An innovative protocol for the study of painting materials involving the combined use of MA-XRF maps and hyperspectral images

X-ray fluorescence (XRF) and reflectance spectroscopy (RS) are commonly used for the characterization of painting materials. It is well known that former provides chemical fingerprint pictorial layers, while latter returns molecular description pigments constituting uppermost layers. Even if these two techniques cannot unveil stratigraphy, their synergetic application describes materials employed realizing panels represents a key turn non-invasive scientific analysis works art. However, potential cross-comparison between XRF RS not fully exploited yet. The measurement points often barely match, they usually few isolated spots spread over whole surface painting; facts limit mutual exchange information data sets can lead to losing details. In this scenario, mapping (MA-XRF) hyperspectral imaging (HRI) provide connection channel promises be decisive tool strengthen relationship and, therefore, deepen knowledge about case studies. Due spatial localization contain, maps only straightforward reference comparing but also three-dimensional collection elemental images. By applying computer vision statistical methods such as spectral angle mapper (SAM), it possible implement an innovative approach exploits features, obtained from spectra, improve comprehension aspects given by RS, vice versa. Once we discussed main issues behind our approach, applied analyze Chariot Race Giorgio De Chirico (1928–1929, oil on canvas, Pinacoteca di Brera, Milan, Italy). results reflect complexity painting, even some spectra identified method peculiar ascribable recognizable pigments, correspondence distributions SAM defines mixture matches artist in his “Small Treatise Pictorial Technique” (De Abscondita, 2019).