anbul 2004
| Set on top
lever arms
enna were
tion of the
acy a small
ble, with a
oto flight at
ye realized,
. The 1.3 h
ition of the
st field was
grammetric
is acquired.
> flight took
c the GPS
t ‘step an
0 determine
MU. Based
its between
ition of the
etween the
rmined. To
control 12
s and in the
of the first
ration site -
Cp
Ze
0.630
0.305
1.408
0.019
0.022
0.060
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B1. Istanbul 2004
Thereby it became obvious that there are systematic shifts
depending upon the flight direction between the angles of the
GPS/INS solution and the outcome of the aerial triangulation.
The shifts of +/- 0.1° in © and +/- 0.45? in « are most probably
related to the different definition of the rotations matrices of the
Applanix system and the ERDAS LPS Software, Cramer &
Stallmann, 2002. The recalculation of the rotations is currently
under way.
Additionally the GPS/INS data provide valuable information to
compare them to the common approach of PFIFF, based solely
on the GPS-information with linear interpolation to the
perspective centres. The time delay between the trigger impulse
and the image exposure is 71 ms, + 0.5 ms. The GPS data was
processed with the Leica Ski Pro: 3.0 software. A comparison
with the integrated GPS/INS data was done at 133 recorded
events. Thereby the positions at the full GPS-second prior to the
recorded trigger event and the XYZ-position at the event itself
were compared, see table 5a. Table 5b shows the influx
(residuals) of the linear interpolation onto the perspective
centres between two GPS-seconds, compared to the GPS/INS
positions.
Table Sa: Differences between GPS and GPS/INS (n = 133).
X y 7
Avg. -0.006 -0.003 -0.553
Std. 0.106 0.061 0.109
Max. 0.177 0.135 -0.258
Min. -0.139 -0.119 -0.697
-0.007
Table 5b: Residuals between GPS and GPS/INS at the
interpolated perspective centres (n = 133).
Avg. -0.001 0.000 0.000
Std. 0.012 0.017 0.007
Max. -0.019 -0.038 -0.023
Min. 0.028 0.060 0.018
The standard deviation of X and Y in table 5a is due to the fact
that the GPS-antenna on top of aircraft was not placed exactly
above the camera. The differences in Z are still under
investigation. The very small residuals in table 5b demonstrate,
that a simple linear interpolation yield precise information of
the perspective centres.
4. CONCLUSIONS AND OUTLOOK
The most important factor for aerial image acquisition is the
reliability of all components, especially for larger projects or for
commercial use. This is some what in contradiction to a
university development where every now and then new
components are added or changed. Nevertheless PFIFF has been
proven to be a reliable system for digital airborne data
acquisition. Currently the bottleneck for high resolution surveys
is the number of images with a total of several gigabytes which
have to be processed and geocoded in a timely manner. This
may be overcome with direct georeferencing. The test flight
revealed that direct georeferencing does not only work high end
photogrammetetric systems but also with digital SLR-cameras.
For low cost systems the direct georeferencing provides
additional flexibility for image acquisition, especially for a
combination of nadir looking and oblique images.
With direct geocoding small digital systems such as PFIFF will
become a competitive alternative due to some special advances
compared to large digital systems such as the DMC or ADS 40.
An important goal for future hardware developments is the
construction of an active mount, which compensates the
difference between the true heading and navigated direction of
the aircraft. The active mount will also reduce the need for large
side laps in urban areas and make the photo flights even
smoother than today.
ACKNOWLEDGEMENTS
Special thanks to Frank Lehmann and Dr.-Ing. Sergej Sujev,
DLR-Berlin, Adlershof for the possibility to use the Applanix
410 and the processing of the GPS/INS data.
5. REFERENCES
Báumker, M. Brechtken, R. and Heimes, F.-J. 1999. Direkte
Georeferenzierung mit dem Luftaufnahmesystem | LEO.-
Internationale Geodátische Woche, Obergurgl 22.02.1999, 10 p.
Cramer, M. and Stallmann, D. (2002): System calibration for
direct georeferencing — International Archives on
Photogrammetry and Remote Sensing IAPRS, Vol. XXXIV,
Com. III, Part A: 79 — 84, IRSPS Commision III Symposium
Graz, Sept. 2002.
Franke, J. and Montgomery, B. 2000. Experiences with a small
format imaging system integrating airborne DGPS.- 20.
Wissenschaftlich-Technische Jahrestagung der DGPF, Berlin
11.-13.10.2000 (= DGPF Publikation der DGPF Band 9), pp.
245 — 255
Grenzdórffer, G. 2002. Konzeption, Entwicklung und
Erprobung eines digitalen integrierten flugzeuggetragenen
Fernerkundungssystems für Precision Farming (PFIFF).-
Deutsche Geodátische Kommission, Reihe C: Vol. 552: 142 p
(PhD-Thesis)
HRSC-TEAM 2003. March. High Resolution Stereo Camera —
Airborne Extended (HRSC-AX). Available http://solarsystem.
dlr.de/FE/ hrsc.shtml
Hinz, A. 1999. The Z/I Imaging Digital Aerial Camera System.-
In: Fritsch, D. und Spiller, R. [Eds.]: Photogrammetric Week
'99.- pp. 109 - 115; Wichmann Verlag.
Holm, M and Rautakorpi, S. 1999. Experiences of automatic
creation of image mosaics and digital surface models using
airborne digital camera data. In Videometrics VI, S.F. EI-
Hakim, A. Gruen, Eds., Proceedings of SPIE Vol 3641-14
(Videometrics VI, 28-29.1.1999, San Jose, CA, USA), pp. 139-
150,.1998.
Mostafa M.M.R. and Schwarz, K.-P. 2000. A multi-sensor
system for airborne image capture and georeferencing.- PE &
RS Vol. 66, No. 12; pp. 1417 - 1423.
Thom, C. and Souchon, J.-P. 1999. The IGN digital camera
system in progress.- In: Fritsch, D. und Spiller, R. Eds.:
Photogrammetric Week '99.- pp. 89 - 94; Wichmann Verlag.
Toth, C. 1999. Experiences with frame CCD arrays and direct
georeferencing.- In: Fritsch, D. und Spiller, R. Eds.
Photogrammetric Week '99: 95 — 108; Wichmann Verlag.
