Share Email Print
cover

Proceedings Paper • new

Method of sensitivity analysis in anomaly detection algorithms for hyperspectral images
Author(s): Adam J. Messer; Kenneth W. Bauer
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Anomaly detection within hyperspectral images often relies on the critical step of thresholding to declare the specific pixels based on their anomaly scores. When the detector is built upon sound statistical assumptions, this threshold is often probabilistically based, such as the RX detector and the chi-squared threshold. However, when either the detector lacking statistical framework or the background pixels of the image violate the required assumptions, the approach to thresholding is complicated and can resolve into performance instability. We present a method to test the sensitivity thresholding to small changes in the characteristics of the anomalies based on their Mahalanobis distance to the background class. In doing so, we highlight issues in detectors thresholding techniques comparing statistical approaches against heuristic methods of thresholding.

Paper Details

Date Published: 5 May 2017
PDF: 9 pages
Proc. SPIE 10198, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXIII, 101980U (5 May 2017); doi: 10.1117/12.2260965
Show Author Affiliations
Adam J. Messer, Air Force Institute of Technology (United States)
Kenneth W. Bauer, Air Force Institute of Technology (United States)


Published in SPIE Proceedings Vol. 10198:
Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXIII
Miguel Velez-Reyes; David W. Messinger, Editor(s)

© SPIE. Terms of Use
Back to Top