nbul 2004
Ose issues
works on
which has
und truth
nply want
t possible
1e need to
But there
) measure
e made to
of detail is
“buildings
wmber of
issue, as it
[dings in a
ss to the
ientate the
'anizations
1uch more
er SUCCESS;
rmance on
d feature
| systems.
small but
for further
ds. inIECT
with new
Not only
concept of
IS, as there
due to low
traction of
ow highly
cels with a
'eased the
mated 2D
tested with
ixel size of
| m ground
measuring
ipproach is
to be an
re modules
dapting the
lications is
1JECT has
pecially in
10 need for
ofessionals
an easy-to-
ction tasks
lecade with
1 to be the
Also non-
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
photogrammetrists can measure buildings, bridges,
streets and parcels, if provided with easy-to-use
interfaces.
e Very few research efforts have found their way to a
commercial system. This is surprising, but most
efforts have been put either to unrealistic full
automation or to improve GUIs, input-output
functionality and not to extend basic kernel software
for automation.
e The lack of standards for object modelling still
hinders a wider range of applications and the
exchange and re-usage of 3D data, but emerging
standards like GML3 could fill this gap.
e. There is still no common language or model for a
building, but initiatives like SIG3D allow deep inside
and provide excellent ideas in this context.
e There is an increasing tendency to rely on unrealistic
types and amount of a priori information for real
work applications. If the approach is too limited the
usage for nation wide or world wide application is
limited as well. There can be observed a new
promising trend to use image analysis to support the
feature extraction in airborne laser scanning data.
* [tis still completely unclear, how updating of 3D city
models is to be performed. Some first ideas and
approaches have been presented, but real world
experience is not yet gathered on a broader scale.
e There are some few empirical investigations on
international level to test and compare feature
extraction software in an independent way.
Organizations like ISPRS and EuroSDR should be
supported much more to conduct those tests and to be
able to disseminate the results. With additional help
those organizations larger empirical tests can be
performed, which have shown in the past the
tremendous impact like in automated AT and direct
georeferencing.
There is on-going rescarch and development in automated
feature extraction. New ideas are presented and it seems that a
new step of development has been reached to combine new
resources and groups to work together towards the common
goal of a reliable automate feature extraction from aerial
imagery and airborne laser scanning data.
REFERENCES
Arefi, H., Hahn, M., Lindenberger, J., 2003. LIDAR data
classification with remote sensing tools. Joint ISPRS
Commission IV Workshop "Challenges in Geospatial Analysis,
Integration and Visualization II", Stuttgart, September 8- 9,
2003.
Baltsavias, E., Grün, A., Gool, Luc van. (Eds), 2001. In:
Proceedings Third International Workshop on Automatic
Extraction of Man-Made Objects from Aerial and Space
Images, Ascona, Switzerland, 10-15 June 2001.
Swets&Zeitlinger Publishers/Balkema, The Netherlands.
CyberCity AG, 2004. http://www.cybercity.tv
EuroSDR, 2004. http:/Avww.eurosdr.org
Gülch, E., 2000. Digital Systems for automated cartographic
feature extraction. In: International Archives of the
Photogrammetry, Remote Sensing and Spatial Information
Sciences, Vol. XXXIII, Part B2, pp. 241-255.
427
Gülch, E., Müller, H., Läbe, T., 2000. Semi-automatische
Verfahren in der photogrammetrischen Objekterfassung.
Photogrammetrie Fernerkundung Geoinformation (PFG),
3/2000, pp. 199-209.
Gülch, E. Müller, H., 2001. New applications of semi-
automatic building extraction. In: Proceedings Third
International Workshop on Automatic Extraction of Man-Made
Objects from Aerial and Space Images. Ascona, Switzerland,
10-15 June 2001. Swets&Zeitlinger Publishers/Balkema, The
Netherlands.
Inpho GmbH, 2004. http://www.inpho.de.
ISPRS WGIII/8, 2004 "Reliability and Performance of
Algorithms":
http://www.isprs.org/technical commissions/wgtc 3.html£wg3/
8.
Kolbe, T.H., Gróger, G., 2003. Towards unified 3D city
models. Proceedings Joint ISPRS Commission IV Workshop
"Challenges in Geospatial Analysis, Integration and
Visualization II.. Stuttgart, September 8-9, 2003.
Ohlhof, T., Gülch, E., Müller, H., Wiedemann, C., Torre, M.
2000.. In: /nternational Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences, ISPRS
Congress, Istanbul, July 2004.
SIG3D, 2004. Special interest
http://www.ikg.unibonn.de/sig3
oup 3D of GDI forum NRW
C. go
Ulm, K., 2002. Improved 3D city modeling with CyberCity-
Modeler (CC-Modeler™) using aerial, satellite imagery and
laserscanner data. In: /nfernational Archives of Photo-
grammetry, Remote Sensing and Spatial Information Sciences,
Vol. XXXIV, Part 5/W10.
Sithole, G., Vosselman, G., 2003. Comparison of filtering
algorithms. Proceedings of the ISPRS working group [11/3
workshop 3-D reconstruction from airborne laserscanner and
InSAR data. Dresden, Germany 8-10 October 2003.
ACKNOWLEDGEMENTS
The support of INPHO GmbH, Stuttgart, in providing access to
the inJECT 1.9 pre release software is gratefully acknowledged.
