Integrated Spatio-temporal Data for Earth Observation: A RDF dataset of Territorial Units with their Land Cover

Ba-Huy Tran (1) , Nathalie Aussenac-Gilles (1, 2) , Catherine Comparot (1, 3) , Cassia Trojahn dos Santos (1, 3)

1 IRIT-MELODI - MEthodes et ingénierie des Langues, des Ontologies et du DIscours
2 CNRS - Centre National de la Recherche Scientifique
3 UT2J - Université Toulouse - Jean Jaurès

In this paper, we propose an ontology-based approach that relies on data spatial and temporal dimensions to semantically integrate open datasets together with Earth Observation data. The resulting dataset provides rich contextual information about EO and makes it possible to search EO images according to this contextual information. We illustrate the approach on three French datasets: administrative units, land registries and their land cover dataset. The resulting dataset is a semantic database and it is exposed through a semantic search interface or can be accessed via a SPARQL endpoint.

Mots clés: Image metadata; Knowledge representation; Ontologies; Semantic data integration; Semantic search

Domaines: Informatique et langage [cs.CL]

Ba-Huy Tran, Nathalie Aussenac-Gilles, Catherine Comparot, Cassia Trojahn dos Santos. Integrated Spatio-temporal Data for Earth Observation: A RDF dataset of Territorial Units with their Land Cover. 6ème Atelier Gestion et Analyse de données Spatiales et Temporelles (GAST @ EGC 2020), Groupe de Travail GAST, Jan 2020, Bruxelles, Belgium. pp.1-9. ⟨hal-02947790⟩