International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
5.2 Topologic model
After the use of the first tool, the frontage faces and the rooftop
contour lines are identified and structured into a database
working on MySQL and EasyPHP. (Figure 13).
Only the structured four-points-faces are recorded in the
database, the initial model is recorded as graphical design and
can be displayed on demand.
Ti
Figure 13. Model in hidden line rendering mode extracted from
database
5.3 Images
The photography of each frontage (Figure 14) can be used into
the second tools to enrich geometrically and thematically the
model. The Figure 15 shows the model geometrically enriched.
Figure 15 : Frontages geometrically enriched
5.4 Texture images
With the orientation parameters defined in the second step, the
third tools realize the computing of texture-images of the
frontage associated to each part of the frontage (Figure 16).
rt
WEE TTT EET
à 3» voy P
552
5.5 Final 3D model
Figure 16 : Frontages geometrically enriched and textured
6. CONCLUSION
The tools described in this paper allow to overpass the first
level of the description of a building. In this first level, the
frontage surfaces are considered as planar vertical faces. From
now, the frontage have acquired a new dimension and a higher
detailing grade because it is now possible to decompose a
frontage into stories (horizontal division) and each story into
different parts according to characteritics of interest for the end
user. Besides, the geometric definition of the frontage is
refined. The different architectural entities that constitute the
frontage (balconie, porch, recess, etc) are identified and can be
quantified with a depth or a shift value from the frontage plane.
In a geometric point of view, the obtained definition level is
more in accordance with the needs of a 3D-GIS and its
applications.
At this time, we did not want to replace the old frontage faces
by the new computed frontage parts. This will not prevent from
3D analyzes.
The use of photographies from the frontage of the modeled
buildings seems to be a relevant way to access to 3D
informations and allows to represent and to use them into
system we can qualify as 3D-systems.
A lot of improvements can be added to the in this paper three
presented groups of tools.
The first tools could be able to define the whole topologic
relations in more complex 3D models.
The second tools could allow the definition of whatever
divisions of the frontage by using a polyline for the division
trace. This would offer the possibilities to describe and
represent phenomenon with a more complex geometry.
The third tools should be able to create rectified textures for
overhanging objects. They also have to integrate algorithms for
resampling and for creating texture-images for complex
divisions.
References from Journals:
Koehl, M. [2003] Modeéles et méthodologies de saisie pour
maquettes numériques 3D, XYZ n? 97 2003, pp.31-37.
References from Other Literature:
Koehl, M. [1999], Modélisation Géométrique et Sémantique en
milieu urbain. Intégration dans un Systéme d'Information
Topographique tridimensionnel, PAD Thesis, 242 pages.
ENSAIS Strasbourg, France.
Gaiotti, V., [2003], Définition et conception d'outils de
modélisation et d'intégration de données dans un S.I.T. 3D
urbain, Dipl-Thesis, 64 pages. INSA Strasbourg, France.
Koehl, M. et Grussenmeyer, P. [1998], 3D data acquisition and
modelling in a Topographic Information System, /APRS, Vol.
32, Part 4, pp.314-320, Stuttgart, Germany.
Molenaar, M. [1990], A formal data structure for three
dimensional vector maps. First European Conference on GIS
(EGIS'90), Vol.2, pp.770-781. Amsterdam, The Netherlands.
fi "yi LI
OY ae - PA
es
