Abstract
The demand for 3D city modelling for various applications continue to grow with the capabilities of 3D city modelling. One of the uses of 3D city models is to facilitate 3D analysis which usually requires information regarding the topology of the objects within the city model. CityGML as the international standard for 3D city modelling maintains topological information with the use of a ‘topology-incidence’ where objects are referenced to each other with the condition that the objects share a common surface. This paper explains the extraction of topological information based on the data structure of the geometries in CityGML files and discusses the usability of the existing topology mechanism of CityGML. The topological information was extracted from the CityGML files using the hierarchical geometric structure of CityGML as a stand-in model to describe the topological properties of the object. The extracted information consisted of building surfaces which have been decomposed to 0D points with their respective identification and coordinates. Based on the extracted topological information and related literature, it was found that the topological information extracted from the geometric structure of CityGML was limited to the locality of the object in question and could not extend beyond the dimension of the primitive.
Highlights
In the past decade, the modelling of cities has continued to flourish moving from 2D drawings to complex 3D models and even venturing towards n-dimensions
The existing topological component of CityGML consisted of a simple topology-incidence which catered to the relations between stored objects that shared a common surface
This paper demonstrated an extraction of topological information from CityGML test data based on the existing data structure
Summary
The modelling of cities has continued to flourish moving from 2D drawings to complex 3D models and even venturing towards n-dimensions. In order to provide a standard for a complete 3D city model, CityGML focuses on five main aspects as shown in Figure 1 which are 3D geometry, semantics, scale or level-of-detail (LoD), appearance and topology. Analytical queries related to adjacency, intersection, connectivity, containment and disjointedness requires information that includes topological properties [12] These analyses provide a foundation for more complex and application-specific uses such as indoor navigation, simulations, and others. This defines how features are constructed based on the geometric model GML3 classes for geometric primitives from 0D to 3D [2]. The extraction of topological information was based on the hierarchical structure of the geometries from the test data CityGML files. The topological information extracted for both test data are shown in Figure 8 and Figure 9
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
