Full text: Proceedings, XXth congress (Part 5)

     
  
  
   
   
  
  
  
  
  
  
  
  
  
  
   
    
   
   
   
    
      
    
    
    
   
    
  
    
      
  
   
   
   
   
  
  
    
   
  
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part BS. Istanbul 2004 
  
global object coordinate system (object reconstruction). This 
system is usually the same as the one in which the control 
points are defined. 
In the project described in this paper we were dealing with 
non-metric images. We did not even know the camera type. No 
control points were measured and very limited information was 
given for scaling the image block. Also the texture information 
in the overlapping areas was not sufficient to measure an 
adequate amount of tie points. And finally, there was no 
information available about the position from which the 
pictures had been taken. 
Thus the usual workflow of traffic accident evaluation could not 
be directly applied and was therefore modified (Figure 1). 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
Scannon- , Approdmabonsf. 
metric Photos Ext. Orientation | 
Import Images & First Block 
Create Pyramids . Adjustment 
Estimate Interior Adjustment T 
; ; Measure Tie and 
Orientation Refinement Control Features 
Y Data Snooping : 
Measure fi 
À True Block Introduce Fictitious 
 Tie-Points- Rescaling Observations 
Datum Definition | | | Final Hybrid 
(Arbitrary Scaling) - Adjustment 
  
  
  
  
  
Figure 1: Project Flow 
* 
2. PROJECT DEFINITION 
Four non-metric images depicting a motorcycle with a deformed 
front wheel fork were given (Figure 2a and Figure 2b). No 
control information (control points or orientation 
approximations of the camera positions) were available and 
there was no data available regarding the camera with which the 
images had been taken. 
      
  
o du A 
igure 2a: n-metric image (a 
Figure 2a: No t ge (a) 
  
Figure 2b: Non-metric image (b) 
The depicted motorcycle was involved in an accident. It had 
crashed frontally into a car's side. The assessor needed to know 
with what speed the motorcycle had been driving in the moment 
of the collision. The only clues to derive such information were: 
e angle a (see Figure 3) of a sharp bend that arose in 
the upper right front fork during the accident. 
e exact range between the front and rear wheel axes of 
the deformed motorcycle! 
If this angle and/or range could be precisely computed, a 
technical specialist would be in position to derive the force 
needed to emerge such a break and thus calculate the unknown 
driving speed. 
  
63l 
Figure 3: Bent off wheel fork 
Due to the complexity of the project and the limited information 
that was provided, a program with a very versatile mathematical 
background was needed, hence common off-the-shelf software 
products for close range photogrammetry did not seem suitable. 
The employed package should be able to handle hybrid 
adjustment techniques (including fictitious observation 
implementation). So, the project was carried out by using the 
program system ORPHEUS (Kager et al. 2002), which is based 
on the photogrammetric adjustment system ORIENT (Kager 
1993).
	        
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.