International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
tries to reduce damages with preventive protective mea-
sures before any high water event and effective disaster re-
action while fighting the high water. The disaster manage-
ment cycle consists of four phases illustrated in figure 5:
Mitigation is the ongoing effort to lessen the impact disas-
ters may have on people and property, preparedness means
the measures before the occurrence of an emergency. Sim-
ulations, training situations and information visualisation
can be done during the first two phases. Response refers to
the immediate actions during the emergency. It is impor-
tant to make adequate decisions during this time-critical
third phase. And finally, recovery is the process of going
back to normal.
Mitigation ——» Preparedness
| |
Recovery Response
Figure 5: Cycle of disaster management
Flood simulations basically rely on calculations based on
measurements of rain and topography. In literature sev-
eral different solutions for flood simulation are presented.
(UNESCO Training Module on GIS-Module H-2, 1999)
and (Casper, M. and Ihringer, J., 1998) e.g. describe tools
for simulating flood using a digital elevation model.
4.2 System, Data and Results
Results of high water simulations can be used as a basis for
planning disaster response measures. Visualisation of re-
sults is needed in order to help the local forces interpreting
them. The 3D visualisation with AR can be useful to ex-
pand the comprehension of the situation during the disaster
(fig. 6).
Figure 6: Virtual water surface nearby a real building, re-
sult of occlusion processing
In the city of Cologne authorities installed a mobile flood
protection wall, shown in figure 7, to protect the old town
against flooding from the river Rhine.
| Source: Sabine Mówes, An der Schutzmauer, http:
//www.stadt-koeln.de/hochwasserschutz/bilder/03808/
index.html
1052
Figure 7: Mobile wall in the city of Cologne !
Figure 8: 2D view of polygon representing an aera pro-
tected with mobile walls
In order to simulate the effect of such mobile walls, the
user can draw a polygon in a 2D view of the flood zone
which represents the virtual position of a mobile wall (fig.
8). In the 3D view of the area (fig. 9) the user can see the
water surface around the protected area.
This approach overlays virtual objects on a video stream
that can be viewed live on a laptop's screen. The GPS re-
ceiver and the inertial sensor provide necessary informa-
tion about position and orientation of the camera's view in
order to overlay the vision of the real world with virtual
buildings and water surface models.
The data used are building models, a digital elevation model
derived from airborn laser scanning and hydrological data
like the amount of rainfall in a region. The laser scan-
ner data is used in different ways. The last pulse data
provides elevation information of the surface with build-
ings but roughly without vegetation. A filter algorithm that
uses an image pyramid method extracts the digital terrain
model out of the laser scanner data (Coelho et al., 2002).
Flood water simulation models use hydrological data to-
gether with digital elevation models in order to estimate
the water surface at a given time. The simulations models
and the hydrological data are provided by the Institut für
Wasserwirtschaft und Kulturtechnik (IWK) at the Univer-
sity of Karlsruhe. The project scheme is shown in figure
10.
Internat
Figure
of the f
1
1
3
AA CT SM
The pr
velopin
usage a
(Breun
is subd
is devel
of spat
data ac
standar
velopin
visuali:
A use «
applica
tion an
bile rea
The
lung von
cation an
