In recent years, several sensing devices capable of identifying unknown chemical and biological substances have been
commercialized. The success of these devices in analyzing real world samples is dependent on the ability of the on-board
identification algorithm to de-convolve spectra of substances that are mixtures. To develop effective de-convolution
algorithms, it is critical to characterize the relationship between the spectral features of a substance and its probability of
detection within a mixture, as these features may be similar to or overlap with other substances in the mixture and in the
library. While it has been recognized that these aspects pose challenges to mixture analysis, a systematic effort to
quantify spectral characteristics and their impact, is generally lacking. In this paper, we propose metrics that can be used
to quantify these spectral features. Some of these metrics, such as a modification of variance inflation factor, are derived
from classical statistical measures used in regression diagnostics. We demonstrate that these metrics can be correlated to
the accuracy of the substance's identification in a mixture. We also develop a framework for characterizing mixture
analysis algorithms, using these metrics. Experimental results are then provided to show the application of this
framework to the evaluation of various algorithms, including one that has been developed for a commercial device. The
illustration is based on synthetic mixtures that are created from pure component Raman spectra measured on a portable
device.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.