%®
N " ’ A — B —— S mr "— — M page nem ee ra er pique Lo A
EE A A TARR En TF DAS EE EE DONC ESSOR BE OA S TETE
magnitude of the granularity. This method will be used in this paper.
Determinations of the relative variances of coordinate observa-
tions as a function of image quality on the image plane were made in
the following steps:-
(ij Derivation of the MTF's of components of the photogrammetric
system,
(ii) Computation of the overall MTF of the system,
(iii) Evaluation of the FL from the intersection of the MTF curve
of the system and the Modulation Sensitivity curve of the visual system,
(iv) Derivation of ideal target sizes and consequently pointing
precisions from graphs of (Trinder, 1973),
(v) Computation of variances for x and y-coordinate observations.
It should be emphasised that the variances derived relate only to
factors of image quality and observer performance, and not to measuring
accuracies of the stereoplotter. Variances derived in this paper must
be combined with variances of the stereoplotter, e.g. between 0.5um
and 5um to obtain the projected variance of coordinate observations
on a specific stereoplotter.
2.3 MIF's of Photogrammetric System Components
In (Trinder, 1974) an investigation was made of the influence of
photogrammetric system components on target selection. MTF's chosen
for that study for film, image movement, and stereoplotter optics will
be adopted for this paper. Several recently determined MTF's have been
used to represent those of the camera lens. The MTF's chosen are:-
(i) Film - Kodak 2402 film (Kodak,1971) with y of approximately
unity. The MIF curve was normalized to 100% modulation at low
frequencies (Figure 1).
(ii) Image Movement - Since the effects of image movement vary
with flying height (i.e. photograph scale) two cases were computed.
The first was for large scale photography (1:7200) and a very large
image movement of 37um at photograph scale while the second case was
for small scale photography (1:50,000) and an image movement of 6um.
An exposure time of 1/300 second with an aircraft speed of 300 km/hr.
were assumed. The MTF of an image movement of 37um is placed with the
index arrow at 100/IM in Figure 1.
(iii) Optics of Stereoplotter - representative figures of MIF of
the stereoplotter optics are given by Hempenius (1969). Clearly this
component is difficult to estimate as little information is available.
The MTF given in Figure 1 is of similar form to the on-axis MTFs of
camera lenses.
(iv) Camera Lens - The combined lens filter functions vary with
orientation of the test targets (either tangential or radial), and the
inclination of the incident ray. Data chosen were those presented by
Rosenbruch (1976) for 2 wide angle cameras, coded 188 and 189, and
that given by Würtz (1969) for a super-wide angle. Rosenbruch has
b
ki
9
h
EL
A
oh
a
b
Al
gl
a
49