very 
his 
the 
am- 
dis- 
ina- 
the 
The 
fer- 
ning 
ere 
am- 
Dre- 
n of 
oad 
ues- 
etry 
dto- 
to 
1 as 
and 
cate 
» to 
wuld 
vey 
vith 
; of 
; on 
; on 
. In 
>, à 
| by 
pto- 
ary 
[he 
me- 
s à 
ient 
Ihe 
ein, 
ma- 
ou. 
del 
phy 
the 
on- 
the 
eli- 
ves 
  
GRAPHICAL OR NUMERICAL PHOTOGRAMMETRY ?, DISCUSSION 135 
us an excellent aid in solving the problem of 
getting the road well fitted in the terrain so that 
later on when the road is to be constructed it will 
just be a part of the landscape. 
When we have done the preliminary planning 
we then use the electronic computer for making 
the calculations which are necessary for judging 
the results from the planning in the model. By 
using photogrammetric instruments in this way 
— those which are cheap in price and easy to 
handle — for the preliminary phase, we also get 
very good help by using photogrammetry on a 
local level in engineering on projects in the field. 
It is a good thing to make it possible for the 
engineer to use the picture intensively without 
being a photogrammetric specialist. The results 
will surely be much better when we have 
educated civil engineers in photogrammetry; 
make them have more photogrammetric special- 
ists in civil engineering. The interest for using it 
arises particularly when the engineer has the 
feeling that he himself performs the real plan- 
ning and when photogrammetry has been in the 
field just what it should be, an aid to civil engi- 
neering. 
There is much more to be said on this sub- 
ject, but my time is up. 
Dr J. M. ZARZYCKI: I would like to illustrate 
some procedures on the results we have obtained 
in Canada in applying numerical photogramme- 
try in the construction of some railways. I 
would particularly like to mention a project of 
a 192-mile long railway in northern Canada 
which goes from the St. Lawrence river, and it is 
completely virgin territory. 
Photogrammetry and electronic computation 
are used for two reasons: first, it provides a 
means of performing very accurate measure- 
ments; secondly, plotting instruments to which 
electronic printers are attached provide means 
of automatic handling of a great amount of 
numerical data. This application is relatively 
simple. The photo construction is done, the 
photography is obtained, cross-sections are 
taken and then the same is done during the 
design stage. 
I should like to show some pictures of the 
project to show you how this is done. 
This slide (figure 6 of my paper) is a pho- 
tograph of a right-of-way before construction. 
This is approximately 100’ clear of the right-of- 
way. It is bush country so that before photog- 
raphy can be obtained the right-of-way has to be 
cleared. After it is, it is photographed again. 
We place markers every 30 metres, and outside 
of the right-of-way we place permanent control 
points. We establish only the elevation of these 
points. We know their position, so when the 
model is set up in the plotting machine the 
scale is established within this point. This can be 
levelled very well. Then on the manuscript we 
plot the position of this point. The position of 
the permanent control point is established by 
photogrammetry. The cross-sections are then 
obtained perpendicular to the central line, and 
the elevations are made around the central line. 
This slide (figure 8 of my paper) shows the 
same area after construction. You can see the 
location of the permanent control points which 
are used to set up the model levels on the scale, 
and there we obtain cross-sections of the final 
grade to determine the final quantities. 
In the preliminary stage when we obtained a 
cross-section, we computed all the data required 
by the railway highway engineers, the quantities, 
the distances and the elevation required for 
staking off the centre line. 
This slide (figure 1 of my paper) is roughly 
an outline of the flow of data. Since this project 
is many, many miles from head office, we have 
the field engineer telling the aircraft where to 
fly, the film is developed and it goes to the field 
engineer to identify the control points and 
provide other information regarding soil and 
cross-sections. Then the information goes to 
head office. Manuscripts are prepared, the plot- 
ting is done with EK3. We have done some ex- 
periments with the curves, but we have found 
that the A8 is better because in other evalua- 
tions it is more economical. We are using the 
IBM 650 for computations. The result is want- 
ed graphically, so the computer produces extra 
cards which are fed to the electronic data plotter 
which plots the cross-sections graphically. It is 
obtained with the plotting machine, and also the 
design cross-section. You have the numerical 
part and then it goes forward. 
I will not show you any instruments. I just 
want to mention some accuracies we have ob- 
tained. In a 7-mile long section, 211 points have 
been checked in elevation. 80% of the points 
have an error from zero to 3 centimetres; 18% 
have an error from 1 centimetre to 6 centi- 
metres; and 2% have an error from 6 centi- 
metres to 9 centimetres. The maximum error 
was 9 centimetres which was 1 : 15,000 of the 
flying height. The photography is obtained be- 
tween 1 : 1,300 to 1 : 1,400, and the plotting is 
done with the A8. 
Mr F. J. DoyLE: I had prepared a paper on 
the standard procedures which are used in 
numerical photogrammetry for highway design 
  
  
[TE 
BT