Deterioration estimation of paintings by means of combined 3D and hyperspectral data analysis

Abstract

Deterioration of artwork, in particular paintings, can be produced by environmental factors such as temperature fluctuations, relative humidity variations, ultraviolet radiation and biological factors among others. The effects of these parameters produce changes in both the painting structure and chemical composition. While well established analytical methodologies, such as those based in Raman Spectroscopy and FTIR Spectroscopy require the extraction of a sample for its inspection, other approaches such as hyperspectral imaging and 3D scanning present advantages for in-situ, noninvasive analysis of artwork. In this paper we introduce a novel system and the related methodology to acquire process, generate and analyze 4D data of paintings. Our system is based on non-contact techniques and is used to develop analytical tools which extract rich 3D and hyperspectral maps of the objects, which are processed to obtain accurate quantitative estimations of the deterioration and degradation present in the piece of art. In particular, the construction of 4D data allows the identification of risk maps on the painting representation, which can allow the curators and restorers in the task of painting state evaluation and prioritize intervention actions.