EO Image Semantics and Interoperability Aspects in support of Content Retrieval and Data Mining

Kurzfassung

The data interoperability is very important for high resolution sensors with direct impact in retrieval and application development. The goal of the paper is to identify the semantic categories that exist in a large TerraSAR-X data-set that covers different urban areas over the world and to define taxonomies for these categories. Traditionally, image data-sets are built by individual groups and are being used to solve specific problems. For different domains (e.g., remote sensing, multimedia, etc.) the available data-sets may have a large variety in the number of retrieved categories.