Salta al contenuto principale
Passa alla visualizzazione normale.

FRANCESCA DI SALVO

FDA dimension reduction techniques and components separation in Fourier-transform infrared spectroscopy

  • Autori: Francesca Di Salvo, Elena Piacenza, Delia Francesca Chillura Martino
  • Anno di pubblicazione: 2020
  • Tipologia: Contributo in atti di convegno pubblicato in volume
  • OA Link: http://hdl.handle.net/10447/514112

Abstract

FTIR spectroscopy is a measurement technique used to obtain an infrared spectrum of absorption of a solid (or a liquid or a gas), for the characterization of specific chemical components of materials. When repeated measures are taken on samples of materials, the result is a collection of spectra representing a set of samples from continous functions (signals) defined in the domain of the frequencies. An unifying approach to the study of a collection of FTIR spectra is proposed to deal with the presence of random shifts in the peaks, the identification of representative spectra and finally the characterization of the observed differences: in the functional data framework, the performance of a new proposal for alignment of curves and their assignment to clusters of shapes with an intrinsic order is presented and discussed.