Automatic clustering of Celtic coins based on 3D point cloud pattern analysis
This addresses a tedious and expert-dependent task in archaeology by automating die clustering for Celtic coins.
The paper tackles the problem of automatically clustering Celtic coins struck by the same die, which is currently done manually by experts, and achieves clustering quality equivalent to expert work on a dataset of 90 coins.
The recognition and clustering of coins which have been struck by the same die is of interest for archeological studies. Nowadays, this work can only be performed by experts and is very tedious. In this paper, we propose a method to automatically cluster dies, based on 3D scans of coins. It is based on three steps: registration, comparison and graph-based clustering. Experimental results on 90 coins coming from a Celtic treasury from the II-Ith century BC show a clustering quality equivalent to expert's work.