16 April 2008 Bayesian performance metrics of binary sensors in homeland security applications
Author Affiliations +
Abstract
Bayesian performance metrics, based on such parameters, as: prior probability, probability of detection (or, accuracy), false alarm rate, and positive predictive value, characterizes the performance of binary sensors; i.e., sensors that have only binary response: true target/false target. Such binary sensors, very common in Homeland Security, produce an alarm that can be true, or false. They include: X-ray airport inspection, IED inspections, product quality control, cancer medical diagnosis, part of ATR, and many others. In this paper, we analyze direct and inverse conditional probabilities in the context of Bayesian inference and binary sensors, using X-ray luggage inspection statistical results as a guideline.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Tomasz P. Jannson, Tomasz P. Jannson, Thomas C. Forrester, Thomas C. Forrester, } "Bayesian performance metrics of binary sensors in homeland security applications", Proc. SPIE 6943, Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense VII, 69430O (16 April 2008); doi: 10.1117/12.778279; https://doi.org/10.1117/12.778279
PROCEEDINGS
12 PAGES


SHARE
Back to Top