Structural Interpretation of High Resolution Images

With increasing resolution of remote sensing data more and more details and structures of landscape objects are observable. An interpretation of such images requires the use of improved methods for object extraction. In this paper several approaches on this topic are presented. The basis of all approaches is an explicit knowledge representation with semantic nets. For different applications concept nets are shown and the strategy to use them. Furthermore a methodology to adapt the concept nets automatically to other resolutions is described and strategies to extend the methods to a multitemporal interpretation.