Přístupnostní navigace
E-application
Search Search Close
Publication detail
HOLUB, D. POŘÍZKA, P. KAISER, J.
Original Title
Cluster analysis of spectroscopic data in the principal component space
Type
conference proceedings
Language
English
Original Abstract
Laser-Induced Breakdown Spectroscopy (LIBS) is commonly used for mapping heterogeneous samples with unknown chemical composition. After adequate pretreatment of spectroscopic data and choosing the right chemometric methods, the visualization of individual matrices is obtained. However, the outcome from heterogeneous samples are blurred, non-distinct clusters. This phenomenon corrupts the possibility to distinguish individual matrices and provide proper clustering. The experiment is designed to simulate an interference, where two different matrices (or elements) meet. The ablation spot hits the boundary of these two compounds, in this case Iron (Fe) and Aluminium (Al), with controlled step and therefore known ablated Fe:Al ratio. The ratio between Iron and Aluminium is determined using Principal Components Analysis (PCA), clustering and computing the Euclidian distance between clusters. The data is pretreated using methods such as detecting and deleting outliers, different kinds of data standartization and several kinds of approximation.
Keywords
LIBS, Principal Component Analysis, Cluster analysis
Authors
HOLUB, D.; POŘÍZKA, P.; KAISER, J.
Released
23. 10. 2019
Publisher
Spektroskopická společnost Jana Marka Marci
Location
Ke Karlovu 2027/3, 120 00 Praha 2 - Nové Město
ISBN
978-80-88195-13-9
Pages from
208
Pages to
Pages count
293
URL
https://emslibs.com/wp-content/uploads/2019/09/b3.2.pdf
BibTex
@proceedings{BUT159467, editor="Daniel {Holub} and Pavel {Pořízka} and Jozef {Kaiser}", title="Cluster analysis of spectroscopic data in the principal component space", year="2019", pages="208--208", publisher="Spektroskopická společnost Jana Marka Marci", address="Ke Karlovu 2027/3, 120 00 Praha 2 - Nové Město", isbn="978-80-88195-13-9", url="https://emslibs.com/wp-content/uploads/2019/09/b3.2.pdf" }