Geoscience Reference
In-Depth Information
buildings in CityGML. The proposed method deals with segmentation of a 3D
building based on its semantic value and surface characteristics, fitted by one of the
predefined primitives. For future work, the segmentation method will be imple-
mented as part of the change detection module that can detect any changes on the
3D buildings, store and retrieve semantic information of the changed structure.
Keywords 3D city models
3D model segmentation
CityGML
1 Introduction
3D city model is a digital representation of a city or urban area that was previously
generated only for visualization purposes to complement the results from 2D
geospatial related analysis. The involvement of major industry players, private and
government organizations as well as participation from the mass public has pro-
pelled the popularity and interests in 3D city modelling. Now, it has become the
new trend in building and urban management while modelling 3D objects are
getting easier with the emergence of user-friendly tools for 3D modelling available
in the market. Currently, many countries around the world have been generating
virtual 3D representation of their major cities. The growing interest in improving
the usability of 3D city models has resulted in the development of various tools for
analysis. Today, 3D city models are generated for various purposes such as for
tourism, location-based services, disaster management and urban planning.
Over the years, there are quite a number of researches have been carried out to
improve the usability of 3D city models in geospatial related analyses such as
noise mapping (Kolbe
2009
), pollution control, disaster management, navigational
and network analysis.
The core entity in 3D city model is buildings since it is the entity that defines a
city. However, other entities such as terrain, landscape, street furniture and vege-
tation also play important roles in representing the city as a whole. Since new
developments, constructions and renovations are always happening all around the
city, the buildings will change over time and their structures become more complex.
However, complex building structures are actually a combination of several basic 3D
primitives such as cubes, cuboids, spheres and cylinders. These combinations can be
easily recognized by human brains naturally but computers do not have the same
skill, hence the need for 3D segmentation for 3D city model objects. Segmentation is
a method to breakdown a complex object into simpler parts and in 3D city models,
each segmented parts usually represent a meaningful part of the building.
This chapter mainly focuses on the 3D segmentation technique for 3D city
model objects (3D buildings) that can be deployed for several analyses in GIS such
as change detection and automatic updating. In
Sect. 2
, this chapter will discuss
about past researches related to CityGML and 3D city models, the building gen-
eration methods, and 3D segmentation on 3D buildings. Meanwhile,
Sect. 3
will
