Performance metrics for the evaluation of hyperspectral chemical identification systems
Name
Truslow-2016-Performance metrics.pdf
Size
1.1 MB
Format
Adobe PDF
Checksum (MD5)
13e5ef6e9fd298a0cf04875223f1c972
Author(s)
Ingle, Vinay
Truslow, Eric O.
Golowich, Steven E.
Manolakis, Dimitris G.
Date Issued
February 2016
Journal
Optical Engineering
Publisher
SPIE--Society of Photo-Optical Instrumentation Engineers
Citation
Truslow, Eric, Steven Golowich, Dimitris Manolakis, and Vinay Ingle. “Performance Metrics for the Evaluation of Hyperspectral Chemical Identification Systems.” Opt. Eng 55, no. 2 (February 10, 2016): 023106. ©2016 SPIE.
Version
Final published version
Abstract
Remote sensing of chemical vapor plumes is a difficult but important task for many military and civilian applications. Hyperspectral sensors operating in the long-wave infrared regime have well-demonstrated detection capabilities. However, the identification of a plume’s chemical constituents, based on a chemical library, is a multiple hypothesis testing problem which standard detection metrics do not fully describe. We propose using an additional performance metric for identification based on the so-called Dice index. Our approach partitions and weights a confusion matrix to develop both the standard detection metrics and identification metric. Using the proposed metrics, we demonstrate that the intuitive system design of a detector bank followed by an identifier is indeed justified when incorporating performance information beyond the standard detection metrics.
MIT Department
Lincoln Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1117/1.oe.55.2.023106
