A machine learning approach to hyperspectral detection of solid targets

We describe and compare two approaches to solid subpixel target detection in hyperspectral imagery. The first approach requires explicit models for both the target and the background, and employs a generalized likelihood ratio in order to obtain a detector that is optimized to those specific models. When this approach is most successful, a closed-form solution is obtained that permits the detector to be efficiently applied. A specific example of this approach is outlined in some detail, leading to the elliptically-contoured finite-target matched filter (EC-FTMF), a variant of the classical FTMF algorithm that uses a multivariate t-distribution instead of a Gaussian as the model for the background. The second approach also requires an explicit model of the target, but does not need a model for the background. In this second approach, matched pairs of data samples are created: for each pixel in the original hyperspectral image, a corresponding pixel is generated by implanting the target into the original pixel. These matched pairs are used as training data for a machine learning algorithm to classify pixels as either non-target or target. Here we use a support vector machine, but the matched pair machine learning (MPML) framework does not restrict the choice of classifier type. Detectors using both approaches are applied both to simulated data (with Gaussian and with multivariate fit distributed backgrounds) and to real hyperspectral data with known, referenced targets.

[1]  Marcus S. Stefanou,et al.  A Method for Assessing Spectral Image Utility , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[2]  E. J. Kelly Performance of an adaptive detection algorithm; rejection of unwanted signals , 1989 .

[3]  James Theiler,et al.  Elliptically Contoured Distributions for Anomalous Change Detection in Hyperspectral Imagery , 2010, IEEE Geoscience and Remote Sensing Letters.

[4]  James Theiler,et al.  Matched-Pair Machine Learning , 2013, Technometrics.

[5]  James Theiler,et al.  Transductive and matched-pair machine learning for difficult target detection problems , 2014, Defense + Security Symposium.

[6]  John P. Kerekes,et al.  Statistics of hyperspectral imaging data , 2001, SPIE Defense + Commercial Sensing.

[7]  Alan Schaum,et al.  Continuum fusion solutions for replacement target models in electro-optic detection. , 2014, Applied optics.

[8]  James Theiler,et al.  Improved matched-filter detection techniques , 1999, Optics & Photonics.

[9]  L.L. Scharf,et al.  Adaptive matched subspace detectors and adaptive coherence estimators , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.

[10]  A. Hayden,et al.  Determination of trace-gas amounts in plumes by the use of orthogonal digital filtering of thermal-emission spectra. , 1996, Applied optics.

[11]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[12]  John Kerekes,et al.  The target implant method for predicting target difficulty and detector performance in hyperspectral imagery , 2011, Defense + Commercial Sensing.

[13]  I. Reed,et al.  Rapid Convergence Rate in Adaptive Arrays , 1974, IEEE Transactions on Aerospace and Electronic Systems.

[14]  James Theiler,et al.  EC-GLRT: Detecting Weak Plumes in Non-Gaussian Hyperspectral Clutter Using an Elliptically-Contoured Generalized Likelihood Ratio Test , 2008, IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium.

[15]  James Theiler,et al.  Simplex ACE: a constrained subspace detector , 2017 .

[16]  Rama Chellappa,et al.  Hybrid Detectors for Subpixel Targets , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  L.L. Scharf,et al.  Adaptive coherence estimation for radar signal processing , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.

[18]  Dimitris G. Manolakis,et al.  A taxonomy of algorithms for chemical vapor detection with hyperspectral imaging spectroscopy , 2005, SPIE Defense + Commercial Sensing.

[19]  A. Fraser,et al.  Characterizing non-Gaussian clutter and detecting weak gaseous plumes in hyperspectral imagery , 2005 .

[20]  James Theiler,et al.  Local background estimation and the replacement target model , 2017, Defense + Security.

[21]  T. Cooley,et al.  Performance evaluation of hyperspectral detection algorithms for subpixel objects , 2010, Defense + Commercial Sensing.

[22]  Daniel R. Fuhrmann,et al.  A CFAR adaptive matched filter detector , 1992 .

[23]  Dimitris G. Manolakis,et al.  Using elliptically contoured distributions to model hyperspectral imaging data and generate statistically similar synthetic data , 2004, SPIE Defense + Commercial Sensing.

[24]  Steven M. Adler-Golden,et al.  Hyperspectral Detection and Identification with Constrained Target Subspaces , 2008, IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium.

[25]  David W. Messinger,et al.  The SHARE 2012 data campaign , 2013, Defense, Security, and Sensing.

[26]  Louis L. Scharf,et al.  The adaptive coherence estimator: a uniformly most-powerful-invariant adaptive detection statistic , 2005, IEEE Transactions on Signal Processing.

[27]  Alan P. Schaum,et al.  Enough with the additive target model , 2014, Defense + Security Symposium.

[28]  Fred A. Kruse,et al.  Analysis of Imaging Spectrometer Data Using $N$ -Dimensional Geometry and a Mixture-Tuned Matched Filtering Approach , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[29]  Alan Schaum Continuum fusion versions of the finite target matched filter for sub-pixel detection , 2013, 2013 5th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS).

[30]  Dimitris G. Manolakis,et al.  Hyperspectral matched filter with false-alarm mitigation , 2012 .

[31]  James Theiler,et al.  Closed-Form Detector for Solid Sub-Pixel Targets in Multivariate T-Distributed Background Clutter , 2018, IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium.

[32]  Gary A. Shaw,et al.  Hyperspectral Image Processing for Automatic Target Detection Applications , 2003 .