Date of Award
Raman spectroscopy, Multivariate analysis, Raman spectroscopy Spectral identity mapping multivariate analysis hyperspectral analysis minerals imaging
With increased use of Raman spectroscopic instrumentation for material analysis there has also been an increase in the amount of acquired Raman spectral data. Because of this, there is a clear need to develop and implement advanced spectral analysis techniques. This is especially true in cases where limited reference data may be available and large data sets need to be interpreted. Raman spectral analysis and standardization techniques, along with a foundation for a comprehensive repository of Raman spectral data, will be described in this thesis. The main focus will be automated analysis and standardization of Raman spectral data using spectral identity mapping (SIM). In addition, details on how to promote widespread access to the analyzed data and SIM techniques will be given. SIM, a statistical spectral analysis method, is useful as either a stand-alone data classification method or as a factor analysis step that precedes other multivariate approaches. SIM for calibrating spectra utilizes multivariate processing algorithms that can differentiate spectra according to the intrinsic nature of their spectral shapes. SIM also enables spectral identity mapping to be performed on unknown samples by calculating a set of scores giving the most likely match for given set of spectral information. A SIM database of calibrated spectra and a proposal to utilize SIM matching algorithms via the internet was developed. There are currently few searchable databases available for Raman spectral data. Also, while there are many internet platforms available to publish spectra, some can be difficult to implement and most do not provide data to users in a way that is educational, engaging and fully oriented to the Raman community. The software libraries given here provide a set of tools for data-basing, searching and interpreting spectral data while encouraging user/client participation in order to grow the spectral libraries. The hope is that the SIM results and the database developed here will promote SIM for more extens
Cannon, Robert William, "Automated Spectral Identification of Materials Using Spectral Identity Mapping" (2013). ETD Archive. 761.