Adaptive anomaly detection using subspace separation for hyperspectral imagery

Kwon, Heesung; Der, Sandor Z.; Nasrabadi, Nasser M.

doi:10.1117/1.1614265

Cited by 149 publications

(78 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The last one is another real HSI dataset with a complex background distribution. For comparison, a series of existing state-of-the-art anomaly detection algorithms is used as benchmarks in the experiments, including global RX (GRX) [10], local RX (LRX) [37], kernel-RX (KRX) [14], CBAD [11], subspace RX (SSRX) [12,13], DWEST [21], BJSRD [24] and CRD [23]. Furthermore, the influences of different parameters settings on each HSI dataset are discussed in detail.…”

Section: Experimental Results and Analysismentioning

confidence: 99%

Hyperspectral Anomaly Detection via Background Estimation and Adaptive Weighted Sparse Representation

Zhu

Wen

2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:Anomaly detection is an important task in hyperspectral imagery (HSI) processing. It provides a new way to find targets that have significant spectral differences from the majority of the dataset. Recently, the representation-based methods have been proposed for detecting anomaly targets in HSIs. It is essential for this type of method to construct a valid background dictionary to distinguish anomaly and background accurately. In this paper, a novel hyperspectral anomaly detection method based on background estimation and adaptive weighted sparse representation has been proposed. Firstly, to obtain the effective background dictionary without anomaly information, a new background dictionary construction strategy is designed. Secondly, the sparse representation based on the constructed background dictionary is utilized on the dataset. Anomalies and background are distinguished through the response of the residual matrix. Thirdly, the residual matrix is weighted adaptively from global and local domains, which makes anomalies and background more discriminative. An important advantage of the proposed method is that it considers the properties of anomalies in both spectral and spatial domains. Experiments on three HSI datasets reveal that our proposed method achieves an outstanding detection performance compared with the other anomaly detection algorithms.

show abstract

Section: Experimental Results and Analysismentioning

confidence: 99%

Hyperspectral Anomaly Detection via Background Estimation and Adaptive Weighted Sparse Representation

Zhu

Wen

2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The RX algorithm is a maximum likelihood anomaly detection procedure that simplifies the clutter to being spatially white. Researchers have also adapted some classic approaches 6 Earlier laboratory experiments using SAR (synthetic aperture radar) imagery 13 produced compelling evidences to suggest that using conventional approaches to compare sample pairs, i.e., two-sample hypothesis tests treating the two samples individually, promotes an intolerable high number of meaningless detections (false alarms [FA]) in areas characterized by region discontinuities in the imagery. At that time, I proposed a post-processing method from mathematical morphology to account for those observations and adapted it to a SAR targetdetection system.…”

Section: Survey Of Prior Artmentioning

confidence: 99%

“…Targets are often found to be anomalous to its immediate surroundings. 6 Kwon, 2003. 25 Schweizer, 2000.…”

Section: Data Descriptionmentioning

confidence: 99%

“…13,14 The approach produced some surprisingly good results using real SAR data. 5 Yu, 1997 6 Kwon, 2003. 7 Stein, 2002 8 Stein, 2002.…”

Section: Survey Of Prior Artmentioning

confidence: 99%

“…However, using profiling one can express g 0 in terms of α and β and then replace g 0 with its new representation back in (6). Using profiling, the method of Lagrange multiplier was proposed 18 to attain the maximization of ζ by fixing (α,β) and then maximizing ζ with respect to g 0 (t i ) for i=1,…,n, subject to constraints 5 Yu, 1997.…”

Section: Logistic Modelmentioning

confidence: 99%

See 2 more Smart Citations

Innovative Statistical Inference for Anomaly Detection in Hyperspectral Imagery

Rosario¹

2004

View full text Add to dashboard Cite

Sensors and Electron Devices Directorate, ARLApproved for public release; distribution unlimited. Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. ii REPORT DOCUMENTATION PAGE

show abstract

Target Detection with Kernels

Nasrabadi

2009

Kernel Methods for Remote Sensing Data Analysis

View full text Add to dashboard Cite

Adaptive anomaly detection using subspace separation for hyperspectral imagery

Cited by 149 publications

References 11 publications

Hyperspectral Anomaly Detection via Background Estimation and Adaptive Weighted Sparse Representation

Hyperspectral Anomaly Detection via Background Estimation and Adaptive Weighted Sparse Representation

Innovative Statistical Inference for Anomaly Detection in Hyperspectral Imagery

Target Detection with Kernels

Contact Info

Product

Resources

About