ic data 
of poor 
ibilities 
renever 
ly inte- 
nds can 
d). The 
cm for 
true to 
cluding 
rder to 
pation, 
»uld be 
ence of 
on the 
o large, 
volving 
hrough 
gre ater 
plotters 
10w the 
to the 
e possi- 
nines in 
e repre- 
| curva- 
erators. 
rvation. 
e ascer- 
plotting 
ten due 
aterials 
  
  
  
6.1. 
1.1. 
8.1. 
9.1. 
Theoretical and practical investigation of the fundamental photogrammetric operations, 
as applied to the supervised tests IV/1, IV/2, IV/3 (A/C/IV/4) 
Professor Hallert has submitted the material of tests ^Revoira" (IV/1), *Wabern" (IV/2) and 
"Renfrew" (IV/3) to a numerical analysis, using grid coordinates and vertical parallaxes, so 
as to get theoretical values for the final errors. These values were compared to the errors determined 
directly at the verification stage of the tests. The results show that by recording the residual 
vertical parallaxes after orientation of the model one can reliably forecast the accuracy of the 
plotted map. Clues about the nature of the error components are also obtained. The tolerances for 
photogrammetric work could well be based on such analysis of parallaxes. A desirable result of 
the London discussions would be the standardization of these analytical checking procedures on 
the basis of international cooperation. 
Convergent versus near-vertical photographs (B.1/Panel 1) 
It cannot be expected that the papers and discussions make possible a final evaluation of the 
practical advantages and disadvantages of these two systems. The available theoretical investi- 
gations and tests do not rest on a sufficiently broad basis to be significant. It will however be 
important to set up a program for future research, which would try to find out what tasks are 
better solved by the one or the other method. 
Graphical or numerical photogrammetry | (B. 1/Panel 2) 
Numerical photogrammetry is still in its infancy, but its development meets now with great 
interest. The results secured up to now are already of benefit to cadastral surveying, to the deter- 
mination of volumes (open air mining), to the design and staking of highways, railroads, canals. 
The application of the numerous recently unveiled pieces of equipment and methods (Stereo- 
comparators, Stereomat, Orthophotoscope, recording devices and computers) can only be dis- 
cussed in general lines, keeping in mind the desirable qualitative and quantitative improvements 
in the production of coordinates and maps. Ultimate goal is to free the personnel from routine 
work, letting them concentrate on such mental activity as implies human judgment. Digital 
photogrammetry transcends photogrammetry itself. It projects automation into the field of 
the users of its results, as shown for instance in the case of civil engineering. The cooperation be- 
tween programmers, photogrammetrists and engineers or land surveyors should be a rewarding 
matter of discussion. One might try for instance to draw a line between what pertains to numerical 
photogrammetry and what should better be left to graphical photogrammetry, considering econo- 
mical as well as qualitative viewpoints. 
Methods for filling in the blanks and for checking the photogrammetric maps  (B.1/Panel 3) 
There is not yet a satisfactory solution to the problems of limiting the blanks left over after 
plotting, or of filling in the blanks by field operations. Better photo-interpretation and identifi- 
cation should be achieved in the first place. The practical importance of these questions is great if 
results of homogeneous accuracy are to be secured. Other important points requiring further 
elucidation refer to tolerances and to methods for checking the accuracy, completeness and relia- 
bility of the results. The discussion should show whether it would be possible and useful to set up 
norms bv international cooperation.