Ciortan, Irina MihaelaMarchioro, GiacomoDaffara, ClaudiaPintus, RuggeroGobbetti, EnricoGiachetti, AndreaSablatnig, Robert and Wimmer, Michael2018-11-112018-11-112018978-3-03868-057-42312-6124https://doi.org/10.2312/gch.20181352https://diglib.eg.org:443/handle/10.2312/gch20181352A critical and challenging aspect for the study of Cultural Heritage (CH) assets is related to the characterization of the materials that compose them and to the variation of these materials with time. In this paper, we exploit a realistic dataset of artificially aged metallic samples treated with different coatings commonly used for artworks' protection in order to evaluate different approaches to extract material features from high-resolution depth maps. In particular, we estimated, on microprofilometric surface acquisitions of the samples, performed at different aging steps, standard roughness descriptors used in materials science as well as classical and recent image texture descriptors. We analyzed the ability of the features to discriminate different aging steps and performed supervised classification tests showing the feasibility of a texture-based aging analysis and the effectiveness of coatings in reducing the surfaces' change with time.Computing methodologiesMachine learning approachesNeural networksApplied computingArts and humanitiesGeneral and referenceMetrics10.2312/gch.20181352147-154