44
SYMPOSIUM PHOTO INTERPRETATION, DELFT 1962
near the limit of resolution. Photographic tone reproduction is the reproduction
of larger details - sometimes called “macro detail” - and is not very much
affected by the scale or the size of this detail. But the reproduction of the
smallest detail near the resolution limit - sometimes called “micro detail” -
is much more complicated. As the image size becomes smaller, the contrast is
also reduced, because in the reproduction of micro detail, size and contrast
must be considered together. In this respect we may draw attention to the
important work of Welander. His paper on the T-functions deserves more
attention than could be given during the short time of its presentation at this
Symposium. At our institute, Hempenius is working at the basics and the appli
cation of the T-functions in aerial image formation.
It is of the utmost importance that the principles of photography as a branch
of applied physics are understood and applied in photo interpretation, because the
photo interpreter is particularly interested in the smallest details, such as
“texture”, and other information near the limit of recognition.
Conclusion
Finally it can be said that “rational planning of aerial photography for a
specific purpose” means:
“optimization of the complete system to that purpose”.
To achieve such optimum conditions, experience and insight into all aspects
are needed. But even then, part of it is still personal judgement. Gradually,
however, we are getting to know more about the possibilities and the tolerances
in survey navigation and we are gaining a better insight into the aerial image
formation process. For photo interpretation it is important to use this complete
knowledge: the application of the total complex to aerial photography leads to
the most economical aerial photograph.
Discussion
J. A. E. Allum (U.K.) : After having gone to the expense of producing the aerial photography
would it not be sensible to produce copy negatives at different scales? Answer: Photographic
enlargements or reductions do not give more information than the original scale. The smallest
scale which is still useful for the required purpose should be chosen. F. Ruellan (France)
endorsed this answer by stating that each purpose requires its own scale.