The final adjustment gives the following results: 
- coordinates of the new points to be determined, 
- standard errors of the coordinates of the new points, 
- residuals of image coordinates of each point on each photograph, 
- number of error equations in the adjustment; which is the same as the number 
of observations, 
- number of unknowns in the adjustment, 
- standard error of unit weight of the adjustment (thestandard error of one 
observation of image coordinates). 
The adjustment does not give directly the corrections to the five control points 
whose geodetic coordinates are known. The discrepancies between geodetic and ana- 
lytically determined points can be calculated from the coordinate residuals of e e 
the image coordinates of each photograph. They can be indicated for instance by 
means of vectors. 
Results of computing and their analysis 
  
It is impossible with the computing program used here to solve more than 172 
unknowns at the same time. Each of the three photographs that were used in the 
adjustment gives 6 unknowns which are their outer orientation elements. These 
18 unknowns can be solved separately and thus they are not included in the 172 
limit. Each new point has three unknowns. The number of new points must be li- 
mited to 57 as this give 171 unknowns. As there are more than 80 points to be 
determined in all three cases I, II and III, the computing in each case had to e * 
be done in two parts. Symbols IA, IB, IIA, IIB, IIIA and IIIB denote the dif- 
ferent computations. The new points to be determined are always chosen evenly 
from the whole area of the photograph for both computation. Of course in both 
computation A and B the known control points are always the same. Parameters 
of the computed results are given in table 1. 
For the 1:3500 scale there are 20 points that have been used both in computa- 
tions A and B. For the 1:6000 scale there are 35 such points enabling the re- 
sults of the computations A and B to be compared. However for the 1:12000 scale 
there are too many points for computation for points common to both computions 
to be included. 
Table 2 gives the results of the comparison. First are given the means of the