In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
Figure 8. Oblique view aerial image of the DB-Tower 
Figure 7. Simulated double bounce reflections overlaid with PS 
Therefore, it is reasonable to assume that the PS are caused by 
structures on this roof and not by scatterers located at the upper 
part of the facade. This assumption is verified by the estimated 
height of the PS (see 
Figure 2). 
It is furthermore possible to assess PS located on double bounce 
lines induced by large scale building features (e.g. curb-to-wall 
scattering). These are displayed in Figure 7. The PS heights 
arranged in five meter classes are overlaid to simplify the 
discrimination of PS on and off the double bounce lines. It is 
apparent, that the PS densities on the double bounce lines are 
not always as high as one may expect due to the strong 
backscattering. An example is marked by the red box in Figure 
7, where there are hardly any PS resident (PS coloured in light 
red). In contrast to that, there are lots of PS on the double 
bounce feature marked by the green box. The reason for that 
is to be investigated. 
As can be seen from the oblique view aerial image in Figure 6 
the virtue of the respective façades differ in terms of material 
and geometry. 
Finally it is remarkable, that there are almost no PS at the 
façades visible in Figure 6, which is apparent from 
Figure 2 and Figure 5. This is surprising, since there are usually 
plenty of PS at building fronts oriented towards the sensor. It is 
conceivable, that this is due to the texture of the façades. 
However, there are at least some PS located on the skyscraper at 
the top of the SAR image (see Figure 1 and 
Figure 2), whose façade looks quite similar as can be seen from 
the oblique view aerial image displayed in Figure 8. 
4. CONCLUSIONS 
In theory a SAR simulator can assist the PSI analysis by helping 
to determine the exact location of the PS. This is true, if very 
precise and detailed 3D building models are available. Although 
3D building models are available for many urban areas, they 
seldom contain the geometry of the façades. The façade is 
mostly represented with an image texture, which is feasible for 
visualization purposes, but not for SAR simulation. To simulate 
the façade reflection, a 3D model containing the façade 
geometry is necessary. 
However, the simulation still proofed to be useful. We could 
demonstrate that the double bouncing between building and 
ground does not cause many PS. Most PS in our test scene are 
caused by scatterers located on the building roofs. 
Without highly detailed building models, SAR simulation can 
only provide very rudimentary assistance for the PSI analysis. 
In future work we will augment the 3D model by close range 
Photogrammetry in order to represent also pillars, doors, 
windows and balconies. In a next step we want to group PS and 
match those against facade structures. 
ACKNOWLEDGEMENTS 
Part of the work was supported by the Research Fellowship for 
International Young Scientists of the National Natural Science 
Foundation of China (Grant No. 60950110351). 
REFERENCES 
Auer, S., Hinz, S., and Bamler, R., 2010. Ray-tracing 
simulation techniques for understanding high-resolution SAR 
images. IEEE Transactions on Geoscience and Remote Sensing 
48(3), pp. 1445-1456. 
Bamler, R., Eineder, M., Adam, N., Zhu, X., Gemhardt, S., 
2009. Interferometric Potential of High Resolution Spacebome 
SAR. Photogrammetrie - Fernerkundung - Geoinformation 5, 
pp. 407-419. 
Becker, S., 2009. Generation and application of rules for quality 
dependent facade reconstruction. ISPRS Journal of 
Photogrammetry and Remote Sensing 64, pp. 640-653. 
Ferretti, A., Prati, C., Rocca, F., 2000. Nonlinear subsidence 
rate estimation using permanent scatterers in differential SAR 
interferometry. IEEE Transactions on Geoscience and Remote 
Sensing 38(5), pp. 2202-2212.