The generation of 3D city models demonstrates an increasing trend due to their usefulness. Many examples of application fields, such as emergency response and urban planning, can be mentioned in this context. With regard to their data acquisition methods, 3D city models reveals different characteristics. For instance, City Geographic Markup Language (CityGML) models provide different geometric, semantic and topological information, whereas digital surface models (DSM) focuses on data visualization, but lack in semantic information. Both models can benefit from each other in terms of a bi-directional information enrichment, especially with regard to the building bodies to be represented. Since no attention has yet been paid to the outstanding benefits, this work should deal with them. In the centre of this work, different distance measurements and algorithms are presented, discussed and implemented in a corresponding programming environment, on the basis of which building surfaces can be identified and segmented in a digital surface model using a CityGML LoD 2 model. In this context, the integration of geometrically high-resolution building facades of the DSM with the semantic city model shall be shown. Thereby, the building surface is to be saved as an alternative representation of the building surface. Various matching parameters are presented and implemented additionally which allow an evaluation of the segmented areas. After a successful integration of the geometrically high-resolution building facades, the presented parameters can also be used as a first clue in the semantic city model to determine whether the alternative representation of the original surface area is preferable. With regard to the solar potential analysis developed at the Chair of Geoinformatics, the results can be significantly improved with the presented work.     «

The generation of 3D city models demonstrates an increasing trend due to their usefulness. Many examples of application fields, such as emergency response and urban planning, can be mentioned in this context. With regard to their data acquisition methods, 3D city models reveals different characteristics. For instance, City Geographic Markup Language (CityGML) models provide different geometric, semantic and topological information, whereas digital surface models (DSM) focuses on data visualizati...     »