ated for 
e image 
ates the 
(Fig. 2), 
ympared 
e, taken 
provides 
uce. 
on Gra- 
) min for 
e digital 
objects 
0 - 2050 
ck pixels 
rt of the 
ons as a 
' surface 
/hich are 
like the 
  
Fig. 8: Comparable image from a distant point of view, 
taken with a Rolleiflex 6006 metric, 
focal length 250 mm 
fence around the platforms on the towers. This objects 
are displayed at wrong positions. Problems like this could 
only be resolved by refining the DSM. 
4. POST PROCESSING 
The generation of orthoimages may be carried out with 
more than one distorted image. The necessary informa- 
tion to fill occluded areas may be extracted from a second 
image. But in this case differences in the radiometric 
information (brightness, contrast etc.) of the two ortho- 
images have to be considered. This may be achieved by 
radiometric mosaicking developed for mapping purposes 
(Scholten 1996). 
An additional post processing technique may be the geo- 
metric comparison of two orthoimages, interactively or 
automatically by matching techniques. If the images have 
been exposed at the same day, local geometric diffe- 
rences of the two orthoimages can be explained as errors 
in the DSM. If the two images are from different exposure 
dates the differences may be the results of changes of 
609 
the object. This approach could be used for automatic 
damage analysis. 
5. CONCLUSION 
At present the generation of the CAD model is the most 
tedious task in the production of orthoimages. New de- 
velopments, such as semi-automatic or automatic digital 
restitution techniques, will accelerate this process and 
provide the necessary data fast and cheap (Streilein 
1994). New users of orthoimages can be anticipated in the 
world of computer graphics. Orthoimages are well suited 
for texture mapping in Virtual Reality Models. 
Architectural photogrammetric data may be used for the 
documentation of historical sites and monuments, for the 
determination of damages, or for civil and building engi- 
neering purposes. The requirements for such data are 
undisputed (Waldháusel 1992) and permanently con- 
firmed by the news about destroyed buildings we see 
every day on TV. 
The rectification of architectural imagery has a very long 
tradition. But thanks to the digital image processing also a 
successful future, and new users can be anticipated. 
REFERENCES 
Li, R., 1993. Generation of geometric representations of 
3D objects in CAD/CAM by digital photogrammetry. ISPRS 
Journal of Photogrammetry and Remote Sensing, 48(5): 
pp. 2-11. 
Marten, W., L. Mauelshagen & R. Pallaske 1994. Digital 
orthoimage-system for architecture representation. 
ISPRS Journal of Photogrammetry and Remote Sensing, 
49(5): pp. 16-22. 
Mayr, W. & Ch. Heipke 1988. A Contribution to Digital 
Orthophoto Generation. In: International Archives of 
Photogrammetry and Remote Sensing, Kyoto, Japan, Vol. 
XXVII, Part. B11, pp. 430-439. 
Scholten, F., 1996. Automated Generation of Coloured 
Orthoimages and Image Mosaics Using HRSC and 
WAOSS Data of the Mars 96 Mission. In: International 
Archives of Photogrammetry and Remote Sensing, Wien, 
Austria, Vol. XXXI, Part. B3. 
Streilein, A., 1994. Towards automation in architectural 
photogrammetry: CAD-based 3D-feature extraction. 
ISPRS Journal of Photogrammetry and Remote Sensing, 
49(5): pp. 4-15. 
Waldhäusl, P., 1992. Defining the future of Architectural 
Photogrammetry. In: International Archives of Photo- 
grammetry and Remote Sensing, Washington, USA, Vol. 
XXIX, Part. B5, pp. 767-770. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996