|
Activity Number:
|
445
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #321230
|
|
|
Title:
|
Identification of Solids in Hyperspectral Images Using Spectral Features from Gaussian Basis Functions
|
|
Author(s):
|
Cory Lanker* and Milton O. Smith
|
|
Companies:
|
Lawrence Livermore National Laboratory and Lawrence Livermore National Laboratory
|
|
Keywords:
|
remote sensing ;
hyperspectral imaging ;
classification ;
wavelets ;
receiver operator characteristic ;
basis functions
|
|
Abstract:
|
We characterize solid materials through spectral features derived from LWIR reflectance spectra in order to improve compositional exploitation in a hyperspectral imaging sensor data cube. Our algorithm performs stepwise fitting of Gaussian basis function center frequencies and bandwidths using nonlinear least squares minimization. Experimental validation of the algorithm is performed using comparable Lorentz oscillator fits and using published values of crystalline and amorphous materials. Our aim is to use the derived basis function parameters to characterize the materials that are present in a data cube pixel. We demonstrate that there are material-specific properties of these spectral features that show subtle variability when considering changes in morphology or measurement conditions. The experimentally verified results include variability in material particle size, measurement angle, and atmospheric conditions for measurements of six pure minerals and their mixtures. We show that this approach has good initial identification results that are extendible across localized experimental conditions. Prepared by LLNL under Contract DE-AC52-07NA27344.
|
Authors who are presenting talks have a * after their name.
