Share Email Print
cover

Proceedings Paper

Performance characterization of a combined material identification and screening algorithm
Author(s): Robert L. Green; Michael D. Hargreaves; Craig M. Gardner
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Portable analytical devices based on a gamut of technologies (Infrared, Raman, X-Ray Fluorescence, Mass Spectrometry, etc.) are now widely available. These tools have seen increasing adoption for field-based assessment by diverse users including military, emergency response, and law enforcement. Frequently, end-users of portable devices are non-scientists who rely on embedded software and the associated algorithms to convert collected data into actionable information. Two classes of problems commonly encountered in field applications are identification and screening. Identification algorithms are designed to scour a library of known materials and determine whether the unknown measurement is consistent with a stored response (or combination of stored responses). Such algorithms can be used to identify a material from many thousands of possible candidates. Screening algorithms evaluate whether at least a subset of features in an unknown measurement correspond to one or more specific substances of interest and are typically configured to detect from a small list potential target analytes. Thus, screening algorithms are much less broadly applicable than identification algorithms; however, they typically provide higher detection rates which makes them attractive for specific applications such as chemical warfare agent or narcotics detection. This paper will present an overview and performance characterization of a combined identification/screening algorithm that has recently been developed. It will be shown that the combined algorithm provides enhanced detection capability more typical of screening algorithms while maintaining a broad identification capability. Additionally, we will highlight how this approach can enable users to incorporate situational awareness during a response.

Paper Details

Date Published: 29 May 2013
PDF: 13 pages
Proc. SPIE 8726, Next-Generation Spectroscopic Technologies VI, 87260F (29 May 2013); doi: 10.1117/12.2015953
Show Author Affiliations
Robert L. Green, Thermo Fisher Scientific Inc. (United States)
Michael D. Hargreaves, Thermo Fisher Scientific Inc. (United States)
Craig M. Gardner, Thermo Fisher Scientific Inc. (United States)


Published in SPIE Proceedings Vol. 8726:
Next-Generation Spectroscopic Technologies VI
Mark A. Druy; Richard A. Crocombe, Editor(s)

© SPIE. Terms of Use
Back to Top