New tools dealing with old issues: from graphical elements to semantic objects

Abstract

For decades, archaeologists have created digital plans in Computer-Aided Design (CAD) or Geographic Information Systems (GIS), as needed for research. Digitising old archaeological maps could be achieved in two ways: by scanning or vectorising the spatial information. Cartographic technologies have made it possible to collect, store, query and visualise spatial data with increasing efficiency, thereby transforming data-recording and data-presentation. Even if the maps recorded in CAD or GIS share a common mode of numerical representation organised in distinct layers, the difference between the two systems lies in the link between the geometry and the alphanumeric information attached to it. CAD produces complex graphical maps but ignores the non-spatial attributes associated with the graphic entities. Therefore, reusing CAD data in GIS is not an easy task without a reorganisation of information based on the construction of semantic objects corresponding to real features. Many tools have been developed with the aim of automatically converting CAD files into GIS formats. Despite great efforts, to date the results are scarcely significant. And the challenge is still open. This is because, without the direct intervention of the researcher, the transition from CAD to GIS risks generating incorrect data. Starting from these premisses, this paper deals with the discussion of a case study focused on the organisation and conversion of cartographical resources related to the Islamic site of al-Balīd in the Sultanate of Oman. The aim is to provide a topographical frame to anchor existing archaeological structures and propose a workflow to convert CAD drawings into GIS files.