Knowledge Based Interpretation of Aerial Images and Maps Using a Digital Landscape Model as Partial Interpretation

The methods for the interpretation of aerial images and maps are usually different although both describe the same landscape. The presented work shows that regarding remote sensing data and maps as different kinds of sensors allows a similar approach for both in the domain of landscape interpretation. The prior knowledge about the landscape objects is represented explicitly by semantic nets. Based on the semantics of the network language a problem independent set of rules controls the scene interpretation. Important is that the scene analysis employs a partial interpretation derived from a Digital Landscape Model (ATKIS DLM 25/1). This partial interpretation is used to generate an initial scene description. Consecutively the scene description is verified in aerial images and maps. Interpretation proceeds iteratively mixing top-down and bottom-up strategies. This paper shows the representation of the knowledge in several semantic layers, the strategies of interpretation, and the methods to interprete aerial images and maps using a DLM as partial interpretation.