Micro-computed tomography for discriminating between different forming techniques in ancient pottery: new segmentation method and pore distribution recognition

Micro-computed tomography is a valuable tool for studying ancient ceramics technology. Analysing pottery 3-D images is a challenging issue, the data being extremely noisy and heterogeneous. Quantitative criteria are introduced for the characterisation of a previously unrecognised pottery building method, the Spiralled Patchwork Technology (SPT). An analytical protocol has been implemented which applies to 3-D reconstructions of ceramic sherds and integrates automatic segmentation of porous systems and shape recognition using Hough transform. It enables discriminating between SPT and other techniques for vessels manufacture, and opens up many perspectives for independent characterisation of ancient technical gestures in ceramic technology.