ARCA (Automatic Recognition of Color for Archaeology): a Desktop Application for Munsell Estimation


Archaeologists are used to employing the Munsell Soil Charts on cultural heritage sites to identify colors of soils and retrieved artifacts. The standard practice of Munsell estimation exploits the Soil Charts by visual means. This procedure is error prone, time consuming and very subjective. To obtain an accurate estimation the process should be repeated multiple times and possibly by other users, since colors might not be perceived uniformly by di erent people. Hence, a method for objective and automatic Munsell estimation would be a valuable asset to the field of archaeology. In this work we present ARCA: Automatic Recognition of Color for Archaeology, a desktop application for Munsell estimation. The following pipeline for Munsell estimation aimed towards archaeologists has been proposed: image acquisition of specimens, manual sampling of the image in the ARCA desktop application, automatic Munsell estimation of the sampled points and creation of a sampling report. A dataset, called ARCA108, consisting of 22848 samples has been gathered, in an unconstrained environment, and evaluated with respect to the Munsell Soil Charts. Experimental results are reported to define the best configuration that should be used in the acquisition phase. Color tolerance values of the proposed framework are also reported.

In Proceedings of 19th International Conference on Image Analysis and Processing (ICIAP 2017).