mensional 
since this 
values are 
with this 
|" reveals, 
hence all 
vays have 
rs. in the 
correction 
A more 
ymputation 
of sight of 
one point, 
each other. 
rs and the 
arch with a 
ments. 
al projects, 
ion objects. 
h the latest 
ries will be 
itrol points, 
radial lens 
nknown as 
system had 
'e shown in 
cation error 
el / nm] 
0.116 
  
m» fJ 
m, f] 
my [7 
0.22 
0.21 
0.06 
  
  
0.21 
0.19 
0.06 
  
0.22 
0.22 
0.06 
  
0.21 
0.18 
0.07 
  
0.22 
0.22 
  
  
0.06 
  
ries. 
mages of the 
stage was set 
the adjusted 
ioning table. 
ient handling 
nt, therefore 
ns may occur 
and are subject to detailed investigation. The sensor, however, 
was set to a scale factor of 84.7 pixels/um, yet the adjustment 
showed 94.6 pixels/um. This deviation is yet to be investigated 
for consistency. Eventually the sensor has to be readjusted, 
since the sensor specifications serve as approximate values for 
the later adjustment. Finally, the whole system setup is 
illustrated in figure 7. 
  
Figure 7. The system configuration visualized with VRML. 
The control points on each level of the pyramid have been 
connected with lines for means of clarification. 
For the same reason, the images have been moved away from 
the x,y-plane by the same distance (cp. figure 6). 
S. CONCLUSIONS AND FUTURE WORK 
The results we have achieved are very promising. The 
calibration pyramid is a very reliable and well recognizable 
object. In connection with the tilting stage it is possible to 
acquire series of images which give accurate information about 
both, the sensor properties and the tables consistency. 
The mathematical developments allow a very flexible system 
configuration. The current system is a very good basis for 
forthcoming evaluations. 
However, deviations and problems are known and are subject to 
investigation. Additionally, as mentioned before, the section in 
space is further investigated for stability and blunder detection. 
Future work will concentrate on the application of the 
introduced techniques to the evaluation of biological specimens. 
The calibration and orientation data will help optimising image 
correlation and will improve the accuracy of the result. 
REFERENCES 
Burkhardt, R., 1981. Die stereoskopische Ausmessung 
elektronenmikroskopischer Bildpaare und ihre Genauigkeit. 
Methodensammlung der Elektronenmikroskopie, Abschnitt 
4.2.2. 
Elghazali, M., 1984. System Calibration of Scanning Electron 
Microscopes. International Archives of Photogrammetry and 
Remote Sensing, Commision V, Vol. XXV, Part A5, pp. 258- 
266. 
Hemmleb, M., Albertz, J., Schubert, M., Gleichmann, A., 
Kóhler, J. M., 1996. Digital Microphotogrammetry with the 
Scanning Electron Microscope. International Archives of 
Photogrammetry and Remote Sensing, Commision V, Vol. 
XXXI, Part B5, pp. 225-230. 
Hemmleb, M., 2001. Photogrammetrische | Auswertung 
elektronenmikroskopischer Bilddaten. Ph.D. Thesis, Technical 
University of Berlin. 
http://edocs.tu-berlin.de/diss/2001/hemmleb matthias.pdf 
Koenig, G., Nickel, W., Storl, J., Meyer, D., Stange, J. 1987. 
Digital Stereophotogrammetry for Processing SEM Data. 
SCANNING Vol. 9, pp. 185-193. 
Kraus, K., 1997. Photogrammetrie - Band 1. Grundlagen und 
Standardverfahren. Dümmler-Verlag, 6. Auflage, Bonn. 
Maune, D. F., 1976. Photogrammetric Self-Calibration of 
Scanning Electron Microscopes. Photogrammetric Engineering 
and Remote Sensing, Vol. 42, No. 9, pp. 1161-1172. 
Moré, J., 2000. Untersuchungen zur kombinierten Ausgleichung 
geodätischer und photogrammetrischer Beobachtungen. 
Diploma Thesis, Technical University of Berlin, 
Photogrammetry and Cartography (not published) 
ACKNOWLEDGEMENTS 
The authors would like to thank the Deutsche 
Forschungsgemeinschaft, for supporting our researches. 
Furthermore, we want to thank Andreas Döring, Institute for 
experimental Physics, University of Ulm, for kindly providing 
AFM measurements on the calibration standards. 
—215— 
na ES SES