Principal Component Analysis of Spectral Data: A Contribution to the Knowledge of the Materials Constituting Works of Art

The use of totally non-destructive techniques such as image spectroscopy for diagnosing paintings makes it possible to obtain a large amount of spectral data that provides information concerning the composition of works of art. Here, we stress how statistical treatments, such as principal component analysis (PCA), applied to 2-D data, can contribute to a better knowledge of the work of art itself and of the distribution of the materials that constitute it. Laboratory tests, as well as applications to actual paintings, will be presented and discussed.