nbul 2004 
fi 2 hz 10m 
:—e— h=30m 
h= 100 m 
——— {oj| 1:2000 
—— {|| 1:5000 
toll 1:10000 
  
resence of 
ir satellite 
yroach and 
neglected, 
nce on an 
ich for the 
age (GSD 
EM (50 m 
3D digital 
cene refers 
rted in its 
ion that is 
jalues that 
s that were 
red” after 
is resolved 
juired from 
»wn. These 
ther sensor 
ve this lack 
eas with a 
in figure 6 
s way, the 
easier for 
data. 
  
  
'espondence 
t) and the 
1. 
  
Figure 6 — An example of Eros Al (1.9 m) orthoprojection. 
Duplication of the radiometric information (on the left) and 
masking of the same hidden area (on the right). 
  
  
Figure 7 — An example of QuickBird (0.61m) orthoprojection. 
Duplication of the radiometric information (upper) and masking 
of the same hidden area (lower). 
5. CONCLUSIONS 
The here proposed case clearly shows the problems connected 
to the use of a precise approach for the orthoprojection of 
satellite images and confirms that altimetric height data of the 
buildings cannot be neglected, above all in urban areas. An 
analysis of the application fields of such approach is basically 
confirmed by the metadata supplied with the image. The mean 
view angle for the test reported in the previous section is about 
10°. The entity of the displacements due to the mean height of 
the buildings, on the basis of the proposed graphs, is therefore 
not negligible for any orthophoto scale. Only the use of a 
DDEM can allow a correct resolution of the problem to be 
made. The suitability of the orthophoto at a certain scale in this 
way remains conditioned by only the quality of the projection 
model. The masking of the hidden areas makes the 
interpretation of orthoprojected satellite images easier, 
especially for unskilled users. 
References: 
Bello, M. G. , 1992, Enhanced training algorithms, and 
integrated — training/architecture selection for multilayer 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part Bl. Istanbul 2004 
perceptron networks, /EEE Trans. on Neural Networks, 3,864- 
875. 
Boccardo, P., Borgogno Mondino, E., Giulio Tonolo, F., 2003, 
High resolution satellite images position accuracy tests, 
IGARSS 2003, Toulouse (proceedings on CD). 
Dequal S., Lingua A., 2002, L’ortofoto di precisione del 
Comune di Torino, Atti VI conferenza nazionale ASITA 2002 
Dowman, I., Tao, V., 2002, An update on the use of rational 
functions for photogrammetric restituition, /SPRS, Vol. 7, N°3, 
September 2002, pp. 26-29 
Giulio Tonolo, F., Poli, D., 2003, Georeferencing of EROS-A1 
high resolution images with rigorous and rational function 
model, /SPRS , October 2003, Hannover, Germany. 
Howard Demuth, Mark Beale, MATLAB Neural Network 
Toolbox User's Guide, The Mathworks, Version 4, pp.5-8,5-30. 
Lingua, A., Borgogno Mondino, E., 2002, High Resolution 
Satellite Images Orthoprojection Using Dense DEM, Spie 2002 
proceedings, Creete 
Open GIS, 1999, Abstract specification. Topic 7: The earth 
imagery case, Open GIS Consortium, March 1999 
Rumelhart, D. E., Hinton, G. E. and Williams, R. J, Learning 
internal representations by error propagation, /n Parallel 
Distributed Processing, vol 1, pp. 318-362. Cambridge, MA: 
MIT Press. 
Tao, C., Hu, Y., 2001, A comprensive study of the Rational 
Function Model for Photogrammetric processing, 
Photogrammetric engineering & Remote Sensing, December 
2001, pp.1347-1357 
Toutin, T., 2004, Geometric processing of remote sensing 
images: models, algorithms and methods, International Journal 
of Remote Sensing, 24, 2003 (to be published) 
Acknowledgements 
The different satellite data that were used for this study were 
acquired as part of a national research project, co-financed by 
the Ministry of the University and Research, entitled “The use 
of satellite images for environmental analysis" (COFIN 2001).