Data Engineering for Materials Identification, Damage Assessment and Restoration of Cultural Objects

Abstract

Cultural objects and art works need ongoing conservation interventions in order to be available for the generations to come. The most object-friendly analysis approaches are based on non destructive techniques (NDTs) that allow both the materials characterization/evaluation as well as the decay detection and assessment of cultural artifacts.

Non destructive testing and evaluation includes the employment of several methods such as the well-established technique of Diffuse Reflectance Spectroscopy with Fiber Optics (FORS). FORS allows the reflectance spectral analysis of the pigments used in artifacts, which leads to their identification. Such techniques produce output with large volumes of data for each different pigment used in objects. In this work, we present a data management solution that contributes with (1) a library of known reference pigments/colors of archaeological objects along with (2) a proposed novel pattern matching technique that allows the automatic classification of any new pigment that is recovered from cultural objects using the FORS measurements. The proposed technique is based on a k-NN classifier. The experimental evaluation results of the proposed technique show that the data processing proposed is both effective and efficient. Feedback for the proposed approach is particularly encouraging as it allows automation and therefore radically decreased time for pigment/color matching and identification.