Spatio-Semantic Comparison of Large 3D City Models in CityGML Using a Graph Database

The OGC open data model for the storage and exchange of virtual 3D city models City Geography Markup Language (CityGML) allows various syntactic ways to define a 3D city object. This on the one hand offers a high degree of flexibility in terms of creating new content-rich city models, but on the other hand complicates the automatic maintenance process of existing large CityGML documents. One often-stated example of such complications is the difficulty observed while attempting to detect possible thematic, geometrical as well as semantic deviations between two CityGML datasets of the same city. Existing studies have indicated that such problems can be solved using graph representations of CityGML documents. However, the question as how this concept can be realized still remains. Thus, this research provides an in-depth solution to this question in three main steps: (1) mapping two arbitrarily large-sized CityGML datasets efficiently onto graphs using a graph database (such as Neo4j), (2) matching mapped graphs based on concrete algorithms and attaching various types of EditOperations designed for updating the older CityGML dataset, and (3) executing attached EditOperations by converting them to transactions conforming to the Web Feature Service (WFS), the standard interface for updating geographical features across the web. The functionality and performance of the developed software is examined and demonstrated using the entire 3D city model of Berlin.