Anomaly and target detection by means of nonparametric density estimation

We describe a novel completely non parametric high-dimension joint density estimation algorithm suited for anomaly and target detection using hyperspectral imaging. The new algorithm is compared against linear matched filter detection schemes with different available sample sizes, background statistics (MVN, GMM and non Gaussian). The new algorithm is shown to be superior in important cases.

[1]  Stefania Matteoli,et al.  Improved covariance matrix estimation: interpretation and experimental analysis of different approaches for anomaly detection applications , 2009, Remote Sensing.

[2]  G. F. Hughes,et al.  On the mean accuracy of statistical pattern recognizers , 1968, IEEE Trans. Inf. Theory.

[3]  Alan P. Schaum.  Advanced hyperspectral detection based on elliptically contoured distribution models and operator feedback , 2009, 2009 IEEE Applied Imagery Pattern Recognition Workshop (AIPR 2009).

[4]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[5]  J. Friedman.  Regularized Discriminant Analysis , 1989 .

[6]  Chris J. Willis.  Hyperspectral target detection using heavy-tailed distributions , 2009, Security + Defence.

[7]  Bor-Chen Kuo,et al.  A covariance estimator for small sample size classification problems and its application to feature extraction , 2002, IEEE Trans. Geosci. Remote. Sens..

[8]  Terry J. Wagner,et al.  Consistency of an estimate of tree-dependent probability distributions (Corresp.) , 1973, IEEE Trans. Inf. Theory.

[9]  S.R. Rotman,et al.  Improved covariance matrices for point target detection in hyperspectral data , 2008, 2009 IEEE International Conference on Microwaves, Communications, Antennas and Electronics Systems.

[10]  José M. F. Moura,et al.  Efficient detection in hyperspectral imagery , 2001, IEEE Trans. Image Process..

[11]  Stanley R. Rotman,et al.  Point target detection in segmented images , 2008, Optical Engineering + Applications.

[12]  C. N. Liu,et al.  Approximating discrete probability distributions with dependence trees , 1968, IEEE Trans. Inf. Theory.

[13]  S. K. Michael Wong,et al.  Comments on Approximating Discrete Probability Distributions with Dependence Trees , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Jing-Long Wang.  EXACT DISTRIBUTION OF THE MLE OF CONCENTRATION MATRICES IN DECOMPOSABLE COVARIANCE SELECTION MODELS , 2001 .

[15]  Brian J. Daniel,et al.  Urchin: an RX-derivative accounting for anisotropies in whitened clutter , 2010, Defense + Commercial Sensing.

[16]  José M. F. Moura,et al.  Hyperspectral imagery: Clutter adaptation in anomaly detection , 2000, IEEE Trans. Inf. Theory.

[17]  Dan G. Blumberg,et al.  Detecting anomalous objects in hyperspectral data using segmentation , 2008, Security + Defence.

[18]  Stefania Matteoli,et al.  Improved estimation of local background covariance matrix for anomaly detection in hyperspectral images , 2010 .

[19]  David A. Landgrebe,et al.  Covariance Matrix Estimation and Classification With Limited Training Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Jon Williamson.  Approximating discrete probability distributions with Bayesian networks , 2003 .

[21]  Jenq-Neng Hwang,et al.  Nonparametric multivariate density estimation: a comparative study , 1994, IEEE Trans. Signal Process..

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

[23]  M. Rahmati,et al.  Estimation of Probability Density Function by Dependence Tree Methods for Pattern Recognition Systems , 2006 .