Abstract
In this study a methodological framework for enhancing the detection and interpretation of archaeological features through near-surface geophysical surveys, in particular Ground Penetrating Radar (GPR) and magnetic gradiometry (MAG) is presented. Consequently a combined approach based on spatial analysis techniques and Artificial Intelligence, specifically Self-Organizing Maps (SOM), is devised to support automatic feature enhancement and recognition. This method has been experienced using GPR and gradiometric surveys, performed in a use case inside the archaeological area of Grumentum (Southern Italy). The results highlight the effectiveness of this approach in improving the readability of complex and heterogeneous geophysical datasets and increase the reliability of archaeological interpretations and in identifying subsurface remains and facilitating their interpretation. It is expected that the approach herein proposed can be promptly generalized and applied to other application fields.