56 BALLISTIC PHOTOGRAMMETRY, DISCUSSION 
  
Fig. 7. Tangential Distortion. Astrotar BC-4. No. 265. 
of magnitude for the setting error of 3 micron. 
If we get a standard deviation of adjustment of 
3 micron it indicates the bias has been removed 
from the adjustment. 
Mr B. HALLERT: What conditions do you 
use in connection with the adjustment? How do 
you determine the standard error of unit weight 
to which, I suppose, you are referring? 
Mr G. H. ROSENFIELD: From the adjustment 
we obtain the residuals from the computation, 
the residuals between the measured point and 
the projection back on the plate of the computed 
point, then from the weighted sum of the 
squares of the residuals divided by the degrees 
of freedom it gives us the mean error of the 
adjustment. 
Mr B. HALLERT: You must have certain 
conditions. You must have conditions which 
have to be fulfilled. 
Mr G. H. ROSENFIELD: We base the ad- 
justment on the projective relationship condi- 
tions, that the ray in object space and the ray 
in image space must coincide. 
Mr B. HALLERT: In other words, all the 
rays are simultaneous. But you have a few 
points in space of the missile. 
Mr G. H. ROSENFIELD: We may have 20 
flash points in one particular area, for example, 
we may have 100 flash points on the plate and 
we will have 6 plates from 6 different cameras, 
we carry them all out simultaneously. 
Mr B. HALLERT: Can we call this part of 
the relative orientation? The intersection of 
rays is t 
determin 
correct? 
orientatk 
mination 
should ii 
space. 
Mr € 
orientatic 
Generall 
them, bu 
a mean « 
that is g 
ror. So v 
with for 
orientatk 
using six 
In this 
error of 
order of 
error ag 
the adju 
with the 
quite a | 
we have 
boards € 
the plate 
With a r 
we may 
to 17, 2 
error in 
using th 
ballistic