Measurements were made on two RC5A cameras to explore the spectral
differences involved in the two calibration methods (figure l 1 ). Transverse colour
(chromatic difference of distortion) was measured at a number of field angles.
Results are shown in figure 2. Then the camera was tested on the photographic
calibrator with »mean noon sunlight» light sources and flattened Aero Super XX
plates. First the Wild 2X yellow filter was fitted. Next, no filter was on the
camera but collimator lights were screened with Wratten 13 filters to transform
the effective spectral sensitivity to about that of the human eye. Results are
shown in figure 3.
The differences in image position between the two calibrations are shown in
figure 4. For comparison, median wavelengths were obtained for curves B and C
of figure 1 and image shifts between these two wavelengths read from the
curves of figure 2, were plotted to give the comparison curve in figure 4. This
procedure is only a rough approximation to the theoretically correct one. How
ever, the agreement between the two curves is sufficient to confirm that the
differences found between the two calibration procedures can be attributed
chiefly to the chromatic differences of distortion of the lens as seen by two
spectrally different receivers.
These results emphasize the necessity of calibrating air survey cameras by a
procedure which closely simulates conditions of use. The error investigated here
arose primarily from the visual test procedure having the wrong chromatic
sensitivity. However, similar errors could arise from other departures from
normal conditions of use.
P. D. Carman tfc H. Brown: Unterschiede zwischen visueller und photographischer
Kalibrierung von Luftbildkainniern.
Die im Jahre 1955 vom National Research Council in Canada mit dem neuen photographischen
Kalibrierungsgerät durchgeführten Kontrollmessungen an Wild RC5A Kammern zeigten systema
tische Differenzen mit den auf einem visuellen Kalibrierungsgerät erlangten Werten der Herstellungs
firma.
Photographisch bestimmte Bildweiten überschritten gewöhnlich die vom Hersteller angegebenen
um 0.01 bis 0.02 mm. Die bei der Photographischen Kalibrierung erhaltenen Extremwerte für die
durchschnittlichen radialen Verzeichnungskurven lagen zwischen 10 und 17 Micron. Das sind im
Mittel 5 bis 6 Micron mehr, als die vom Hersteller angegebenen Werte.
Die Messungen wurden an zwei RC5A Kammern durchgeführt, um die spektralen Differenzen,
die durch die zwei Kalibrierungsmethoden bedingt sind, festzustellen (Abb. 1). In einer Anzahl
von Bildwinkeln (Abb. 2) wurden die chromatischen Differenzen der Verzeichnung gemessen. Dann
wurde die Kammer in der üblichen Art photographisch und auch in einer der visuellen Methode
entsprechenden Weise kalibriert (Abb. 3). Die Unterschiede in der Bildlage, bedingt durch die zwei
verschiedenen Kalibrierungsmethoden, stimmten überein mit den Werten, die man durch Berech
nung aus den gemessenen chromatischen Differenzen und den mittleren Wellenlängen von zwei
entsprechenden Kurven erhalten hatte (Abb. 4).
Diese Resultate betonen die Notwendigkeit, Kalibrierungen von Luftbildkammern unter annähernd
gleichen physikalischen Bedingungen durchzuführen, wie man sie später auch im praktischen
Gebrauch der Kammern vorfindet.
B. Hallert: A new
method for the
determination of the
distortion and the
inner orientation.
(Publ. I Hi)
B. Hallert: A new method for the determination of the distortion and the
inner orientation of cameras and projectors. The Grid Method.
The paper deals with the determination of some important systematic errors
in photogrammetric pencils of rays, and the complete determination of the inner
orientation of all kinds of cameras and projectors, X-ray instruments etc.
The method is founded upon the photography or projection of regular plane
grids, which are measured with high precision. The projective connection be
tween the grid points is used for the determination of the qualities of the central
projection in question. The points are combined in a special way and the method
of the least squares is used for the determination of corrections to the 6 degrees
of freedom of the pencils of rays in space from the discrepancies in the coordinate
measurements. Particularly the radial distortion and asymmetries of the pencils
1 The figures are not printed here.