CONVERGENT VERSUS VERTICAL PHOTOGRAPHY, AUTHOR'S PRESENTATION 119
results, however, proved to be hardly any better. This gives the impression that also
here the inner orientation could be the cause and that the camera is guilty!
The conclusion of all considerations is that at present a fully justified comparison I
of convergent and vertical photography is impossible. As long as we have such a great |
difference in precision between the various experiments the geometrical proportions are I
of a very limited value. The exception is perhaps the use of the twincamera in aerial |
survey for highway construction, since there a larger scale of photography has less |
influence on the economy.
Nevertheless it must be considered an important problem to find out which are the
real qualities of convergent photography in order to determine in which cases its appli-
cation will have advantages above vertical photography. I agree, however, fuly with a
statement, which Hallert made in a letter to me on April 14: “Special tests of convergent
aerial photographs should be performed for determinations of systematic errors and
estimations of the residual irregular errors.
Measurements of y-parallaxes in connections with restitution should regularly be
performed (in particular the residual parallaxes after finishing the relative and
absolute orientation). The theoretical connections between the errors of the fundamental
operations and the final results should be derived for different assumptions”.
It is evident that all this is missing. Perhaps there will be a chance to fill this gap
by the proper use of the new material taken in Germany for the commissions B and C
of the O.E.E.P.E., among this material are a great many convergent models as well as
vertical photographs of different types. The sense of this paper is then no other than
to show how poor the position of research in photogrammetry still is at present notwith-
standing all the theory of errors published during the last 20 years.
Author’s Presentation of the Paper
in the Meeting held on Thursday, 8th September, 1360
Mr President, Ladies and Gentlemen, first 1
have two remarks. I regret very much that not-
withstanding the fact that the first two issues
of “Photogrammetria” were ready on 16th
August, and have been sent from Amsterdam,
they will perhaps arrive today in London. They
were kept in Holland until 2nd September some-
where, and after that time, between the Hook
of Holland and this building, I do not know
where. However, it has been promised that they
will be arriving today for those who want to
have them here.
The second remark is that I will try to make
this a proper discussion and not a series of
lectures prepared at home. This means that I
will make a few short remarks and then ask the
public to raise questions, not to make speeches
but only to raise certain questions which must
be answered by the members of the panel. Then
perhaps there might be a short discussion about
the same questions amongst the members of the
panel who perhaps also can reply to some of
my questions because there are quite a lot that
remain.
The conclusion of my story is that at the
moment when I finished this report there was
not enough sufficient reliable data available to
evaluate the merits of each of the two systems.
If you look at the geometrical position, that
viewpoint is without any doubt in favour of
convergence. You need less pairs and the
coverage is larger, but that is only one viewpoint.
Another viewpoint is the precision which can be
obtained. A second consideration is the time
necessary, and the last consideration is the over-
all cost of the equipment, because we should
remain with our feet on the ground in all photo-
grammetric problems.
What I have said already from the geomet-
rical point of view is not so difficult. Bruck-
lacher has delivered a paper to me which I
have included, in its main points, in my small
report. There you find the advantages from the
geometrical point of view. But I went to pieces
entirely at the moment I tried to evaluate the
precision. I regretted very much that, from let
us say the most important colleagues, at present
advocating convergent photography, the US.
Geological Survey, I did not get any more com-
ments on this aspect, as it was such a good
method. That is what I knew already. We have
been using it in Delft since 1932, but I might