Image Spectroscopy is proposed, among other spectrophotometric methods, as a non-invasive, effective tool for monitoring colour alterations of ancient remains. The method consists in acquiring a sequence of digital images of the subject in narrow spectral bands, from the Visible to the near-IR range. The reflectance spectrum of any detail is derived from the image sequence, requiring no contact with the surface. The same measurement is repeated after a lapse of time and the spectra obtained in succession on the same regions are compared using multivariate analysis to detect any systematic deviation. Trends of colour alteration can thus be detected, for further investigation or conservation interventions. Other well-established spectroscopic techniques in the visible range, like Colorimetry and Integrating Sphere Photometry are surely well suited to the same aim: Colorimetry is helpful and easy in interpretation of colour changes, while a contact spectrophotometer gives surely the most reliable quantitative spectral measurements. Nevertheless, quantitative Image Spectroscopy has some peculiar advantages in the conservation field: it performs measurements on each pixel of the whole acquired image and not on sampling points; it requires no contact with the investigated surface and, as a last relevant issue, spectral data analysed by multivariate techniques are much more sensitive to faint alterations than three-coordinate colorimetry is. An instrument for Image Spectroscopy will be presented, with some applications devoted to conservation of contemporary art paintings, showing that for periodical, no-contact monitoring and detection of colour progressive alteration, Image Spectroscopy is the natural choice.
Image Spectroscopy for non-invasive monitoring and diagnostics of archaeological remains
PETRUCCI, Ferruccio Carlo
2007
Abstract
Image Spectroscopy is proposed, among other spectrophotometric methods, as a non-invasive, effective tool for monitoring colour alterations of ancient remains. The method consists in acquiring a sequence of digital images of the subject in narrow spectral bands, from the Visible to the near-IR range. The reflectance spectrum of any detail is derived from the image sequence, requiring no contact with the surface. The same measurement is repeated after a lapse of time and the spectra obtained in succession on the same regions are compared using multivariate analysis to detect any systematic deviation. Trends of colour alteration can thus be detected, for further investigation or conservation interventions. Other well-established spectroscopic techniques in the visible range, like Colorimetry and Integrating Sphere Photometry are surely well suited to the same aim: Colorimetry is helpful and easy in interpretation of colour changes, while a contact spectrophotometer gives surely the most reliable quantitative spectral measurements. Nevertheless, quantitative Image Spectroscopy has some peculiar advantages in the conservation field: it performs measurements on each pixel of the whole acquired image and not on sampling points; it requires no contact with the investigated surface and, as a last relevant issue, spectral data analysed by multivariate techniques are much more sensitive to faint alterations than three-coordinate colorimetry is. An instrument for Image Spectroscopy will be presented, with some applications devoted to conservation of contemporary art paintings, showing that for periodical, no-contact monitoring and detection of colour progressive alteration, Image Spectroscopy is the natural choice.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
