Share Email Print
cover

Proceedings Paper

Bayesian paradox in homeland security and homeland defense
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In this paper we discuss a rather surprising result of Bayesian inference analysis: performance of a broad variety of sensors depends not only on a sensor system itself, but also on CONOPS parameters in such a way that even an excellent sensor system can perform poorly if absolute probabilities of a threat (target) are lower than a false alarm probability. This result, which we call Bayesian paradox, holds not only for binary sensors as discussed in the lead author's previous papers, but also for a more general class of multi-target sensors, discussed also in this paper. Examples include: ATR (automatic target recognition), luggage X-ray inspection for explosives, medical diagnostics, car engine diagnostics, judicial decisions, and many other issues.

Paper Details

Date Published: 3 June 2011
PDF: 10 pages
Proc. SPIE 8019, Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense X, 801911 (3 June 2011); doi: 10.1117/12.883323
Show Author Affiliations
Tomasz Jannson, Physical Optics Corp. (United States)
Thomas Forrester, Physical Optics Corp. (United States)
Wenjian Wang, Physical Optics Corp. (United States)


Published in SPIE Proceedings Vol. 8019:
Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense X
Edward M. Carapezza, Editor(s)

© SPIE. Terms of Use
Back to Top