Scientific investigation in the cultural heritage field is generally aimed at the characterization of the constituent materials and the conservation status of artworks. Since the 1990s, reflectance spectral imaging proved able to map pigments, reveal hidden details and evaluate the presence of restorations in paintings. Over the past two decades, hyperspectral imaging has further improved our understanding of paints and of its changes in time. In this work, we present an innovative hyperspectral camera, based on the Fourier transform approach, utilising an ultra-stable interferometer and we describe its advantages and drawbacks with respect to the commonly used line- and spectral-scanning methods. To mitigate the weaknesses of the Fourier transform hyperspectral imaging, we propose a strategy based on the virtual extension of the dynamic range of the camera and on the design of an illumination system with a balanced emission throughout the spectral range of interest. The hyperspectral camera was employed for the analysis of a painting from the “Album of Nasir al-din Shah”. By applying analysis routines based on supervised spectral unmixing, we demonstrate the effectiveness of our camera for pigment mapping. This work shows how the proposed hyperspectral imaging camera based on the Fourier transform is a promising technique for robust and compact in situ investigation of artistic objects in conditions compatible with museum and archaeological sites. Graphic abstract

The famous Persian illustrated manuscript's page called "Humay meets the Princess Humayun" from Louvre Museum was investigated using multispectral imaging, X-ray fluorescence mapping and Raman spectroscopy for a deepen study of its constitutive materials and painting techniques and to evaluate the state of conservation of the artwork. Cross image treatments were also performed in order to extract further knowledge from the artwork. Due to the extreme fragile nature of the object and its outstanding value in the collection of Louvre's Department of Decorative Arts, noninvasive and contactless diagnostic methods were chosen. Hypercolorimetric Multispectral Imaging (HMI), a portable 7-band multispectral system and its native digital image processing (DIP) software for image treatment, was used to integrate X-ray fluorescence (XRF) imaging mapping to multispectral and colorimetric dataset. It allowed to study the rich color palette, identify original materials and restoration interventions and eventual degradations of both pictorial layer and ancient paper. Punctual Raman spectroscopy was performed to confirm hypotheses of pigments characterization made on infrared and ultraviolet false color images (IRFC, UVFC). Image treatments crossing multispectral data and analytical punctual techniques investigated possible correlation between datasets providing interesting and easy-to-read distribution maps for chemical elements, which are characteristic for pigment identification. In particular, pigments not detectable through XRF for light-element limit of detection as ultramarine blue (lapis lazuli) and organic dyes were successfully identified and mapped. This integrated approach of multisource imaging data is a valuable tool for restorers and conservators, encouraging new protocols of noninvasive diagnostics for cultural heritage.

No abstract