PHOTOGRAMMETRIC ENGINEERING 
7 
provide a visual check on the stability of the 
origin of the machine coordinate system and 
a means to damp out vibrations which could 
be disturbing not only for the pointing proc 
ess, but which, because of the physical prin 
ciples inherent in the Ferranti-counters, 
could cause recording errors. 
As a further improvement in this direction, 
the inclusion of memory circuitry is en 
visaged, which would, in addition, dispense 
with the need for carriage locking and in 
crease the measuring economy by reducing 
the period of immobility by 80%. 
Addressing himself to the problem of 
economy, the author suggests modifications 
in the auxiliary photographic registration 
equipment. Some of these improvements are 
already in operation, whereas others are con 
sidered for future application. 
Mr. Szangolies, in his paper, presents a 
study of problems created by the need for 
registration of coordinates in modern photo- 
grammetric techniques. There can be no 
doubt that both economical considerations 
and the desire to eliminate gross errors in 
reading and recording are sufficient reasons 
for giving full attention to these problems. 
The paper correctly emphasizes the increas 
ing demand for flexible, human operator- 
oriented methods for supplementing the 
registration of the actual measurements by a 
rather large amount of additional informa 
tion such as point number, point type, 
weighting factors, etc. The author’s emphasis 
is well placed, because the potential pro 
vided by electronic computers cannot be used 
effectively in production work unless the 
various types of raw data can be identified 
and assimilated in the computer by means of 
common indexes. A lack of this kind of flexi 
bility can be particularly troublesome in the 
area of numerical photogrammetric tri 
angulation, where the data evaluation is 
based on sophisticated mathematical models 
which are capable of incorporating a wide 
variety of constraints, many of which refer 
to the same parameters. 
Another important point in the paper is the 
problem of providing flexibility in the output 
format of the coordinate reader in order to 
assure compatibility with the input format 
of a particular computer. 
The importance of clearly arranged tabula 
tions, supplemented by plenty of typewriter 
headings in clear text, cannot be over 
emphasized, because these types of output 
are essential for the editing, checking and 
analysis of intermediate or final results. 
The remaining three papers are concerned 
strictly with the economical aspect of num 
erical photogrammetric methods. For various 
reasons, some of which could be considered 
controversial, the authors share a common 
point of view. The core of their arguments is 
that there are certain problems in aerial tri 
angulation, and in the determination of 
spatial coordinates of individual points at 
large scales, where accuracy requirements for 
the final results can be met if the mean error 
of unit weight for the corresponding coordi 
nate measurements is ± 5ju, ± 10/x, or even 
greater. Clearly, as outlined at the beginning 
of this discussion, such error bounds allow the 
use of analogue type restitution equipment. 
For such cases, the application of numerical 
photogrammetry can be justified only if the 
cost of the numerical method is less than the 
cost of the analogue approach. 
Experiments have been conducted in vari 
ous countries by different agencies which, in 
general, lend support to this position. 
Such results have encouraged the VEB Carl 
Zeiss, Jena, to develop a data processing sys 
tem that is tailored to meet somewhat re 
duced accuracy requirements. This equip 
ment, which is being marketed as the Stecom- 
eter with automatic registration, is dis 
cussed in the paper by Dr. Weibrecht. The 
instrument is built around the quite re 
nowned pre-war Zeiss Stereocomparator 
1818. The new instrument is equipped with 
automatic registration and has an accuracy 
of ±2/i for the measurement of coordinates 
and parallaxes. This result is all the more re 
markable, because the instrument, built as an 
economy-solution, and in addition having 
features which facilitate operations, can 
handle raw material up to 23 cm. X 23 cm. 
The maximum magnification of 18 times, with 
a field of view of 17.5 mm. diameter, is suffi 
cient to make full use of the stated measuring 
accuracy. 
The adaptability of the instrument is en 
hanced by providing 6 selective, illuminated 
measuring marks with different sizes and 
shapes. There may be a problem when 
pricked points have to be measured, an 
operation demanding a black dot as a meas 
uring mark. The means for auxiliary photo