un
1
J
* 3004
* 2004
* 1004
0 T T 1 T 1 T 1 1 1
50 100: 150 r mm
- 100
- 2004
- 300-
Fig. 2. Total lens distortion curve for the unit Aviogon camera - infra filter
measured in the laboratory.
radial distance r in mm 20 40 60 70 80 90
distortion e in mm/100 —10 —18 —24 —26 —25 —22
100, 110. 120. 130 140 150
—18 —10 0 +12 +28 +47
As expected, the lens distortion 1s considerable. If intersecting the r-
axis in the point r — 120 mm, the curve has an absolute minimum equal
to — 0,26 mm for r — 70 mm, approximately. At the outer edges the
distortion 1s positive and rapidly increasing.
At the same time it was agreed with professor Hallert at the Royal
Institute of Technology, Stockholm, the propounder of "the y-parallax
method", to determine the lens distortion curve in the calculation way
according to this method. So three consecutive infra-pictures, thus rep-
resenting two pairs of stereoscopic pictures, were placed at his disposal.
The two lens distortion curves are shown in fig. J.
As it appears from the curves, they have quite another shape and cha-
racter than the curve of the Wild company. Instead of the considerable
lens distortion that distinguishes such an infra-filter, these curves in-
dicate a quite small distortion, suggestive of the Aviogon distortion but
still smaller. Thus the positive maximum is about 6 pu and the negative
one about 5 y respectively 9 pu, the latter showing a certain dispersion,
too. Further the mathematical symbols are quite inverse and the curves
intersect the r-axis twice instead of once.
Summing up, I have only to state that I had to use the laboratory lens
distortion curve determined by the Wild company. The preliminary
measurings of the model deformation had already shown that the dis-
tortion of the unit camera-filter must be considerable. By the following
theoretical study I have attained definite knowledge of that problem.
11