The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-4
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4, 483–490, 2018
https://doi.org/10.5194/isprs-archives-XLII-4-483-2018
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4, 483–490, 2018
https://doi.org/10.5194/isprs-archives-XLII-4-483-2018

  19 Sep 2018

19 Sep 2018

TOWARDS AN ONTOLOGY FOR THE STRUCTURING OF REMOTE SENSING OPERATIONS SHARED BY DIFFERENT PROCESSING CHAINS

G.-A. Nys1, J.-P. Kasprzyk1, P. Hallot2, and R. Billen1 G.-A. Nys et al.
  • 1Geomatics Unit, Department of Geography, ULiège - Place du 20 Août, 4000 Liège, Belgium
  • 2LNA-DIVA, Faculty of Architecture, ULiège - Boulevard de la Constitution, 41 4020 Liège, Belgium

Keywords: Ontology engineering, Remote sensing, NoSQL database, Graph mining, Knowledge discovery

Abstract. This paper proposes an ontology to structure and describe processing chains in the remote sensing field. These chains are made up of elementary elements (operations) organized in collections. The collection notion, including information about order and repeatability of the elements, is widely defined by using the relations between their constituting items and relations to the whole data store. Applications of the ontology are illustrated with web services provided by a platform for users and providers of processing chains. A graphical interface facilitates data integration in a RDF triple store. Thanks to the management of metadata (ISO19115-3), relevant information can be requested by intelligent search engines. Graph analysis, errors management and consistency rules are computed in order to gather coherent information from the different sources. Results of these analyses are then used by machine learning algorithms for new knowledge discovery.