A background refinement method based on local density for hyperspectral anomaly detection

Abstract

For anomaly detection, anomalies existing in the background will affect the detection performance. Accordingly, a background refinement method based on the local density is proposed to remove the anomalies from the background. In this work, the local density is measured by its spectral neighbors through a certain radius which is obtained by calculating the mean median of the distance matrix. Further, a two-step segmentation strategy is designed. The first segmentation step divides the original background into two subsets, a large subset composed by background pixels and a small subset containing both background pixels and anomalies. The second segmentation step employing Otsu method with an aim to obtain a discrimination threshold is conducted on the small subset. Then the pixels whose local densities are lower than the threshold are removed. Finally, to validate the effectiveness of the proposed method, it combines Reed-Xiaoli detector and collaborative-representation-based detector to detect anomalies. Experiments are conducted on two real hyperspectral datasets. Results show that the proposed method achieves better detection performance.

Keywords

hyperspectral imagery / anomaly detection / background refinement / the local density

References

[1] ManolakisD, ShawG. Clutter detection algorithms for airborne pulse-Doppler radar [J]. IEEE Signal Processing Magazine, 2010, 19(1): 29-43

[2] BajorskiP. Target detection under misspecified models in hyperspectral images [J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2012, 5(2): 470-477

[3] ZhaoC-H, LiJ, MeiF. A kernel weighted RX algorithm for anomaly detection in hyperspectral imagery [J]. Journal of Infrared and Millim Waves, 2010, 29(5): 378-382

[4] SteinD W J, BeavenS G, HoffL E. Anomaly detection from hyperspectral imagery [J]. IEEE Signal Processing Magazine, 2002, 19(1): 58-69

[5] ZhaoC-h, WangY-l, LiX-hui. A real-time anomaly detection algorithm for hyperspectral imagery based on causal processing [J]. Journal of Infrared Millim Waves, 2015, 34(1): 114-121

[6] ChangC I, ChiangS S. Anomaly detection and classification for hyperspectral imagery [J]. IEEE Transactions on Geoscience & Remote Sensing, 2002, 40(6): 1314-1325

[7] RobeyF C, FuhrmannD R, KellyE J. A CFAR adaptive matched filter detector [J]. IEEE Transactions on Aerospace & Electronic Systems, 1992, 28(1): 208-216

[8] RenH, DuQ, ChangC I. Comparison between constrained energy minimization based approaches for hyperspectral imagery [C]. IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data (IEEE 2003), 2003, Greenbelt, USA, IEEE: 244248

[9] ChenY, NasrabadiN M, TranT D. Simultaneous joint sparsity model for target detection in hyperspectral imagery [J]. IEEE Geoscience & Remote Sensing Letters, 2011, 8(4): 676-680

[10] ReedI S, YuX. Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution [J]. IEEE Transactions on Acoustics, Speech and Signal Processing, 1990, 38(10): 1760-1770

[11] KwonH, NasrabadiN M. Kernel RX-Algorithm: A nonlinear anomaly detector for hyper-spectral imagery [J]. IEEE Transactions on Geoscience & Remote Sensing, 2005, 43(2): 388-397

[12] BanerjeeA, BurlinaP, DiehlC. A support vector method for anomaly detection in hyperspectral imagery [J]. IEEE Transactions on Geoscience & Remote Sensing, 2006, 44(8): 2282-2291

[13] KhazaiS, HomayouniS, SafariA. Anomaly detection in hyperspectral images based on an adaptive support vector method [J]. IEEE Geoscience & Remote Sensing Letters, 2011, 8(4): 646-650

[14] SaklaW, ChanA, JiJ. An SVDD-based algorithm for target detection in hyperspectral imagery [J]. IEEE Geoscience & Remote Sensing Letters, 2011, 8(2): 384-388

[15] LiW, DuQ. Collaborative representation for hyperspectral anomaly detection [J]. IEEE Transactions on Geoscience & Remote Sensing, 2015, 53(3): 1463-1474

[16] BillorN, HadiA S, VellemanP F. BACON: blocked adaptive computationally efficient outlier nominators [J]. Computational Statistics & Data Analysis, 2000, 34(99): 279-298

[17] DuB, ZhangL-pei. Random-selection-based anomaly detector for hyperspectral imagery [J]. IEEE Transactions on Geoscience & Remote Sensing, 2011, 49(5): 1578-1589

[18] GaoL-r, GuoQ-d, PlazaA. Probabilistic anomaly detector for remotely sensed hyperspectral data [J]. Journal of Applied Remote Sensing, 2014, 8(1): 083538

[19] LouC, ZhaoH-jie. Local density-based anomaly detection in hyperspectral image [J]. Journal of Applied Remote Sensing, 2015, 91095070-095070

[20] MoleroA J M, GarzonE M, GarciaI. Analysis and optimizations of global and local versions of the RX algorithm for anomaly detection in hyperspectral data [J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2013, 6(2): 801-814

[21] LiW, DuQ. Unsupervised nearest regularized subspace for anomaly detection in hyperspectral imagery [C]. IEEE, Geoscience and Remote Sensing Symposium (IGARSS 2013), 201410551058

[22] LiW, TramelE W, PrasadS. Nearest regularized subspace for hyperspectral classification [J]. IEEE Transactions on Geoscience & Remote Sensing, 2014, 52(1): 477-489

[23] PlazaA, MartinezP, PlazaJ. Dimensionality reduction and classification of hyperspectral image data using sequences of extended morphological transformations [J]. IEEE Transactions on Geoscience & Remote Sensing, 2005, 43(3): 466-479

[24] HerwegJ A, KerekesJ P, WeatherbeeO. Algorithms and technologies for multispectral, hyperspectral, and ultraspectral imagery XVIII (SPIE 2012) [C]. SPIE Proceedings: SPIE., 2012839028

[25] MoleroA J M, GarzonE M, GarciaI. Anomaly detection based on a parallel kernel RX algorithm for multicore platforms [J]. Journal of Applied Remote Sensing, 2012, 6(1): 623-632