Full text: Proceedings, XXth congress (Part 1)

  
    
    
  
  
  
   
  
  
  
  
  
  
   
       
   
  
   
   
    
  
   
   
    
    
    
    
  
   
   
    
   
   
   
    
  
  
  
   
     
   
   
     
    
   
   
   
  
    
  
  
   
  
   
    
  
   
   
   
  
    
     
     
       
Istanbul 2004 
ig tasks (e.g. 
> properties 
the operating 
is a Bell 206 
'Xto-electronic 
a systems in 
onfigurations 
mbination of 
Therefore, a 
» fulfil these 
ric dynamics 
nnel. As an 
petween 256 
)96 different 
scomes clear 
1 areas, very 
nage due to 
urface (e.g. 
idows). An 
sary. 
ling to the 
bout 0.2 Hz) 
car velocity 
have to be 
2s). 
a number of 
ment of the 
data transfer 
stems were 
ining traffic 
nera system 
of a typical 
  
  
     
x 1079 
| 
figuration 
re mounted 
'k mounts 
| (airplane). 
target area 
remaining 
'en for long 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B1. Istanbul 2004 
   
X is E kA. ue 
Figure 2. Integration of the equipment (IMU, VIS-Camera and 
Camera-Laptop) in a helicopter 
The influence of image compression was partly investigated for 
the projects up to now. This topic remains to be considered if 
data transfer rate limits image acquisition. So for the project 
LUMOS an image compression with fixed factor of 2 was 
applied without significant influence of the quality of image 
processing. 
Parameters describing pixel positions on the focal plane (so- 
called interior orientation) are necessary for georeferencing of 
the image data. They were determined during a calibration 
process in optical laboratories at DLR. 
3.2.2 Infrared camera 
Cameras working in the infrared range of the electromagnetic 
spectrum have the advantage to be applicable even at night. In 
most cases the spectral texture in the infrared allows an easier 
image data interpretation as with visible cameras. The main 
disadvantages of infrared cameras are the small number of 
pixels (so the swath or the ground resolution has to be reduced) 
and high costs. Table 2 shows the parameters of the used 
thermal IR camera. 
  
  
Parameter Value 
Detector MCT, cooled at 77?KIR 18 
Type MK III, Barr & Stroud ltd. 
  
768 x 500 
15,28? x 10,20? 
Number of pixels 
Field of view 
  
  
  
  
  
  
  
  
Radiometric dynamics 8 Bit 
Spectrum 8 — 14 um 
Frame rate 28 Hz 
Ground sampling 0.5 m 
distance, flight height 
3500ft 
Swath width 380 m 
  
Table 2. Parameters of a typical IR camera configuration 
3.3 Inertial measurement unit 
Onboard or real-time georeferencing requires the exact 
knowledge of all six parameters of the so-called exterior 
orientation (three translations x, y, z, three rotations 0,0,' 
of the camera for each snapshot. Depending on the desired 
accuracy of data products, these parameters have to be 
determined with an accuracy in the range of one ground pixel 
distance and one instantaneous field of view (field of view for 
one pixel). 
A typical technical solution for direct and precise 
measurements of the exterior orientation parameters of the 
camera during imaging is to use an integrated GPS/Inertial 
system, which combines efficiently inertial sensors technology 
and GPS technology. DLR owns such a system (POS-AV 410 
of Applanix Corp.), which fulfils the required specifications 
(Lithopoulos 1999, Scholten et al 2001). The system consists of 
an Inertial Measurement Unit (IMU) LN200 and a control unit 
with integrated GPS receiver. The IMU sensor head was 
mounted close to the camera. The IMU realizes measurements 
of accelerations and angular velocities and records movements 
of the camera/IMU. The time synchronized IMU and GPS data 
are processed in a control unit within an inertial navigation 
algorithm. The system provides real time output of position 
and orientation with a rate up to 200 Hz. In combination with a 
differential GPS correction, an absolute accuracy for position 
of 0.5 to 2 meters and for attitude of 0.015 to 0.05 deg can be 
obtained. 
Image and attitude data have to be recorded synchronously. 
Therefore, Applanix' trigger pulses were monitored and used 
for commanding the image acquisition process. 
3.4 Data Transmission 
Onboard captured images and related orientation data are sent 
to the ground station via radio transmission. The data 
transmission channel consists of the transmitter and antenna 
onboard and receiving station on ground. The data transmission 
rate is the most limiting factor for the specification of the real 
time airborne monitoring system. It defines the maximum 
image acquisition rate. The transmission rate values were 2 
and 4 Mbps for LUMOS and Eye in the Sky configurations 
respectively. 
3.5 Ground Station 
Image data received on the ground station are transmitted to 
the traffic computer for image processing. Within the LUMOS 
project a special network of three ground stations with a 
distance of about 25 km was developed and installed to ensure 
the coverage of whole Berlin-City with surrounding area. 
Directional radio link lead the data from each station to the 
server, where the best of the three data packages is selected 
and provided to the image processing. For the project “Eye in 
the Sky” another technical realisation was applied. It is based 
on a digital transmission system with a mobile GPS antenna 
placed on the roof of the building for tracking. 
4. DATA PROCESSING 
4.1 Direct georeferencing 
The real time orientation data stored in a control PC describe 
the actual position of the camera. This position is given by 
longitude, latitude and ellipsoid height with respect to the 
geodetic datum WGS84 and the rotation angles of IMU 
measurement axes given by roll, pitch, and heading with 
respect to the local tangential coordinate system. 
The misalignment between the IMU and camera axes (bore 
sight angles) has to be estimated offline once per system 
installation in the airborne platform using the traditional aero 
triangulation method. 
During the measurement flights, real time orientation data, 
misalignment angles and camera parameters (interior 
orientation) define a transformation from image space to object 
space and vice versa. Assuming a medium terrain height, the 
position of the vehicles can be estimated. Consequently, for 
each pixel of interest a (x,y)-tuple can be determined and each
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.