Naslov
Method of and system for blind extraction of more than two pure components out of spectroscopic and spectrometric measurements of only two mixtures by means of sparse component analysis

Autori
Kopriva, Ivica ; Jerić, Ivanka ; Smrečki, Vilko

Nositelj prava
Institut Ruđer Bošković, Zagreb, Hrvatska, Zagreb, Hrvatska

Sažetak
The present invention generally relates to a computer-implemented system for processing data for the purpose of blind extraction of more than two pure components from two mixtures recorded in the fields of spectroscopy and spectrometry. Specifically, the invention is related to the application of the method of sparse component analysis, also known as underdetermined blind source separation, to blind decomposition of spectroscopic data consisting of two mixtures X into more than two pure components S and concentration matrix A. Spectroscopic data refers to data gathered by nuclear magnetic resonance (NMR) spectroscopy, electron paramagnetic resonance (EPR) spectroscopy, infrared (IR) spectroscopy, ultraviolet (UV) spectroscopy, Raman spectroscopy or mass spectrometry. Two mixtures are either analyzed in a recording domain or in a first new representation domain by using linear transform T1, wherein pure components in the first new representation domain are sparser than in the recording domain. The number of pure components and mixing matrix are estimated in either the recording domain or the first new representation domain by means of a data clustering algorithm. The pure components are estimated by means of linear programming, convex programming with quadratic constraint ( -norm based constraint) or quadratic programming method with -norm based constraint in either the recording domain, the first new representation domain or the second new representation domain, wherein the second new representation domain is obtained through another linear transform T2 and the second new representation domain must be the domain where the results are presented. The estimated pure components are ranked using negentropy based criterion. Components with negentropy measure that differs 10 orders of magnitudes or more from the negentropy of the majority of the components are classified as outliers and eliminated. If pure components are estimated in the first new representation domain, inverse transform is applied to estimate pure components to transfrom them back into recording domain of the two mixtures.

Izvorni jezik
Engleski

Znanstvena područja
Kemija, Računarstvo

POVEZANOST RADA