The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part BI. Beijing 2008 
1053 
20 double images and an overlap of p=60% / q=80%. The 20cm 
block consisted of three lines east/west with 20 double images 
and an overlap of p=60% / q=64%. 
The block layout was chosen to optimally fit to the distribution 
of ground control points in the Vaihingen/Enz test field. This 
causes the changes in the overlap conditions. The quite high 
side laps are not necessary for regularly production flights, they 
were only realized to obtain very large overlaps between 
individual camera head imagery for later stable overall system 
calibration. 
The individual block layout is shown in Figure 3. For the 7cm 
block only the three flight lines without cross strips are given. 
The figures also illustrate the corresponding distribution of 
control and check point information, where the later used 
control points are given in red (arranged in five control point 
chains perpendicular to main flight direction) and the remaining 
signalised points for independent accuracy check are coloured 
in grey. These check point configurations were utilised for the 
traditional bundle adjustments, based on control points only. In 
addition, direct georeferencing was performed, where all 
available signalised object points served as independent 
absolute accuracy control. 
The mission was conducted on December 19 th 2007 by Weser 
Bildmessflug GmbH, Bremerhaven, Germany. For this flight, a 
Cessna 206 survey aircraft was operated. 
Due to the date and time of this flight mission (19th December 
2007, 11:15 to 12:40 LCL) the sun angle was only between 17 
and 18 degrees that results in poor colour quality and long 
shadows. Bright sunlight and light haze gave also a reduced 
contrast on the outer edges of the frames (Figure 4). 
4. COMPARISON OF DIFFERENT GPS 
TRAJECTORIES 
Figure 4: Top: Image pointing to the left. Bottom: Image 
pointing to the right. 
The GPS and IMU processing was done with AEROoffice 5.1c 
from IGI and GrafNav 8.10 from Novatel Inc. Calgary, Canada. 
Three different trajectories were created. For trajectory A the 
GPS base station “0384 Stuttgart” from the German permanent 
reference station network SAPOS (www.SAPOS.de) was used, 
trajectory B used a virtual base station provided by SAPOS and 
trajectory C was processed using precise orbits and clock 
information with the precise point processing (”PPP”) method 
provided inside the GrafNav software (Kouba & Heroux, 2000). 
Base station “0384 Stuttgart” was located about 25 km to the 
south-east of the test field centre. The virtual base was 
calculated to be in the centre of the area. 
During the flight over the test field, the number of satellites was 
between 5 and 9 with an average number of about 7 available 
satellites. 
The GPS processing of trajectories A and B showed a 
difference between the forward and the reverse solution of max. 
3 cm for the horizontal, and of max. 13cm for the vertical 
component. For trajectory C these differences were 5cm and 
25cm, respectively. Based on the different GPS trajectories, 
GPS/IMU trajectories were processed inside AEROoffice. 
Figure 5: Comparison between solution A and solution B (base 
station vs. virtual base station). 
Figure 5 shows the position and attitude differences between 
solution A and solution B (base station vs. virtual base station).