Ontology-Based Geospatial Approaches for Semantic Awareness in Earth Observation Systems

Abstract

AbstractCurrent work towards making earth observation systems semantically aware attempts to improve user experience by allowing more flexibility in the way that users interact with earth observation systems. Such improvements may occur directly by making data discovery more semantically-flexible, and indirectly in providing intelligent functionality that removes some of the load from users in interpretation of data and processes. Semantic awareness in earth observation systems may be addressed from four different angles: semantics and information modelling; semantic data management; semantic data discovery and semantic data processing. Each of these areas is the subject of ongoing and developing research in the broader geospatial community, has been applied in a number of different situations and systems, and presents particular challenges for earth observation systems. The Global Earth Observation System of Systems (GEOSS) is a large, global, heterogeneous earth observation system and provides a case study of the use of different methods for achieving semantic awareness in each of these four areas. Furthermore, an example architecture for an earth observation system that involves multiple aligned ontologies illustrates the challenges posed by real world, heterogeneous systems. In combination, the review of related work, applications and challenges in each of the four areas, together with the GEOSS case study and example architecture provide an indication of the state of the art in semantic research as it applies to earth observation system. Furthermore, this summary provides a hint towards the future for semantics in earth observation systems and the need for additional work in this area.KeywordsGeographic Information SystemResource Description FrameworkDomain OntologyResource DiscoveryCore OntologyThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.