1051 
RESULTS OF A PERFORMANCE TEST OF A 
DUAL MID-FORMAT DIGITAL CAMERA SYSTEM 
Jens Kremer a ’ *, Michael Cramer b 
a Ingenieur-Gesellschaft fur Interfaces (IGI), 
Langenauer Str. 46, 57223 Kreuztal, Germany, J.Kremer@igi-systems.com 
institute for Photogrammetry (ifp), Universitaet Stuttgart 
Geschwister-Scholl-Str. 24D, 0174 Stuttgart, Germany, Michael.Cramer@ifp.uni-stuttgart.de 
Commission I, ThS-2 
KEY WORDS: medium format, dual-head camera, direct georeferencing, empirical test, geometric performance 
ABSTRACT: 
The low weight and the relatively low cost of medium format digital cameras have pushed the use of those units for aerial survey. 
These medium format cameras are often used as secondary sensors together with other aerial sensors like LiDAR systems. The 
rising number of pixels per camera leads to an increasing interest in medium format systems as main sensors, especially for smaller 
survey aircraft. While the number of pixels across flight direction is not critical for capturing linear objects, like power lines or 
pipelines, the relatively small number of pixels compared to large format systems increases the necessary flying effort for 
photogrammetric blocks. In case larger blocks have to be flown efficiently, it is possible to combine two or more of such medium 
format cameras (dual- or multi-head solutions). This combination of medium format cameras increases the possible image strip 
widths and therefore reduces the flying time and distance for block projects. A performance test of a dual-head medium format 
digital camera system flown over the Vaihingen/Enz test field of the Institute for Photogrammetry of Stuttgart University is 
presented. The operated Dual-DigiCAM-H/39 consisted of two 39Mpixel cameras. To increase the image width across flight 
direction, the two cameras were mounted to look to the side at an oblique angle of +14.8° and -14.8°, respectively. This 
configuration results in an effective image width of 13650 pixel. The dual camera system was operated together with a 
CCNS/AEROcontrol navigation- and GPS/IMU system. The GPS/IMU trajectory was processed with different GPS methods and the 
different trajectories are compared. The overall system performance was evaluated based on the analysis of independent check point 
differences. 
1. INTRODUCTION 
Compared to analogue film based cameras, large format digital 
aerial systems have many obvious advantages. Nevertheless, 
the introduction of these systems did not change two of the 
main challenges for the producers of aerial images. The present- 
day large format digital aerial cameras are of high weight and 
volume, which forbids the flexible use in smaller and cheaper 
aircraft like small single engine aircraft and ultra-light aircraft. 
Furthermore the relatively high price of those systems is a 
problem for smaller aerial survey companies and institutions. 
These reasons have pushed the development of medium format 
digital aerial cameras. Those cameras do not offer the same 
high number of pixels per exposure, but they are able to provide 
good image quality at lower cost and weight. While the smaller 
number of pixels across flight direction is not critical for 
capturing linear objects, like power lines or pipelines, it in 
creases the necessary flying effort for photogrammetric blocks. 
In case larger blocks have to be flown efficiently, it is possible 
to combine two or more of such medium format cameras (dual- 
or multi-head solutions). This combination of medium format 
cameras increases the possible image strip widths and therefore 
reduces the flying time and distance. 
In the following a performance test of such a dual-head medium 
format digital camera system is presented. The operated Dual- 
DigiCAM-H/39 consisted of two 39Mpixel medium format 
cameras mounted with an oblique angle of +14.8° and -14.8°, 
respectively. This configuration results in a width of 13650 
pixel and still provides a sufficient overlap in between the two 
neighbouring images. The two cameras were triggered 
synchronously within lOps by a CCNS4 navigation system. 
Together with the camera, an AEROcontrol-IId GPS/IMU 
system was operated to measure position and orientation of the 
cameras at the instant of the exposures. 
The test took place over the Vaihingen/Enz testfield of the 
Institute for Photogrammetry. This test field has 172 precisely 
measured and signalised 3-D ground control points. 
Additionally approximately 70 natural points are available. The 
test flight configuration consisted of three photogrammetric 
blocks flown at three different altitudes. This empirical test 
flight material allows a detailed and independent analysis of the 
overall system performance. 
The first part of this paper describes the used sensor system. 
In the second part, the effect of different GPS processing 
techniques on the accuracy of the trajectory determination is 
shown. 
Corresponding author