Semantics-Driven Remote Sensing Scene Understanding Framework for Grounded Spatio-Contextual Scene Descriptions

Earth Observation data possess tremendous potential in understanding the dynamics of our planet. We propose Semantics-driven Remote Sensing Scene Understanding (Sem-RSSU) framework for rendering comprehensive grounded spatio-contextual scene descriptions enhanced situational awareness. To minimize semantic gap remote-sensing-scene understanding, puts forward transformation scenes by using semantic-web technologies to Knowledge Graphs (RSS-KGs). The knowledge-graph representation has been formalized through development a Ontology (RSSO)—a core ontology an inclusive product. RSS-KGs are enriched both spatially and contextually, deductive reasoner, mining implicit relationships between land-cover classes scenes. Sem-RSSU, at its core, constitutes novel Ontology-driven Spatio-Contextual Triple Aggregation realization algorithms transform KGs render natural language descriptions. Considering significance informed decision-making from remote sensing during flood, we selected it as test scenario, demonstrate utility this framework. In that regard, contextual domain knowledge encompassing Flood (FSO) developed. Extensive experimental evaluations show promising results, further validating efficacy