Variable subspace model for hyperspectral anomaly detection

Detection of anomalies in a large area is an important objective in remote sensing using hyperspectral imaging system. Some conventional anomaly detectors for such an application have a user-specified parameter which is the dimension of the clutter subspace. The range of possible values for this parameter is typically large. In this paper, an anomaly detector with a different parameter is proposed. The anomaly detector partials out the effect of the clutter subspace from a pixel by predicting each spectral component of a pixel using a linear combination of the clutter subspace. The dimension of the clutter subspace can vary from one spectral component to another one and is determined by a parameter based on the maximized squared correlations. The Mahalanobis distance of the resulting residual is defined as the anomaly detector. The experimental results are obtained by implementing the anomaly detector as a global anomaly detector in unsupervised mode with background statistics computed from hyperspectral data cubes with wavelengths in the visible and near-infrared range. The results show that the parameter in the anomaly detector has a significantly reduced number of possible values and the values can be determined automatically in an iterative mode.

[1]  N. Nasrabadi,et al.  Kernel RX : A new nonlinear anomaly detector , 2005 .

[2]  Francisco J. Prieto,et al.  Using Angles to Identify Concentrated Multivariate Outliers , 2001, Technometrics.

[3]  Alan P. Schaum,et al.  Joint hyperspectral subspace detection derived from a Bayesian likelihood ratio test , 2002, SPIE Defense + Commercial Sensing.

[4]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[5]  Mark Bernhardt,et al.  Hyperspectral clutter statistics, generative models, and anomaly detection , 2006, SPIE Defense + Commercial Sensing.

[6]  E. M. Winter,et al.  Anomaly detection from hyperspectral imagery , 2002, IEEE Signal Process. Mag..

[7]  E. Lo.  Outlier Detection Algorithm for Hyperspectral Imagery Based on Conditioning on Background Subspace , 2009 .

[8]  Alan P. Schaum,et al.  Application of stochastic mixing models to hyperspectral detection problems , 1997, Defense, Security, and Sensing.

[9]  Xiaoli Yu,et al.  Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution , 1990, IEEE Trans. Acoust. Speech Signal Process..

[10]  John Fisher,et al.  Hyperspectral analysis and target detection system for the Adaptive Spectral Reconnaissance Program (ASRP) , 1998, Defense, Security, and Sensing.

[11]  Chein-I Chang,et al.  Anomaly detection and classification for hyperspectral imagery , 2002, IEEE Trans. Geosci. Remote. Sens..

[12]  Edisanter Lo.  Hyperspectral outlier detector based on conditional distributions , 2010, Defense + Commercial Sensing.

[13]  A. Schaum.  Hyperspectral detection algorithms: from old ideas to operational concepts to next generation , 2006, SPIE Defense + Commercial Sensing.

[14]  John P. Kerekes,et al.  Unresolved target detection blind test project overview , 2010, Defense + Commercial Sensing.

[15]  John Ingram,et al.  Hyperspectral anomaly detection based on minimum generalized variance method , 2008, SPIE Defense + Commercial Sensing.

[16]  P. Rousseeuw,et al.  A fast algorithm for the minimum covariance determinant estimator , 1999 .

[17]  Maria Petrou,et al.  Spectral Unmixing With Negative and Superunity Abundances for Subpixel Anomaly Detection , 2009, IEEE Geoscience and Remote Sensing Letters.

[18]  Edisanter Lo.  Maximized subspace model for hyperspectral anomaly detection , 2011, Pattern Analysis and Applications.

[19]  Vic Barnett,et al.  Outliers in Statistical Data , 1980 .

[20]  E. Lo.  Hyperspectral Anomaly Detector based on Variable Number of Predictors , 2010 .

[21]  Alan P. Schaum,et al.  A hyperspectral anomaly detector based on partialing out a clutter subspace , 2009, Defense + Commercial Sensing.

[22]  A. P. Schaum,et al.  Hyperspectral anomaly detection beyond RX , 2007, SPIE Defense + Commercial Sensing.