Multi-spectral and Thermography Imaging Techniques for the Investigation of a 15th Century Wall Painting

When planning the restoration of an artwork, the good practice involves the evaluation of the item healthiness before starting the common operation of cleaning, consolidation, etc., possibly through non-invasive techniques that supply meaningful information about the whole item. Motivated by this need, a plethora of imaging techniques are used in cultural heritage diagnostic typically borrowed from other applications – e.g. medical diagnostics, nondestructive testing, etc., and then tailored for inspecting cultural heritage objects. In the inspection of a painting, hyper- and multi- spectral techniques are commonly used to analyze the outer layers (varnish, pictorial and drawing) while X-ray, tomography, and many other can be employed to investigate its inner structure. Although highly desirable, a single technique providing all the info about a painting is still not available, thus it is of great interest defining protocols that could optimally exploit the complementarities of a limited number of techniques. To this aim, the present paper shows the combined use of the Hypecolorimetric Multispectral Imaging (HMI) and that of the Pulse-Compression Thermography (PuCT) on a 15th century wall painting attributed to the Italian artist Antonio del Massaro, also known as Pastura, and representing the Madonna with the Child and the Saints Jerome and Francis. In particular, HMI is a multispectral imaging method working from the ultraviolet to the near infrared region, exploiting advanced processing based on artificial intelligence to define hypercolorimetric coordinates. Such approach guarantees a thorough analysis of the outer layers, underlining previous restorations, varnish alterations and allowing the pigments to be classified from a comparison with a large database. The PuCT method adopted here has been tailored for the specific needs of artworks’ inspection and it allows for a safe imaging of the multilayer structure of paintings, and hence the stratigraphy analysis, through a suitable processing of the time-domain thermal response. The capabilities and the complementarities of the two techniques, whose info can also be fused through postprocessing techniques, are illustrated in detail in this paper. A false-color imaging approach is also proposed to improve the readability and analysis of the thermography results.