension of the previously proposed generalized composite kernel (GCK) approach, and in the second framework, we have used a voting strategy. The developed modified composite kernels combine different spatial and spectral profiles without considering any weight parameter. We have used the multinomial logistic regression (MLR) algorithm for the classification of hyperspectral images in this work. We have developed different spatial profiles: extended morphological profile (EMP) and extended multiattribute profile (EMAP) in this work, and further integrated them with the spectral profile for spectral-spatial classification. Experimental results with the AVIRIS Indian Pines, ROSIS Pavia university area and AVIRIS Salinas images indicate that the proposed framework leads to an improvement in classification accuracy. The obtained overall accuracy increases by 2–3% for AVIRIS Indian Pines scene, 1–2% for ROSIS Pavia University Area scene and 1–3% for AVIRIS Salinas scene with the first proposed framework. In case of the second proposed framework with a sufficient number of training samples per class, the overall accuracy increases by 3–7% for AVIRIS Indian Pines scene, 2.5% for ROSIS Pavia University Area scene and 7–8% for AVIRIS Salinas scene. For this framework (Modified-CKV with MLR) we have obtained the overall accuracy of 96.95% for the Indian Pines scene, 99.33% for the Pavia University Area, and 97.76% for the Salinas scene with 75 training samples per class, which is less than 2% of the available labelled samples. With the results obtained, we have also concluded that with the linear combination approach for the construction of composite kernels, the overall accuracy improves without any increase in computational complexity.

Keywords

Image classification Hyperspectral images Extended morphological profile Extended multiattribute profile Generalized composite kernel Multinomial logistic regression 

Zusammenfassung

Integration von spektralen und räumlichen Merkmalen für die Klassifizierung von hyperspektralen Bildern mit einem modifizierten Composite Kernel Framework. Dieser Artikel stellt zwei modifizierte Composite-Kernel-Frameworks für die hyperspektrale Bildklassifikation vor. Das erste vorgeschlagene Konzept ist eine einfache Erweiterung des bereits verfügbaren generalisierten Composite-Kernel (GCK) Ansatzes, während wir beim zweiten eine Voting Strategy verwendet haben. Die entwickelten modifizierten Composite-Kernels verknüpfen unterschiedliche räumliche und spektrale Profile ohne Berücksichtigung von Gewichtungen. Wir haben dafür den Multinomial Logistic Regression (MLR) Algorithmus für die Klassifizierung der hyperspektralen Bilddaten verwendet und unterschiedliche räumliche Profile entwickelt – das Erweiterte Morphologische Profil (EMP) und das Erweiterte Multiattributive Profil (EMAP) – und diese mit spektralen Profilen für die spektral-räumliche Klassifizierung integriert. Experimentelle Ergebnisse mit den Test-Szenen Indian Pines (AVIRIS), Universitätsgelände Pavia (ROSIS) und Salinas (AVIRIS) zeigen, dass das vorgeschlagene Framework zu einer Verbesserung der Klassifizierungsgenauigkeit führt. Der erzielte Gesamtgenauigkeit steigt um 2 – 3% für die Szene Indian Pines (AVIRIS), 1 – 2% für die Szene Universitätsgelände Pavia (ROSIS) und 1 – 3% für die Szene Salinas (AVIRIS). Im Falle des zweiten vorgeschlagenen Frameworks und einer ausreichenden Anzahl von Trainingsgebieten für jede Klasse erhöht sich die Gesamtgenauigkeit um 3–7% für die Szene Indian Pines (AVIRIS), 2 – 5% für die Szene Universitätsgelände Pavia (ROSIS) und 7 – 8% für die Szene Salinas (AVIRIS). Für dieses Framework (Modified-CKV mit MLR) haben wir die Gesamtgenauigkeit von 96,95% für die Szene Indian Pines, 99,33% für die Szene Universitätsgelände Pavia und 97,76% für die Salinas-Szene mit 75 Trainingsproben pro Klasse erhalten, was weniger als 2% der verfügbaren Samples entspricht. Die Ergebnisse zeigen, dass sich durch Linearkombination von Composite Kernels die Gesamtgenauigkeit erhöht, ohne dass dadurch die Komplexität der Berechnungen steigt.

