Abstract
The ability to automatically classify hyperspectral imagery is of fundamental economic importance to the mining industry. A method of automated multi-class classification based on multi-task Gaussian processes (MTGPs) is proposed for classification of remotely sensed hyperspectral imagery. It is proved that because of the illumination invariance of the hyperspectral curves, the covariance function of the Gaussian process (GPs) has to be non-stationary. To enable multi-class classification of the hyperspectral imagery, a non-stationary multi-task observation angle-dependent covariance function is derived. In order to test MTGP, it was applied to data acquired in the laboratory and also in field. First, the MTGP was applied to hyperspectral imagery acquired under artificial light from samples of rock of known mineral composition. Data from a high-resolution field spectrometer are used to train the GPs. Second, the MTGP was applied to imagery of a vertical rock wall acquired under natural illumination. Spectra from hyperspectral imagery acquired in the laboratory are used to train the GPs. Results were compared with those obtained using the spectral angle mapper (SAM). In laboratory imagery, MTGP outperformed SAM across several metrics, including overall accuracy (MTGP: 0.96–0.98; SAM: 0.91–0.93) and the kappa coefficient of agreement (MTGP: 0.95–0.97; SAM: 0.88–0.91). MTGP applied to hyperspectral imagery of the rock wall gave broadly similar results to those from SAM; however, there were important differences. Some rock types were confused by SAM, but not by MTGP. Comparison of classified imagery with ground truth maps showed that MTGP outperformed SAM.
Similar content being viewed by others
References
Ben-Dor E, Patkin K, Banin A, Karnieli A (2002) Mapping of several soil properties using DAIS-7915 hyperspectral scanner data—a case study over clayey soils in Israel. Int J Remote Sens 23:1043–1062
Bierwirth P, Huston D, Blewett R (2002) Hyperspectral mapping of mineral assemblages associated with gold mineralization in the central Pilbara, western Australia. Econ Geol 97:819–826
Chabrillat S, Goetz AFH, Krosley L, Olsen HW (2002) Use of hyperspectral images in the identification and mapping of expansive clay soils and the role of spatial resolution. Remote Sens Environ 82:431–445
Clark RN, Roush TL (1984) Reflectance spectroscopy: quantitative analysis techniques for remote sensing applications. J Geophys Res 89:6329–6340
Clark RN, Swayze GA, Livo KE, Kokaly RF, Sutley SJ, Dalton JB, McDougal RR, Gent CA (2003) Imaging spectroscopy: Earth and planetary remote sensing with the USGS tetracorder and expert systems. J Geophys Res Planets 108:5131
Cloutis EA (1996) Hyperspectral geological remote sensing: evaluation of analytical techniques. Int J Remote Sens 17:2215–2242
Congalton RG, Oderwald RG, Mead RA (1983) Assessing Landsat classification accuracy using discrete multivariate-analysis statistical techniques. Photogramm Eng Remote Sensing 49:1671–1678
Cudahy TJ, Ramanaidou ER (1997) Measurement of the hematite:goethite ratio using field visible and near-infrared reflectance spectrometry in channel iron deposits, western Australia. Aust J Earth Sci 44:411–420
Daughtry CST, Hunt ER, McMurtrey JE (2004) Assessing crop residue cover using shortwave infrared reflectance. Remote Sens Environ 90:126–134
Elvidge CD (1988) Vegetation reflectance features in AVIRIS data. In: Sixth thematic conference on remote sensing for exploration geology. Environmental Research Institute of Michigan, Houston, TX, pp 169-182
Goetz AFH (2009) Three decades of hyperspectral remote sensing of the Earth: a personal view. Remote Sens Environ 113:S5–S16
Hecker C, van der Meijde M, van der Werff H, van der Meer FD (2008) Assessing the influence of reference spectra on synthetic SAM classification results. IEEE Trans Geosci Remote Sens 46:4162–4172
Hudson WD, Ramm CW (1987) Correct formulation of the kappa coefficient of agreement. Photogramm Eng Remote Sensing 53:421–422
Jiang L, Zhu B, Rao X, Berney G, Tao Y (2007) Discrimination of black walnut shell and pulp in hyperspectral fluorescence imagery using Gaussian kernel function approach. J Food Eng 81:108–117
Kohram M, Sap M (2008) Composite kernels for support vector classification of hyper-spectral data. MICAI 2008 Adv Artif Intell 5317:360-370
Kokaly RF, Despain DG, Clark RN, Livo KE (2003) Mapping vegetation in Yellowstone National Park using spectral feature analysis of AVIRIS data. Remote Sens Environ 84:437–456
Kruse FA, Lefkoff AB, Dietz JB (1993) Expert system-based mineral mapping in northern Death Valley, California/Nevada, using the airborne visible/infrared imaging spectrometer (AVIRIS). Remote Sens Environ 44:309–336
Mather PM (2004) Computer processing of remotely-sensed images, 3rd edn. Wiley, Chichester
Melkumyan A, Nettleton E (2009) An observation angle dependent non-stationary covariance function for Gaussian process regression. Lect Notes Comput Sci 5863:331–339
Melkumyan A, Ramos F (2011) Multi-kernel Gaussian processes. In: The international joint conference on artificial intelligence (IJCAI’11), Barcellona, Spain
Morris RC (1980) A textural and mineralogical study of the relationship of iron ore to banded iron formation in the Hamersley iron province of western Australia. Econ Geol 75:184–209
Morris RC, Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of western Australia: a critical review. Aust J Earth Sci 58:417–451
Murphy R, Schneider S, Monteiro S (2014) Mapping layers of clay in a vertical geological surface using hyperspectral imagery: variability in parameters of SWIR absorption features under different conditions of illumination. Remote Sens 6:9104–9129
Murphy RJ (1995) Mapping of jasperoid in the Cedar Mountains, Utah, USA, using imaging spectrometer data. Int J Remote Sens 16:1021–1041
Murphy RJ, Monteiro ST, Schneider S (2012) Evaluating classification techniques for mapping vertical geology using field-based hyperspectral sensors. IEEE Trans Geosci Remote Sens 50:3066–3080
Murphy RJ, Wadge G (1994) The effects of vegetation on the ability to map soils using imaging spectrometer data. Int J Remote Sens 15:63–86
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
Rasmussen C, Williams CK (2006) Gaussian processes for machine learning. MIT Press, Cambridge
Scheuerer M, Schaback R, Schlather M (2013) Interpolation of spatial data - a stochastic or a deterministic problem? Eur J Appl Math 24:601–629
Schneider S, Murphy RJ, Melkumyan A (2014) Evaluating the performance of a new classifier - the GP-OAD: a comparison with existing methods for classifying rock-type and mineralogy from hyperspectral imagery. Journal of Photogrammetry and Remote Sensing 98:145–456
Vane G, Goetz AFH (1988) Terrestrial imaging spectroscopy. Remote Sens Environ 24:1–29
Acknowledgments
This work has been supported by the Rio Tinto Centre for Mine Automation and the Australian Centre for Field RoboticsTable captions.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chlingaryan, A., Melkumyan, A., Murphy, R.J. et al. Automated Multi-class Classification of Remotely Sensed Hyperspectral Imagery Via Gaussian Processes with a Non-stationary Covariance Function. Math Geosci 48, 537–558 (2016). https://doi.org/10.1007/s11004-015-9622-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11004-015-9622-x