In the operational airport environment, the rapid identification of potentially
hazardous materials such as improvised explosive devices, chemical warfare agents and
flammable and explosive liquids is increasingly critical. Peroxide-based explosives pose
a particularly insidious threat because they can be made from commonly available and
relatively innocuous household chemicals, such as bleach and hydrogen peroxide.
Raman spectroscopy has been validated as a valuable tool for rapid identification of
chemicals, explosives, and narcotics and their precursors while allowing "line-of-sight"
interrogation through bottles or other translucent containers. This enables safe
identification of both precursor substances, such as acetone, and end-products, such as
TATP, without direct sampling, contamination and exposure by security personnel.
To date, Raman systems have been laboratory-based, requiring careful operation
and maintenance by technology experts. The capital and ongoing expenses of these
systems is also significant. Recent advances in Raman component technologies have
dramatically reduced the footprint and cost, while improving the reliability and ease of
use of Raman spectroscopy systems. Such technologies are not only bringing the lab to
the field, but are also protecting civilians and security personnel in the process.
In recent years a number of analytical devices have been proposed and marketed specifically to enable field-based material identification. Technologies reliant on mass, near- and mid-infrared, and Raman spectroscopies are available today, and other platforms are imminent. These systems tend to perform material recognition based on an on-board library of material signatures. While figures of merit for traditional quantitative analytical sensors are broadly established (e.g., SNR, selectivity, sensitivity, limit of detection/decision), measures of performance for material identification systems have not been systematically discussed. In this paper we present an approach to performance characterization similar in spirit to ROC curves, but including elements of precision-recall curves and specialized for the intended-use of material identification systems. Important experimental considerations are discussed, including study design, sources of bias, uncertainty estimation, and cross-validation and the approach as a whole is illustrated using a commercially available handheld Raman material identification system.
Proceedings Volume Editor (3)
This will count as one of your downloads.
You will have access to both the presentation and article (if available).