Résumé
Satellite imagery can be exploited for any number of thematic analyses for Earth observation purposes. Characterization activities using remotely acquired data are currently made complicated by different limitations relating to, as an example, the meaningful mapping between multi-sensor data or the adding of the geospatial context to satellite information. We argue that describing satellite images through a metadata application profile may leverage capabilities to promote easy-to-use data for further in-depth thematic analysis. Accordingly, an application profile conforming to the Dublin Core application profile (DCAP) guidelines and dedicated to Earth observations(EO) is being developed. More specifically, we discuss RDF-compliant machine-processable aspects of the EO application profile (EOAP) in terms of the DCMI Description Set Profile(DSP) model. Additionally, a methodological approach to represent a DSP model using UML profiling activities is proposed.