Abstract
The use of kernel-based methods has recently become a popular approach to many hyperspectral image applications such as anomaly detection (AD). A well-known kernel-based anomaly detector in hyperspectral images is the Kernel-Reed-Xiaoli (KRX) algorithm. This study aims at improving the performance of the original KRX. To meet this purpose, a modified version of the KRX that assumes a spherical covariance matrix for the background class is proposed. Performance comparison is accomplished using two real hyperspectral data sets that contain subpixel and multipixel targets. Experimental results demonstrate that the use of the modified KRX, compared to the original KRX, improves the detection and computational efficiency for AD.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bajorski P, Ientilucci EJ, Schott JR (2004) Comparison of basis vector selection methods for target and background subspaces as applied to subpixel target detection. Proc SPIE 5425:97–108. doi: 10.1117/12.835956
Rajasimman Balasubramanian UAB, Saravanavel J, Gunasekaran S (2013) Ore mineral discrimination using hyperspectral remote sensing—a field-based spectral analysis. Arab J Geosci 6(12):4709–4716
Banerjee A, Burlina P, Diehl C (2006) A support vector method for anomaly detection in hyperspectral imagery. IEEE Trans Geosci Remote 44(8):2282–2291. doi: 10.1109/TGRS.2006.873019
Basener WF, Messinger DW (2009) Enhanced detection and visualization of anomalies in spectral imagery. Shen, S.S. & Lewis, P.E. (eds.): Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XV, SPIE 7334. doi: 10.1117/12.818672
Camastra F (2002) Kernel methods for unsupervised learning, Ph. D. dissertation, University of Genova
Camps-Valls G, Bruzzone L (2009) Kernel methods for remote sensing data analysis. Wiley, New York
Carlotto MJ (2005) A cluster-based approach for detecting man-made objects and changes in imagery. IEEE Trans Geosci Remote 43(2):374–387. doi: 10.1109/TGRS.2004.841481
Chang C-I (2005) Orthogonal subspace projection revisited: a comprehensive study and analysis. IEEE Trans Geosci Remote 43(3):502–518. doi: 10.1109/TGRS.2004.839543
Etteieb S, Louhaichi M, Kalaitzidis C, Gitas IZ (2012) Mediterranean forest mapping using hyper-spectral satellite imagery. Arab J Geosci 6(12):5017–5032
Goldberg H, Kwon H, Nasrabadi NM (2007) Kernel eigenspace separation transform for subspace anomaly detection in hyperspectral imagery. IEEE Geosci Remote Sens Lett 4(4):581–585. doi: 10.1109/LGRS.2007.903083
Heiden U, Heldens W, Roessner S, Segl K, Esch T, Mueller A (2012) Urban structure type characterization using hyperspectral remote sensing and height information. Landsc Urban Plan 105(4):361–375
Hoffbeck J, Landgrebe D (1996) Covariance matrix estimation and classification with limited training data. IEEE Trans Pattern Anal 18(7):763–767. doi: 10.1109/34.506799
Kwon H, Nasrabadi NM (2010) Optimal kernel bandwidth estimation for hyperspectral kernel-based anomaly detection. IGARSS. 2812 – 2815. doi: 10.1109/IGARSS.2010.5649775
Kwon H, Nasrabadi NM (2007) A comparative analysis of kernel subspace target detectors for hyperspectral imagery. EURASIP J Adv Signal Process 2007(1):193–193. doi: 10.1155/2007/29250
Kwon H, Nasrabadi NM (2005) Kernel RX-algorithm: a nonlinear anomaly detector for hyperspectral imagery. IEEE Trans Geosci Remote 43(2):388–397. doi: 10.1109/TGRS.2004.841487
Manolakis D, Shaw G (2002) Detection algorithms for hyperspectral imaging application. IEEE Signal Process Mag 19:29–43
Matteoli S, Carnesecchi F, Diani M, Corsini G (2007) Comparative analysis of hyperspectral anomaly detection strategies on a new high spatial and spectral resolution data set. SPIE 6748. doi: 10.1117/12.738062
Matteoli S, Diani M, Corsini G (2010) A tutorial overview of anomaly detection in hyperspectral images. IEEE Aerosp Electron Syst Mag 25(7):5–28. doi: 10.1109/MAES.2010.5546306
