Automatic Recognition of Color Pigments from Raman Spectrum Analysis

Giovanni Maria Farinella, Giovanni Gallo, Anna Maria Gueli, Filippo Stanco

Abstract


In this paper, an algorithm to automatically recognize pigments from Raman spectrum analysis is presented. The Raman spectroscopy is a technique to confirm the presence of specific pigments in a painting. Machine learning techniques are applied over an existing database of Raman spectra in order to find a set of characteristic features. These information are necessary to automatically recognize unknown pigments by using ad-hoc
similarity measure.

[DOI: 10.1685/CSC09211] About DOI

Keywords


Cultural Heritage, Raman Spectra, Pigments, Conservation and Restoration.

Full Text:

PDF


DOI: https://doi.org/10.1685/



Creative Commons License   Except where otherwise noted, content on this site is
  licensed under a Creative Commons 2.5 Italy License