Classificare in modo automatico i frammenti ceramici: ArchAIDE, un’utopia realizzata

Pottery classification is a time-consuming activity because it is based on the comparison between the shape or decoration of a potsherd and the shape or decoration of vessels listed in paper catalogues. Traditionally, it requires an extensive and heterogeneous bibliography, often accessible only in libraries, and rarely available in warehouses during fieldwork and post-excavation analysis. The ArchAIDE project, funded by the European Union’s Horizon 2020 - Research and Innovation Programme, developed an automated recognition system for potsherds, using the latest image recognition technology. Using a potsherd photo, the program compares the form and decoration with its internal database against a trained neural network and formulates one or more possible identifications (max 5) for the archaeologist to consider.