References

  1. Bach F, Lanckriet G, Jordan M (2004) Multiple kernel learning, conic duality and the smo algorithm. In: Proceedings of the 21st international conference on machine learning, Banff, Alberta, Canada, pp 1–6Google Scholar
  2. Benediktsson JA, Palmason JA, Sveinsson JR (2005) Classification of hyperspectral data from urban areas based on extended morphological profiles. IEEE Trans Geosci Remote Sens 43:480–491CrossRefGoogle Scholar
  3. Bioucas JM, Figueiredo M (2009) Logistic regression via variable splitting and augmented lagrangian tools. Tech. rep., Instituto Superior Tecnico, Lisbon, PortugalGoogle Scholar
  4. Bioucas-Dias JM, Nascimento J (2008) Hyperspectral subspace identification. IEEE Trans Geosci Remote Sens 46:2435–2445CrossRefGoogle Scholar
  5. Bohning D (1992) Multinomial logistic regression algorithm. Ann Inst Stat Math 44:197–200CrossRefGoogle Scholar
  6. Borges JS, Bioucas-Dias JM, Marcal ARS (2011) Bayesian hyperspectral image segmentation with discriminative class learning. IEEE Trans Geosci Remote Sens 49:2151–2164CrossRefGoogle Scholar
  7. Boser BE, Guyon IM, Vapnik VN (1992) A training algorithm for optimal margin classifiers. In: Proceedings of the fifth annual workshop on computational Learning theory (COLT ‘92), New York, USAGoogle Scholar
  8. Bruzzone L, Chi M, Marconcini M (2006) A novel transductive svm for the semisupervised classification of remote sensing images. IEEE Trans Geosci Remote Sens 44:3363–3373CrossRefGoogle Scholar
  9. Camps-Valls G, Bruzzone L (2005) Kernel-based methods for hyperspectral image classification. IEEE Trans Geosci Remote Sens 43:1351–1362CrossRefGoogle Scholar
  10. Camps-Valls G, Gomez-Chova L, Munoz-Mari J, Vila-Frances J, Calpe-Maravila J (2006) Composite kernels for hyperspectral image classification. IEEE Geosci Remote Sens Lett 3:93–97CrossRefGoogle Scholar
  11. Foundation PR (1994) Hyperspectral images. Online, https://engineering.purdue.edu/~biehl/MultiSpec/hyperspectral.html. Accessed 18 Nov 2017
  12. Giles Foody M (2002) Status of land cover classification accuracy assessment. Remote Sens Environ 80:185–201CrossRefGoogle Scholar
  13. Group CI (2011) Hyperspectral remote sensing scenes. http://www.ehu.eus/ccwintco/index.php?title=Hyperspectral_Remote_Sensing_Scenes. Accessed 21 Nov 2017
  14. Gu Y, Wang C, You D, Zhang Y, Wang S, Zhang Y (2012) Representative multiple kernel learning for classification in hyperspectral imagery. IEEE Trans Geosci Remote Sens 50:2852–2865CrossRefGoogle Scholar
  15. He L, Li J, Liu C, Li S (2018) Recent advances on spectral-spatial hyperspectral image classification: an overview and new guidelines. IEEE Trans Geosci Remote Sens 56:1579–1597CrossRefGoogle Scholar
  16. Hughes GF (1968) On the mean accuracy of statistical pattern recognizers. IEEE Trans Inf Theory 14:55–63CrossRefGoogle Scholar
  17. Krishnapuram B, Carin L, Figueiredo M, Hartemink A (2005) Sparse multinomial logistic regression: fast algorithms and generalization bounds. IEEE Trans Pattern Anal Mach Intell 27:957–968CrossRefGoogle Scholar
  18. Lanckriet G, Cristianini N, Bartlett P, Ghaoui L, Jordan M (2004) Learning the kernel matrix with semidefinite programming. J Mach Learn Res 5:27–72Google Scholar
  19. Li J, Bioucas-Dias JM, Plaza A (2010) Semi-supervised hyperspectral image segmentation using multinomial logistic regression with active learning. IEEE Trans Geosci Remote Sens 48:4085–4098Google Scholar
  20. Li J, Bioucas-Dias JM, Plaza A (2011) Hyperspectral image segmentation using a new bayesian approach with active learning. IEEE Trans Geosci Remote Sens 49:3947–3960CrossRefGoogle Scholar
  21. Li J, Bioucas-Dias JM, Plaza A (2012) Spectral-spatial hyperspectral image segmentation using subspace multinomial logistic regression and markov random fields. IEEE Trans Geosci Remote Sens 50:809–823CrossRefGoogle Scholar
  22. Li J, Marpu PR, Plaza A, Bioucas-Dias JM, Benediktsson JA (2013) Generalized composite kernel framework for hyperspectral image classification. IEEE Trans Geosci Remote Sens 51:4816–4829CrossRefGoogle Scholar
  23. Liu T, Jin X, Gu Y (2016) Sparse multiple kernel learning for hyperspectral image classification using spatial-spectral features. In: 2016 Sixth International conference on instrumentation measurement, computer, communication and control (IMCCC), pp 614–618,  https://doi.org/10.1109/IMCCC.2016.180
  24. Lu D, Weng Q (2007) A survey of image classification methods and techniques for improving classification performance. Int J Remote Sens 28:823–870CrossRefGoogle Scholar
  25. Mura MD, Benediktsson JA, Bruzzone L (2010a) Morphological attribute profiles for the analysis of very high resolution images. IEEE Trans Geosci Remote Sens 48:3747–3762CrossRefGoogle Scholar
  26. Mura MD, Benediktsson JA, Waske B, Bruzzone L (2010b) Extended profiles with morphological attribute filters for the analysis of hyperspectral data. Int J Remote Sens 31:5975–5991CrossRefGoogle Scholar
  27. Plaza A, Benediktsson JA, Boardman JW, Brazile J, Bruzzone L, Camps-Valls G, Chanussot J, Fauvel M, Gamba P, Gualtieri A, Marconcini M, Tilton JC, Trianni G (2009) Recent advances in techniques for hyperspectral image processing. Remote Sens Environ 113:S110–S122.  https://doi.org/10.1016/j.rse.2007.07.028 http://www.sciencedirect.com/science/article/pii/S0034425709000807, imaging Spectroscopy Special IssueCrossRefGoogle Scholar
  28. Rakotomamonjy A, Bach F, Canu S, Grandvalet Y (2008) Simplemkl. J Mach Learn Res 9:2491–2521Google Scholar
  29. Scholkopf B, Smola A (2002) Learning with kernels: support vector machines, regularization, optimization and beyond. MIT Press, CambridgeGoogle Scholar
  30. Tuia D, Matasci G, Camps-Valls G, Kanevski M (2010) Learning relevant image features with multiple kernel classification. IEEE Trans Geosci Remote Sens 48:3780–3791CrossRefGoogle Scholar

Copyright information

© Deutsche Gesellschaft für Photogrammetrie, Fernerkundung und Geoinformation (DGPF) e.V. 2019

Authors and Affiliations

  1. 1.Department of Electronics EngineeringSVNIT SuratSuratIndia

Personalised recommendations