Development of methods for satellite monitoring of cultural heritage sites: A preliminary study

The increasingly intensive use and modification of the landscape as a result of modern demands for efficient infrastructure and land use (agricultural production, mining, energy sources, leisure/tourism facilities) exerts growing pressure on areas and sites associated with our cultural heritage. The use of modern support technologies is imperative if such rapid changes are to be balanced against the sustainable management of this resource. This pilot project directly addresses these issues by initiating the development of a basis for a sustainable, up-to-date and cost-efficient decision-support methodology which relies upon satellite remote-sensing, mapping and monitoring of cultural heritage sites. This report concerns the automatic processing of satellite images in order to detect candidate cultural heritage sites. Using publicly available land use maps, the treatment can be confined to agricultural land. We then describe how images of these regions can be preprocessed in order to remove confusing ground structures. Segmentation (clustering) techniques are then used to group pixels into possible candidate sites, a supervised classification step is finally used to separate between interesting and uninteresting sites. Our principal conclusion is that automated processing of satellite images for detection of candidate sites is feasible, although with clear limitations. Based on our findings, it seems reasonable to investigate the use of satellite data for this purpose further.