An ontology-based approach to environmental signature detection : the vessel tracking case study

An Ontology-based approach to environmental signature detection with particular reference to navigating vessels is presented. The final attempt will be to check if environmental markers taken from MODIS sensors (hence at moderate spatial resolutions, from 250m up to 1 km) can be reversely associated with the event of a vessel transit or presence. To fulfill this requirement, although at an early stage, a top-down approach is proposed. From the detection system designer's point of view the first activity is the definition of the problem ontology. Following there is the phase where string patterns recognized by the ontology are searched through the available dataset. The employed search patterns are suggested from the tracks provided by AIS data source. Preliminary experiments seem to endorse the proposed approach.