53 
of the principal points can be found in a similar way as was demonstrated 
above under 1.225. Hence we find the expressions (142) and (145). | 
In these expressions we then substitute the expressions (197)— (202). i 
After some rearrangements we then find 
ee 
  
p (n — p) 
— y a € M | 
Qr ry = in (20 p (n — p) + 61} (207) | 
pin =p), i | 
— ; F 4 On € Q | 
Qryry = 9 {5 p (n — p) 4 25} (208) 1 
4 
] 
The standard errors of the principal points along the strip are then i 
found from 
my : $e] Qn Ry (209) j 
| 
my = So | Qryry (210) | 
s, is the standard error of the y-parallax measurements. The expressions 1 
(209) and (210) are graphically demonstrated for s, — 1 in the diagrams 
11—12. 
3. Ordinary aerial strip triangulation 
  
  
In principle the triangulation is performed as follows. 
The first model — 1,0 in a strip is established in a stereoscopic plotter 
with the aid of relative orientation. The method of independent photo- 
graphs is normally used. The absolute orientation is assumed to be 
  
at least approximately performed. 
The model 0,1 is then established with the aid of relative orientation. 
Only the orientation elements of photograph 1 are used, however. | 
  
  
  
  
Hence, the method of dependent photographs is used for the relative | 
orientation. The elements of the absolute orientation will consequently | 
partly become determined by the relative orientation. | 
The scale of the model 0,1 will be determined with respect to the 1 
model 1,0 by changing the base 0,1 until the elevation of a model | 
point near the principal point of photograph 0 gets the same elevation j 
in the model 0,1 as it has in the model — 1,0. Then the model 1,2 is I 
established and connected in the same way as the model 0,1. The l 
triangulation is assumed to be continued until the model n,n + 1. il 
! I 
Control points are assumed to be available only in the models — 1,0 1 
and n,n + 1 as demonstrated in fig. 1. | 
| 
hi 
i 
H 
En 
| 
UF 
| 
i]