&lt;title&gt;Performance comparison of phenomenology-based features to generic features for false alarm reduction in UWB SAR imagery&lt;/title&gt;

Marble, Jay A.; Gorman, Jason J.

doi:10.1117/12.357640

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A standard approach to discriminating between real objects and false alarms is to measure the size of the object [4]. This can be done in two or three dimensions through a process that isolates and extracts the landniine signature, then measures the desired features.…”

Section: D Size/shape Featuresmentioning

confidence: 99%

Landmine signatures in ground-penetrating radar

2001

Self Cite

View full text Add to dashboard Cite

The Mine Hunter/Killer system employs a ground penetrating radar (GPR) sensor. Twenty antennas sample a 3m swath to measure a 3D depth return from the earth as the vehicle moves forward in a lane. Data has been collected on shallow and deep, metal and low metal landmines. Sample signaturesfrom a metal and plastic cased landmine buried at 6 " are presented. In each example a hyperbolic signature is observed. Twofeature sets that exploit the hyperbolic shapeforfalse alarm reduction arepresented. The first uses a pixel clustering technique to isolate the hyperbola in three dimensions. A vector of size/shape features is extracted and combined with a quadratic polynomial discriminant (QPD) into a single value. The secondfeature set utilizes the radon transform. The radon transform sums the tails ofthe hyperbola allowing the algorithm to dfferentiate between surface clutter, which tends to be oriented horizontally in depth, and the diagonals ofthe hyperbola. Performance curvesfor both the 3D size/shapefeatures and the radon feature are presented.

show abstract

Section: D Size/shape Featuresmentioning

confidence: 99%

Landmine signatures in ground-penetrating radar

2001

Self Cite

View full text Add to dashboard Cite

show abstract

“…[6] apply a non-parametric density estimator to characterize the behavior of data using distributions of signal energy over subapertures at different resolutions, and then use these densities for discrimination of target, tree, and clutter classes. Marble and Gorman [7] use phenomenology based features derived from scattering physics predictions combined in a quadratic classifier to discriminate between targets and trees. Finally, Nguyen e t d .…”

Section: Introductionmentioning

confidence: 99%

Matched subspace detectors for discrimination of targets from trees in SAR imagery

Sharma

Moses²

Conference Record of the Thirty-Fourth Asilomar Conference on Signals, Systems and Computers (Cat. No.00CH37154)

View full text Add to dashboard Cite

Several approaches that exploit forest clutter structure We investigate the use of subspace-based detectors for discriminating vehicles from trees in low frequency synthetic aperture imagery. We model tree scattering as structured isotropic interference responses and model dominant vehicle scattering as dihedral responses. We form linear subspaces of tree and target responses, and apply subspace-based detection methods developed by Scharf and Friedlander. Analysis on synthetic tree and target models show the viability of this approach. Preliminary results on measured imagery provide lower performance, suggesting the need for improved data calibration and improved scattering models of trees at low frequencies.

show abstract

<title>Performance comparison of phenomenology-based features to generic features for false alarm reduction in UWB SAR imagery</title>

Cited by 2 publications

References 0 publications

Landmine signatures in ground-penetrating radar

Landmine signatures in ground-penetrating radar

Matched subspace detectors for discrimination of targets from trees in SAR imagery

Contact Info

Product

Resources

About