(181) 
THE VERTICAL PROBLEM IN AERIAL PHOTOGRAPHY 
By Dr. Ulrich K. Heidelauf, Consultant for Reconnaissance, 
Photographic Reconnaissance Laboratory, WABC, U.S.A.F. 
Ever since aircraft have been employed carrying equipment to furnish 
photogi aphy foi surveying purposes, the problem of determining the exterior 
orientation of the photographs has existed. Determination of the nadir point 
is a major pai t of this general problem. The complex problem is customarily 
solved by spatial triangulation, referring the geometry of the photography to 
accurately surveyed check points on the ground. The procedure of providing 
such check points is costly and time-consuming and sometimes impossible. I 
think it is generally agreed that considerable improvement and simplification 
of methods and procedures presently employed in aerial photogrammetry 
could be achieved if the elements of exterior orientation could be made reliablv 
known. 
I shall exclude in my considerations the problem of determining absolute 
or relative altitudes, which has been very forcefully approached on the basis of 
electronic altimetry in Canada as well as in the United States and has shown 
very promising results. People more competent in this particular field of tech 
nology than I should report on the result of their work as soon as it is publicly 
presentable. I shall also exclude the problem of azimuthal stabilization and 
reference which may become a matter of a separate presentation at some later 
time and shall confine myself to the problem of verticality, or nadir point 
recording, or tips and tilts, or whatever name seems to fit the problem best. 
Basically, there are two different approaches to the problem of determin 
ing the nadir point in aerial photographs. The first is to measure deflections of 
the optical axis from the true vertical position at the instant of exposure and 
to record them on the photographs. The second procedure is to prevent such 
deflections by employing proper servo control to the camera mount to contin 
uously maintain coincidence between the true vertical and the optical axis of 
the camera. Theoretically, the two approaches are equivalent, since they both 
provide knowledge of the same elements of exterior orientation. However, 
some people seem to think that, for some reason or another, recording is better 
than controlling the camera in the direction of the vertical as a matter of prin 
ciple, because I believe the prevention of errors is better than tolerating and 
recording them and making their elimination a matter of a separate procedure. 
The servo control is no problem, since parallelism between gyro reference line 
and camera axis can be maintained easily within 1 of arc under photogramme- 
tric flight conditions. Servo control of cameras is anyway required under mili 
tary condition of operation for other reasons and may therefore as well pi o- 
vide continuous verticality. It can be seen that either one of the two appioaches 
must provide or have reliable knowledge of the true vertical direction in flight 
in order to operate properly. This, however, is exactly the point wheie both 
methods have failed so far. 
A great number of photogrammetrists all ovei the woild seem to have 
been accepting, as an established fact, that the true vertical cannot be pi oduced