An ontology controlled data fusion process for a query language
Tobias Horney, Erland Jungert, Martin Folkesson · 2003
Query languages designed for acquisition of data from multiple sensor data sources where the data generally are of heterogeneous type requires a number of internal functionality that is not available in traditional query languages. The required functionality can, for instance, be a method for multi-sensor data fusion, methods for query optimization and refinement. A further required technique relates to the problem of selecting sensors that eflciently can respond to the various queries without laying the responsibility of the sensor management on the users. The approach taken in this work has been to introduce an ontological knowledge-based system that can support selection of both sensors and object recognition algorithms as well as the control of the sensor data fusion process.