International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
  
determined by using reference points. The calibration 
procedure, as shown in chapter 3.2 delivers a relation between 
image coordinates (i,j) and object points (X,Y,Z). Now, all 
3D-LS distance data can be transformed in the panoramic 
coordinate system and by that its pixel position in the 
panoramic image can be computed. For this position the actual 
grey value of the panoramic camera is correlated to the 
instantanuous laser image point. Figure 6 depicts the 
combination of both images. The ground sampling at the wall is 
about 1.5 mm. The fact that colour shifts are not visible in the 
image verifies the calculated alignment between both systems. 
The strong red shift is caused by the black cable carrying board 
(s. Figure 3). This board causes poor 3D laser scanner data due 
to bad reflecting properties. 
  
  
  
    
  
  
Figure 6. Merged images of 3D-LS and Panoramic Camera 
4. POSLAS-PANCAM 
Before the POSLAS-PANCAM was used in field experiment 
the recording principle was studied in a laboratory environment. 
4.1  POSLAS-PANCAM in Laboratory 
The functioning of POSLAS-PANCAM (PLP-CAM ) was first 
verified by surveying the test field described in chapter 3. 
During this experiment a robot was used as a moving platform. 
As GPS reception was impossible in the laboratory the position 
data and orientation data were obtained from the camera 
tracking system ARTtrack2 which comprises two CCD 
cameras. The IMU measurement data were also recorded 
parallely. This means, redundant orientation information is 
available and the accuracy of the orientation system can be 
| Targetss s 
IMU NN 
     
Figure 7. PLP-CAM carried by robot 
510 
  
verified. Figure 7 shows the robot with PLP-CAM. Figure 8 
depicts one of the two tracking cameras. The robot is remotely 
controlled by a joystick. 
   
  
Figure 8. PLP-CAM robot and tracking camera 
The results of this experiments were used to develop algorithms 
to integrate the data of the three independently working 
systems. It could be shown, that the data sets can be well 
synchronized. Furthermore, the whole system could be 
calibrated by using the targets (s. Figure 3). 
4.1 POSLAS-PANCAM in the Field 
For a field experiment the PLP-CAM was mounted in a 
surveying van. The GPS-antenna of POS was installed on top of 
the vehicle. The car drove along the fassade of the Neue Schloss 
in Stuttgart (s. Figure 9). 
  
  
M in front of Neues Schloss Stuttgart 
As the range performace of the 3D-LS was too low, only image 
data of the CCD-line camera and the POS-Data were recorded. 
The left image in Figure 10 shows the recitification result on the 
basis of POS data alone. By applying image processing 
algorithms the oscillation can be reduced and a comprehensiv 
correction is achieved by using external laser scanner data 
recorded independently during another survey (s. right image in 
Figure 10). 
The high performance of the line scan camera is documented in 
Figure 11 and Figure 12. An heraldic animal at SchloB Solitude 
in Stuttgart was surveyed by PLP-Cam. The object is hardly 
recocknizable (s. Figure 12) from the original PANCAM data. 
However, after correcting the data a high quality image is 
obtained. The zoomed in part illustrates the high camera 
performance. 
Inte 
  
IEEE NU 
Figi 
  
The 
usir 
be « 
mal 
bec 
diff 
dev 
ver 
equ 
ver 
exp 
and 
GP