Description du projet
Lutter contre le pillage des sites archéologiques
La recrudescence des fouilles illégales de sites archéologiques et le trafic d’antiquités qui s’ensuit est une industrie de plusieurs milliards d’euros qui s’est considérablement développée en lien, par exemple, avec les troubles au Moyen-Orient. De plus, le pillage des sites archéologiques s’est exacerbé pendant la pandémie de COVID-19. Le projet OPTIMAL, financé par l’UE, développera un moyen automatique d’identifier les activités de pillage. Il adoptera une approche d’apprentissage automatique basée sur un transport optimal pour détecter automatiquement les pillages (passés et présents) directement sur des séries temporelles lidar aéroportées. Le projet rassemblera également un ensemble de données lidar qu’il rendra accessibles au public et qui seront utilisées gratuitement pour identifier les activités illégales.
Objectif
"Illegal excavation of archaeological sites aimed at collecting historical material culture (""looting"") is a pressing problem on a global scale. The global upsurge of in the illegal excavation of cultural heritage sites (e.g. in connection to turmoils in Middle East or due to the impossibility of monitoring inaccessible areas, like in South America) and the subsequent trafficking of antiquities, exacerbated by the Covid lockdown, calls for the timely development of automatic means for identifying looting activities. The OPTIMAL (OPtimal Transport for Identifying Marauder Activities on LiDAR) project aims to tackle this challenge by developing an efficient and principled Machine Learning (ML) approach based on Optimal Transport to automatically detect looting (past and present) directly on airborne Light Detection And Ranging (LiDAR) point cloud time-series. OPTIMAL proposes, for the first time, the use of LiDAR for monitoring and assessing the damages of looting based on LiDAR’s unique ability to penetrate forest canopies and enabling to see a range of looting-related features under the canopy (e.g. shape and depth of the lootings pits) that otherwise would remain hidden due to vegetation covers. OPTIMAL will create and make publicly available the first multi-temporal LiDAR dataset for illegal activities’ identification to foster the interest of MLs researchers in developing new methods to tackle challenges in landscape archaeology and to evaluate the developed ML approach. Results of this interdisciplinary research will be widely disseminated within Cultural Heritage, Remote Sensing and Machine Learning communities and to others that can exploit OPTIMAL’s results. A communication strategy will be designed to ignite enthusiasm for technological advancements for the protection of our Heritage."
Champ scientifique
- natural sciencescomputer and information sciencesartificial intelligencecomputer vision
- engineering and technologyenvironmental engineeringremote sensing
- humanitieshistory and archaeologyarchaeology
- natural sciencesmathematicsapplied mathematicsstatistics and probability
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Mots‑clés
Programme(s)
Régime de financement
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinateur
16163 Genova
Italie