A main issue in the archaeological research is to deal with colors of soils and artefacts, especially pottery. Since, in many cases, color information are crucial for the interpretation of cultural products, to avoid risks of a too subjective autoptic recognition Munsell system is commonly used in archaeology.This method widely used in other fields, like geology and anthropology, is based on the matching between the real color and its standardized version on Munsell Charts. But it has significant limitations, due to the influence of cultural background, color sensibility and education, that can mislead archaeologists in their daily work.In this paper a semi-automatic method of color detection on selected regions of digital images of ancient pottery is presented. This tool, whose encouraging experimental results are widely discussed in the contribute, is aimed to prevent eventual subjective errors during color identification and to speed up the process of identification itself. In order to emphasize the relativity of Munsell system, a statistical analysis was carried out on a group of potsherds selected for this research, pointing out the range of different colors identified on a single specimen by different observers.The starting point of the experiment was to take digital pictures of specimens together with the Gretag-Macbeth Color Checker Chart, whose chromatic values have been objectively established. The digital image is processed with color balancing techniques aimed to restore the original value of Macbeth patches, in order to eliminate distortions coming from acquisition process. After the color correction, several regions of interest are selected via ‘point and click’ for the identification of surface color, the algorithm converts RGB values in Munsell data. The reliability of our tool is also verified comparing this chromatic values with the color specification of pottery sherds performed with a spectrocolometer using the CIELAB space to evaluate the differences.The results obtained and percentages of successful matching with Munsell color identification coming from the statistical analysis, seem to open new perspectives for the development of a full automatic system with a GUI interface aimed to facilitate significantly some aspects of the archaeologist's work.