GIS Approach for Mosaic Tesserae Recognition by Supervised Image Classification

This article describes the methodology used to automatically classify the constituent elements of a monochromatic mosaic from the Roman era. The classification involved the recognition of the tesserae (fragments of black and white materials for the floor ornamentation) of the mosaic using tools and algorithms implemented in Geographic Information System software (GIS). The analyses were performed using sample images expressly created by varying the colors on regular tiles in order to test the validity of the supervised classification in the ArcGIS environment. Subsequently, the same methodology was applied on mosaics from an ancient Roman archaeological site at Saltara; the results obtained from the classification process were compared to those obtained through visual reconstruction and manual vectorization. To assess the final quality and to validate the models obtained, a series of quality indicators were considered. The experimental results demonstrated that this approach offers effective classification in a short amount of time, making it a valuable tool for supporting work in the field of cultural heritage and, in particular, for the restoration and conservation activities of these artifacts

Archaeology · Computer vision · Contextual image classification · Geography · Mosaic · Computer Science · Digital Imaging for Blood Diseases · Artificial Intelligence