Matteoli S, Diani M, Corsini G (2009) Different approaches for improved covariance matrix estimation in hyperspectral anomaly detection. Riunione Annuale GTTI: 1–8
Matteoli S, Veracini T, Diani M, Corsini G (2013) Models and methods for automated background density estimation in hyperspectral anomaly detection. IEEE Trans Geosci Remote 51(4):2837–2852. doi: 10.1109/TGRS.2012.2214392
Mei F, Zhao C, Huo H, Sun Y (2008) An adaptive kernel method for anomaly detection in hyperspectral imagery. in Proc. 2nd Int Symp Intell Inf Technol 874–878. doi: 10.1109/IITA.2008.106
Moguerza JM, Muñoz A (2006) Support vector machines with applications. Stat Sci 21(3):322–336
Moradizadeh M, Momeni M, Saradjian MR (2013) Estimation and validation of atmospheric water vapor content using a MODIS NIR band ratio technique based on AIRS water vapor products. Arab J Geosci 6(6):1799–1806
Nasrabadi NM (2008) Regularization for spectral matched filter and RX anomaly detector. Proc. SPIE6966, Algorithms and technologies for multispectral, hyperspectral, and ultraspectral imagery XIV, 696604. doi: 10.1117/12.773444
Playle N (2006) Detection of landmines using hyperspectral imaging. Proc. SPIE 6217, Detection and remediation technologies for mines and minelike targets XI, 62170A. doi: 10.1117/12.664044
Ranney KI, Soumekh M (2006) Hyperspectral anomaly detection within the signal subspace. IEEE Geosci Remote Sens Lett 3(3):312–316. doi: 10.1109/LGRS.2006.870833
Reed IS, Yu X (1990) Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution. IEEE Trans Acoust Speech 38(10):1760–1770. doi: 10.1109/29.60107
Saleh AM, Belal AB, Arafat SM (2013) Identification and mapping of some soil types using field spectrometry and spectral mixture analyses: a case study of North Sinai, Egypt. Arab J Geosci 6(6):1799–1806. doi: 10.1007/s12517-011-0501-6
Shawe-Taylor J, Cristianini N (2004) Kernel methods for pattern analysis. Cambridge Univ. Press, Cambridge, U.K
Schölkopf B, Smola A (2002) Learning with kernels. MIT Press, Cambridge, MA
Schweizer S, Moura JMF (2001) Efficient detection in hyperspectral imagery. IEEE Trans Image Process 10:584–597. doi: 10.1109/83.913593
Snyder D, Kerekes J, Fairweather I, Crabtree R, Shive J, and Hager S (2008) Development of a web-based application to evaluate target finding algorithms. Proceedings of the 2008 I.E. International Geoscience and Remote Sensing Symposium (IGARSS), 2: 915–918 (freely available data [online]: http://dirsapps.cis.rit.edu/blindtest/). doi: 10.1109/IGARSS.2008.4779144
Stein DWJ, Beaven SG, Ho LE, Winter EM, Schaum AP, Stocker AD (2010) Anomaly detection from hyperspectral imagery. IEEE Signal Process Mag 19(1):58–69. doi: 10.1109/79.974730
Tax DMJ (2001) One-class classification, Ph.D. dissertation, Delft University of Technology
Tiwari S, Sanjeev A, Anh T (2008) FastKRX: a fast approximation for kernel RX anomaly detection. Proc SPIE 6953:695316. doi: 10.1117/12.779586
Webb AR, Copsey KD (2011) Statistical pattern recognition, 3rd edn. Wiley, New York
Wilson TA (1998) Automatic target cueing of hyperspectral image data. Ph.D. dissertation, Air Force Institute of Technology
Acknowledgments
The authors would like to thank the Digital Imaging and Remote Sensing group Center for Imaging Science, Rochester Institute of Technology, Rochester, NY, for providing the Target Detection Test data sets, Dr. John P. Kerekes for his valuable help in providing truth locations of the blind test targets.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Khazai, S., Mojaradi, B. A modified kernel-RX algorithm for anomaly detection in hyperspectral images. Arab J Geosci 8, 1487–1495 (2015). https://doi.org/10.1007/s12517-013-1218-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12517-013-1218-5