ion 
  
In connection with the computations the photographs from the Avio- 
gon camera proved to give unexpected high standard errors, see further 
Table 5. This proved to be caused by the lacking flatness of the plates 
of the size 23 X 23 cm. The deflections of the plates were measured, 
however, and the radial distortion amounts could be corrected. 
3.2 Determination of the camera constant (calibrated focal length) 
and the principal point of the photographs 
First an approximate value of the principal distance was determined 
from one of the circles. Then the radial distortion was determined with 
respect to this value. Finally the radial distortion curve was transformed 
with respect to a certain zero-point. Simultaneously a correction to the 
preliminary principal distance was obtained and added. In this way the 
calibrated focal length was determined. For an accurate determination 
of the calibrated focal length it is necessary to know the elevation of 
the external perspective center above the grid on the ground. For sym- 
  
  
Fig. 8. Cross section through an Aviotar lens from which the asymmetry is 
evident. 
metrical lenses the conditions for such a determination are rather simple 
but get more complicated for unsymmetrical lenses for instance Aviotar. 
See Fig. 8. Since the distance camera — ground is comparatively short 
in comparison with the flying altitude at aerial photography corrections 
to the computed calibrated focal length can be calculated and applied. 
On the other hand, for approximately vertical photography and com- 
paratively flat ground small systematic errors of the calibrated focal 
length are of minor importance. In Table 3 the determined calibrated 
focal lengths of the cameras Eagle, RMK P 10 and RC5-Aviotar are 
demonstrated. The corresponding radial distortion curves were de- 
monstrated in Diagrams 1—5. In Table 3 also the calibrated focal 
lengths from laboratory determinations are demonstrated, 
  
—— 
MER I= 
Ts