as small-detail reproduction is concerned, not 
of course as far as large sized detail and macro 
detail and tone-reproduction are concerned. 
The density compression shall take place in 
all areas which carry high light and shadow 
details of interest, and for that reason the un- 
sharpness of the mask for the scanning spot 
shall be small, in the order of the detail itself, 
say one-tenth to one millimetre. An important 
point to note is that existing dodging printers 
do not yet fulfil these requirements, and are 
consequently not yet of full value to photogram- 
metry as far as economy of micro details is 
concerned. We hope that it is possible for im- 
provements to be made in the near future. 
In cases where the photo scale is determined 
by the recognisability of small detail, this has 
a great bearing on the survey economy, because 
the photoscale is determined then by the repro- 
duction of detail. The necessity of adapting the 
photo scale can account for a factor of two in 
the photo scale; consequently in a factor two 
even in the number of photographs. This work 
has shown that these conclusions are only valid 
for very small detail exclusively. For small 
objects the resolution criterion is basically 
wrong, and we must use there a contrast trans- 
mission if we want to judge performance. The 
positive process, including dodging, can only be 
used by contrast transmission as well as judging 
the contents of photographs and judging inter- 
pretation of aerial photographs. 
Mr A. J. WATSON: I would like to say that in 
my field, which is the fluorododge printing field, 
we are researching towards the fineness as 
mentioned by the previous speaker, and towards 
the overall compensation which will approach 
nearly those limits by principally projection 
methods rather than by contact methods, which 
in turn will give greater resolution results. We 
are continuing such development at the present 
moment. 
Mr G. C. Bnock: Referring to Mr Corten's 
paper which he has just summarised, I am not 
sure whether Mr Corten said at the platform 
what he said in the paper, which seems to imply 
that the paper print can be superior to the dia- 
positive. I think it should be quite clearly un- 
derstood that we have to specify the character- 
istics of the diapositive and that if you have a 
diapositive of fine enough material or grain you 
will certainly be able to do as well as you can 
on a paper print. In other words, I would sug- 
gest that the experiments done at I T C do not 
extend to a positive transparency material of 
   
104 PROGRES ACCOMPLIS, DISCUSSION 
fine enough grain and high enough resolving 
power. 
I would expect from the shapes of the fre- 
quency response curves of negative and positive 
materials that if you want to reproduce cor- 
rectly all the detail in a given negative, you have 
to use a positive material whose resolving power 
is at least three times that of the negative emul- 
sion — at least three times, perhaps more — and 
commonly available diapositive materials are 
nothing like as good as that. However, they can 
be obtained and I have seen plenty of evidence 
that given the right kind of positive material all 
the detail in a good negative can be evaluated. 
I think it is merely a matter of choosing the con- 
stants directly, it is not a fundamental disagree- 
ment with Mr Corten’s paper. 
Mr F. L. CorTEN: We have indeed not used 
various positive materials, but we used the best 
positive material which is commonly available 
and commonly used in aerial survey, but I did 
not know there were better transparencies 
available. 
It was not the intention, by the way, to state 
that the performance of a paper print can be 
better than a diapositive transparency. How- 
ever, I would like to make another remark: it 
is dangerous to confuse discussion about reso- 
lution and performance, that is to confuse dis- 
cussion about resolution and about contrast 
transmission. It is a very dangerous thing. I 
think it would be much better in future if 
when we speak about resolution we just mean 
that minute micro detail only, and if we want 
to judge image performance as it interests the 
photogrammetrist — that is details in the pho- 
tograph which are of the order of size of 
a tenth of a millimetre or one millimetre — 
we do not speak of resolution any more as 
a criterion, but that it must basically be judg- 
ed by something like contrast transmission 
where the frequency does play a role. 1 know 
we have discussed that and we agreed on this 
point, but I wanted to state this in the audito- 
rium. One of the conclusions of this paper - 
one of my conclusions, as I told you — was that 
we started using the resolution as a criterion but 
we found that particularly in judging positive 
processes, scanning printers and so on, this cri- 
terion was basically wrong and should not be 
used, but that exclusively contrast transmission 
should be used. 
As far as the remark about Kel-O-Wat is 
concerned I am in complete agreement. We also 
measured Kel-O-Wat contrast transmission of 
the Kel-O-Wat printer, and we came to the con- 
  
  
  
  
  
  
  
  
   
  
  
  
  
  
  
  
  
  
  
  
  
  
   
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
   
  
  
  
  
  
  
   
  
  
  
  
  
   
   
   
  
  
  
  
  
   
   
clu: 
size 
the 
case 
feed 
tran 
part 
few 
colo 
like 
Ami 
equi 
in c 
EXP 
in Oo 
com 
the 
char 
spot 
IS tc 
thes 
ther: 
impr 
men: 
spot 
of, f 
Archives