Share Email Print

Proceedings Paper

On clairvoyant and universal tests for hyperspectral target detection
Author(s): D. Manolakis; E. Truslow; M. Pieper; A. Weisner; R. Bostick; T. Cooley
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The purpose of this paper is to discuss some recent approaches to composite hypothesis testing in the context of hyperspectral target detection applications. The primary tool for the development of practical detection algorithms is the generalized likelihood ratio test (GLRT), which does not have any “built-in” optimality criterion. The GLRT computes maximum likelihood estimates of the unknown parameters and uses them in the place of the true parameters. In this paper we review the asymptotic optimality properties of GLRTs and use them to discuss a family of universal tests known as competitive Bayes and Neyman-Pearson tests. Then, based on this background, we review the family of clairvoyant fusion algorithms and their applicability to hyperspectral target detection.

Paper Details

Date Published: 6 September 2019
PDF: 11 pages
Proc. SPIE 11130, Imaging Spectrometry XXIII: Applications, Sensors, and Processing, 1113004 (6 September 2019); doi: 10.1117/12.2529979
Show Author Affiliations
D. Manolakis, MIT Lincoln Lab. (United States)
E. Truslow, MIT Lincoln Lab. (United States)
M. Pieper, MIT Lincoln Lab. (United States)
A. Weisner, National Air and Space Intelligence Ctr. (United States)
R. Bostick, National Air and Space Intelligence Ctr. (United States)
T. Cooley, Air Force Research Lab. (United States)

Published in SPIE Proceedings Vol. 11130:
Imaging Spectrometry XXIII: Applications, Sensors, and Processing
Emmett J. Ientilucci, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?