ii 
| 
H 
| 
iH 
E 
El 
f 
4 
N 
  
The RC5a camera used in the test is fitted with two heating rings, one each 
around the front and back lens flanges which cut in when the temperature falls 
below 89C. The first flight was made with a window and the second without 
a window. Each test covered six photographic runs made in two hours time. 
With the window installed the outside air was between -24°C and -32°C, the 
outer side of the window registering 2° higher and the inner sides 3° higher. 
The camera bay air showed a constant decrease in temperature from 10°C to 
-9' C at the end of the 4th pun, approximately 1 hour 20 minutes after camera 
standby, leveling off at -9'C to the end of the photographic run. Simultane- 
ously, the outside of the filter dropped to -2°C while the focal plane rose 
to +16% and dropped to +12°c, maintaining a difference between 14 and 15°C. 
The total temperature difference of the imaging forming system from outside 
of the window to the focal plane is 42°C. 
In the second flight, there was no window but the camera was sealed with a 
rubberized fabric membrane to prevent the penetration of outside air to the 
camera bay. At the first run the outside of the filter reached -17°C, which 
was 18 higher than the outside air. It remained within 2 (to 19%) for the (9 
remainder of the photographic runs. Simultaneously the focal plane tempera- 
tures dropped from #15" C to +11°C. Differences across the camera image volume 
were, therefore, 32°C to 29°C, with no condition of equilibrium. 
Worten's test and his general discussion lead to the question as to how accuracy 
of geometry under such varying conditions can really be known. Unfortunately, 
it is not possible to calibrate under conditions of the environment of each 
survey, to obtain this data. The first task, therefore, is to deteriine the 
geometric and image quality variables for certain known survey environments, 
then instrumenting cameras so that temperatures and pressures can be known, 
a measure of correction can be applied. The use of control targets would 
also increase the accuracy. 
Ziemann, of Canada's National Research Council, discusses changes in image 
geometry attributable to changes in temperature which in turn change camera 
dimensions. Four particular situations he notes which can occur at any stage 
of the imaging process are: changes in the camera body, during the film 
flattening in the aerial camera, due to aging, and during measurement. (à 
His point that the fiducial marks needed to reconstruct the image (position) 
at the instant of exposure requires that the marks maintain their position at 
any temperature the camera is likely to be subjected to, emphasizes the con- 
cern of WG-3. Ziemann goes into well needed detail in describing handling of 
film in the camera and through the processing and drying stages; all of which 
can introduce distortions which are not in the optics or due to lack of platen 
flatness. 
Humidity, of course, changes the film dimensions and changes due to film 
handling (mishandling) can be significant, giving appreciable errors. 
Surveyors and film processors not conversant with these errors might de- 
velop individual methods of control by reviewing and gaining an understand- 
ing of the problem from Ziemann's paper. "Image Geometry - Factors Con- 
tributing to Its Change" delivered at the XII International Congress of 
Photogrammetry in Ottawa, Ontario, Canada, July 1972. 
20