Published: June 24, 2019
Charles H. Camp
In this work, a new software library is presented for performing multivariate curve resolution (MCR)analysis, a chemometric method for elucidating signatures of analytes (endmember extraction) and their relative abundance (regression) from a series of mixture measurements, without necessarya priori knowledge of abundances or signatures of the analytes for each of the input measurements. This software library, written in Python, enables users to create an MCR processing pipeline with their choice ofconstraints (e.g., non-negative abundances), their choice of regressors, such as least-squares or ridge regression, and their choice of error functions to minimize. Further, users can apply different constraints and regressors for endmember extraction and regression. Finally, this library enables users to use their own developed constraints, regressors, and error functions or import them from existing libraries.
Citation: Journal of Research (NIST JRES) -
NIST Pub Series: Journal of Research (NIST JRES)
Pub Type: NIST Pubs
chemometrics, endmember extraction, multivariate curve resolution, quantitative analysis, spectral unmixing.
Created June 24, 2019, Updated June 24, 2019