e 66 
® 
>) 
NP 
ü 26 
rl 
0,01 mm 
idjustment 
the discre- 
tograph 2. 
tude. In 
hs, com- 
narks the 
| average. 
1e affine 
al marks. 
on of the 
dr‘ microns 
r'mm 
  
— film (including refraction and curvature, air test ), 3 experiments 
== = = plates (tower test), 4 experiments 
— —7 according to the manufacturer 
Diagram 30. Example of comparison between radial distortion curves of the lens 
(the manufacturer) of plates from tower tests and of films from air tests. 
Evidently the distortion determination from the tower agrees very 
well with the laboratory determination by the manufacturer. Consider- 
able differences are evident from the air test. The standard error of 
the radial distortion amounts is of the order of magnitude 2 and 3 
microns for the tower and air tests respectively. 
6. Summary of the tests of photographs from aerial cameras 
The primary purpose of the performed tests was to investigate the 
practical application of the grid method for the determination of non- 
projective systematic disturbances (primarily radial distortion) in the 
photographs from a number of different cameras under varying con- 
ditions. The grid method has been applied to tests of vertical photo- 
graphs from a tower (about 125 meters above the grid) as well as from 
an airplane (about 5000 meters above the grid). The photographs from 
the tower were taken on film and plates, through different filters and 
in different temperatures. Five different cameras and seven different 
lenses were used for the tests, 
The photographs from the air were taken on film only. 
In the films also affine deformations were determined. 
The results of the tests can be summarized as follows: 
6.1 The grid method has proved to be of great practical value for! 
the determination of certain non-projective systematic disturbances in 
photographs from aerial cameras after photography under varying con- 
ditions. Since the grid method is entirely founded upon the method of 
the least squares the standard error of unit weight of the image coor- 
dinate measurements is automatically and uniquely obtained. This 
standard error characterizes the basic geometric quality of the photo- 
graph and can be used for a unique determination of the accuracy of 
functions of the image coordinate measurements, i. e. also the deter- 
mined systematic disturbances. It can also be used as a kind of geo- 
metric resolving power. 
55