gital
eng
JSA,
ith a
S of
JSA,
hical
^ the
. 30,
MERGING DTM AND CAD DATA FOR 3D MODELLING PURPOSES
OF URBAN AREAS
Siyka Zlatanova Nedkova*, Michael Gruber** and Michael Kofler**
"International Institute for Aerospace Survey and Earth Sciences
7500 AA Enschede, The Nederlands
**Institute for Computer Graphics
Graz University of Technology
Münzgrabenstr. 11, A-8010 Graz, Austria
ISPRS WG IV/1
Key Words: 3D Urban Database, 3D GIS, 3D CAD, DTM, 3D Modelling
Abstract
We present in this paper a method for merging terrain data from DTM and CAD data. The method
uses data from DTM which are reorganized to a set of 3D polygon in a TIN representation. The CAD
objects, i.e. buildings and surface features, are fused with DTM after applying a special procedure.
The used techniques ensure the consistency of the data model in terms of avoiding undesirable gaps
and occlusions between the DTM and CAD features. A partial simplification of the buildings facilitates
the process of texturing and manipulating the data.
In the next steps we will strive fore possible automation of the entire procedure. This shall support a
more or less autonomous and cheap manner for building 3D photo realistic models of urban aereas.
1.Introduction.
Three dimensional (3D) digital models became a
very important issue in the last few decades due to
the increasing demand for a realistic presentation of
the real world. A 3D model gives better presentation
and facilitates the processes of planning, controlling
and decision making especially in the urban areas.
Furthermore modern graphics workstations which are
available on the market allow already to handle and
visualize large amounts of data. Since there is need
for 3D models and sufficient hardware, the task of
the researchers now is to develop technologies for
automatic construction of 3D models using different
data sources.
Speaking about 3D models, it is always of great
importance to clarify what is the purpose of the
model, or in other words, what is the area of
expected applications. The organization of the data
in the model can be totally different if one needs
only a realistic view of the real world with additional
tools for flying trough, etc., compare to needs, for
example, spatial analysis to be provided. Another
aspect which cannot be forgotten is the accuracy of
the data. Depending on the application, it may
appear that there is no need for very accurate data
and there is a possibility for simplification of the
data sets. This, without any doubts, will facilitate the
process of data handling. The goal of this paper is
to contribute a part to the technology for building a
3D city model, namely the merging of terrain data
and objects for photo realistic visualization and
manipulation, which will supply users with:
311
e photorealistic presentation with possibilities for
navigation through the model;
e abilities to model, design and analyze the data of
a city scape, e.g. arrange buildings and man-
made objects, allow queries, etc.
The investigations are focussed into two important
tasks. Broadly speaking, the first task is to
construct geometry and second one is to texture
the model. The aim of the approach is to make use
of various data sets comming from different
Sources, e.g. 2D maps, aereal images, Digital
Terrain Model (DTM), etc. The joint efforts of the
researchers at the International Institute for
Aerospace Survey and Earth Sciences(ITC),
Enschede and the Institute for Computer Graphics
(ICG), Graz will result finally in a automatic or semi-
automatic procedure for creating, storing,
manipulating and analyzing 3D city models.
2. Data sets
Various methods for constructing 3D models are
object of studies and different approaches
concerning data extraction and data merging are
developed (see [2,5,7,11]). The way of constructing
the 3D model depends on the data sets which are
used. For example, the work carried out in ITC, is on
the construction of models only from aerophoto
images, applying 3D digitizing for gathering the
required information. Existing DTM, 2D CAD data
and aerophoto images are the source data used at
the ICG.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996