of 20 cycles/mm has been chosen because here the system is suffi-
ciently sensitive to defocusing and the contrast is still high
enough over the total field to guarantee an exact measurement.
Figure 6 gives the tangential MTF-values of the lens of Fig. 1 as
a function of the field angle for 5, 10, 20, 30 and 40 cycles/mm.
It shows the similarity of the curves.
Instead of representing the modulation transfer as a function of
field angles for a fixed spatial frequency, it is possible to show
the spatial frequencies as a function of field angles for fixed
modulation transfer. This has been done in Fig. 7 for the same lens
as in Fig. 1 and Fig. 6, at tangential azimuth for madulation
transfer values of 0.8, 0.6, 0.4 and 0.2. These curves are very
similar to resolution tests on low contrast targets.
Such curves as shown in Figures 1 to 5 or in Fig. 6 help very much
to reveal shortcomings as well as good aspects when comparing the
lenses. However, one spatial frequency, even if it is a characte-
ristic one, or one modulation level does not provide the same in-
formation as the total number of OTF curves which are necessary
for a more detailed quality comparison.
In Fig. 8 to Fig. 11, MTF- ve ea two R C
n Fi to Fi curves| evaluate wo Reseau Cameras .
& a ven for fiers angles at Congential and radial azimuth, ET
the wavelength A = 546 nm in the plane of best MTF in axis and
parallel to the reseau plate. It can be seen that the OTF curves
of camera No. 189 for all field angles in tangential azimuth are
slightly better for all frequencies. For radial azimuth and for 20 ©
field angles, the MTF-curve of the camera No. 188 is better for
all frequencies (Fig. 10, lower part). Because of the similar
construction of these lenses in the cameras, the shape of MTF-
curves is the same; therefore the MTF curves are either better or
worse for all spatial frequencies. The differences between the two
cameras can also be seen by comparing only the MTF-values for one
frequency, in this case 20 c/mm (Fig. 12).
