Specific spectral bands for different land cover contexts to improve the efficiency of remote sensing archaeological prospection: The Arpi case study

Archaeological remains can be automatically extracted by remote sensing data when their spectral characteristics are discernible from their background; however, problems arise when extracting and classifying archaeological spectral features relative to subsurface structures, as archaeological remains do not have unique shape or spectral characteristics. Under the unidentifiable spectral condition, the main aim of this research is to verify the effectiveness of limited and specific spectral bands, retrieved by a hyperspectral remote sensing methodology described in [1], for the detection of surface anomalies related to archeological structures. The archaeological relevance of these extracted spectral anomalies is determined by an expert using traditional photo-interpretation methods. The results, compared with the archaeological knowledge of this area, stress the usefulness of the methodology to identify the specific spectral ranges to detect surface anomalies related to subsurface archaeological structures as a function of the surfacing land cover. The selected area for this study is located in the Arpi archaeological area (Italy) and the hyperspectral imagery used are Multispectral Infrared and Visible Imaging Spectrometer (MIVIS) airborne data. The paper highlights how airborne hyperspectral remote sensing can be an effective and cost-efficient tool to perform a preliminary analysis of subsurface remains in archaeological areas by prioritizing and localizing the sites where one can apply near surface geophysical and archaeological surveys. (C) 2009 Elsevier Masson SAS. All rights reserved.