Optimization of use-wear detection and characterization on stone tool surfaces.

Debates and doubt around the interpretation of use-wear on stone tools called for the development of quantitative analysis of surfaces to complement the qualitative description of traces. Recently, a growing number of studies showed that prehistoric activities can be discriminated thanks to quantitative characterization of stone tools surface alteration due to use. However, stone tool surfaces are microscopically very heterogeneous and the calculated parameters may highly vary depending on the areas selected for measurement. Indeed, it may be impacted by the effects from the raw material topography and not from the altered zones only, if non-altered part of the surface is included in the measurement. We propose here to discuss this issue and present a workflow involving the use of masks to separate worn and unworn parts of the surface. Our results show that this step of extraction, together with suitable filtering, could have a high impact on the optimization of the detection and thus characterization of use traces. This represents the basis for future automatic routines allowing the detection, extraction and characterization of wear on stone tools.