> air base
of aerial
r stations
'acy with
y Shoran.
ve to ten
ir aspect
oughly is
1 horizon
> record:
uch new,
(now the
hing that
ling. We
1e neces-
se aerial
ary data
geodetic
auxiliary
auxiliary
ds which
ography,
e. Then
em, and
esting is
cameras.
he auxi-
ographs
the time
' vehicle
of time
18, for a
iknowns
ites can
describ-
ou very
n about
L
of Mr
ditional
report,
of Geo-
should
tion of
n.
t of the
mme
AERIAL TRIANGULATION WITH AUXILIARY DATA, DISCUSSION 103
introduction of auxiliary data provide a greater
elasticity than the instrumental one. Experience
at our Institute led us to the conclusion that it
is always preferable to introduce auxiliary data
as an element of absolute orientation of models,
especially owing to the present scarce precision
of the data themselves. Furthermore, we wish
to point out that the introduction of auxiliary
data is very simple in analytical triangulation,
since it requires only a slight modification of the
general computation programme. Also, for solar
triangulation which usually appears to be rather
laborious, the preparation of a suitable and
simple calculation programme allows us to de-
termine quickly from the solar co-ordinates on
the photograph the data necessary for the ca-
mera orientation.
Let us examine now in more detail the
statoscopic triangulation and the solar one. In
regard to statoscopic triangulation, we have
already had the opportunity to remark that the
idea considered in Dr Brandenberger’s report,
of an improvement of the statoscopic data to
be obtained when considering the A rotations
of the fixed camera in instrumental triangu-
lation, is purely illusory. In fact, it gets rid of
the bz absolute value introduced and takes the
statoscopic triangulation back to one which is
similar to the triangulation with bz = 0.
On the contrary, a method of correction of
at least a part of the statoscopic accidental er-
rors — which we successfully experienced re-
cently with analytical methods — is as follows.
First of all, the strip is calculated and the
heights of the camera are recorded. The dif-
ference will be made between those heights,
after the linear distribution of the closure er-
ror, and the ones obtained from the statoscope.
A diagram of the same differences will be
drawn. An interpolation curve is built, allowing
us to calculate — even if it is empirically —
sufficiently approximate values of accidental
errors of the statoscope. However, this is not
a new method since it is also used with radar
profile methods.
Finally, each model is turned taking into
account the corrected bz statoscopic compo-
nents thus determined. The analytical method
lends itself particularly to the application of the
simple proceedings because the accidental er-
rors of each model are really very small.
The results which have been obtained until
now are encouraging, since the mean height
error of the camera, with a simple statoscopic
triangulation, in a 100 kilometre strip taken
from 4,500 metres altitude, equal to + 3.6
metres, in this way is reduced to + 2.8 metres.
Experiments are continuing on longer strips
and also on very rugged terrain.
As far as solar triangulation is concerned,
the analytical method allows us to apply the
solar data more widely than is done instru-
mentally. In fact, using the plotting machine
we are practically obliged to compute omega
from the knowledge of while analytically it
is possible and may be convenient to derive
q or o from the free bridging and to calculate
the other two rotations through the co-ordinates
of the image of the sun.
By analysing the problem in theory, the
conditions have been determined in which we
may use one or other of the indicated pro-
ceedings so as to be sure to attain a sufficient
accuracy.
Of course, good results from these pro-
ceedings depend, for the most part, on the pre-
cision of auxiliary instruments. We have been
pleased to read about the results obtained by
Dr Brucklacher with statoscopic triangulation.
As for the sun, which already supplies good
proximation, we deem it possible to increase
precision to the extent of obtaining the direc-
tion of the rays of the sun with an approxima-
tion similar to the one of the direction going to
the point of the ground.
We have begun at our Institute to experi-
ment with the different proceedings of the in-
troduction of solar data and their coupling
with statoscopic ones, although the photogra-
phic material at our disposal was not taken in
the most suitable conditions. In the near future,
we will carry out a more intense experimental
research on long strips.
Prof W. SCHERMERHORN: This is a very im-
portant subject in aerial triangulation. It has
already occasioned this morning a discussion
between myself and Mr De Masson D'Autume.
A lot of work is being done in this field in dif-
ferent places. Only part of this has been made
public. If possible, next Wednesday there will
be another meeting of Commission III in the
afternoon. There will be two successive hours.
At 1.30 we shall have a discussion on the report
of experimental works organised by Commis-
sion III and then at 2.30 a discussion on the
general report of Commission III and the pre-
paration of resolutions. I hope that it will be
possible to curtail those discussions to such an
extent that we may give more space to this
most important subject; perhaps we can cut
down a little on those which are less important.
|
lil
ui
il 1
