Semantic modelling of Earth Observation remote sensing.

Abstract

Earth Observation (EO) based on Remote Sensing (RS) is becoming increasingly important, offering a robust technological framework for advanced applications in various domains like climate change, precision agriculture, smart urban planning, safety, and more. This leads to the development of data-driven software tools for enhanced data processing, analysis, and visualization, often with improved computing capabilities. However, the challenge of knowledge-driven approaches in remote sensing persists, despite their advantages in domain knowledge representation, data standardization, and semantic integration. Ontologies and semantic web technologies have proven successful in many fields, including Earth Observation, but there's significant room for improvement in remote sensing, particularly in addressing the unique nature and structure of satellite and airborne data products. This article addresses this need by proposing a semantic model for consolidating, integrating, reasoning, and linking data (including meta-data) in the context of satellite remote sensing products for EO. The authors have developed an OWL ontology and created an RDF repository to enable advanced SPARQL querying. While the proposal is intended for remote sensing data products in general, the primary focus is on the Sentinel 2 satellite mission from the Copernicus Programme of the European Space Agency (ESA). The article showcases four distinct use cases to illustrate the potential of the proposed semantic model in terms of ontology integration, federated querying, data analysis, and reasoning.