stanbul 2004 
°NES). 
  
  
Yofen are: 
terskirchen, 
ith a point 
it 5: km x5 
m spacing) 
part, height 
'om contour 
of about 5 
n x 30 km) 
ut 2 meters, 
id geodetic 
ence DEMs 
Table 2. Main characteristics of reference DEMs. 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
Location e DEM | Source | DEM 
i $ Spacing Size E 
*t (m) (kmxkm)| € x 
ROS ep 2 
55 23 
Fr = 3 
9 
= < 
Prien |Smooth,| 5x5 Laser 5x5 0.5 
weakly Scanner 
inclined 
Gars |Smooth,| 5x5 Laser 5x5 0.5 
weakly Scanner 
inclined 
Peters- | Smooth, | 5x5 Laser 5x5 05 
kirchen | weakly Scanner 
inclined 
Taching | Smooth, | 5x5 Laser 5x5 0.5 
weakly Scanner 
inclined 
Inzell- | Rough, | 25x25 | Laser 10x1.3 0.5 
North | strongly Scanner 
inclined 
Inzell- | Rolling, | 25x25 | Contour | 10x7.7 | 5.0 
South | strongly lines . 
inclined 
Vilsbi- | Rough, | 50x50 | Photo- | 50x30 2.0 
burg | weakly gramme- 
inclined try 
  
  
208 
. SPOT-S/HRS forwar 
BE 
d scene with some 
  
  
elements in white (towns, rivers and lakes) and 
location of reference DEMs (in black) with same 
notation used in Table 2. 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part Bl. Istanbul 2004 
3. POINT MEASUREMENTS 
From the available 81 object points, only 41 have been 
identified in the images. In order to locate them in the images a 
digital map at 1:50,000 scale (Topo50) was used. The 
coordinates were given in the Gauss-K rueger system. 
The exact image coordinates of the points have been measured 
with unconstrained Least Squares Matching developed at IGP 
(Baltsavias, 1991), by measuring the points in the master image 
manually. The final point distribution is shown in Figure 3. 
  
4. IMAGES ORIENTATION 
The HRS (High-Resolution Stereoscopic) instrument of SPOT- 
5 uses linear arrays that scan a single image line at an instant of 
time in the so-called pushbroom mode. Consequently each line 
of the HRS image is acquired at a different exposure station 
with different orientation elements. For the orientation of this 
kind of imagery two approaches, based on rigorous models and 
rational function models, are used. 
The rigorous model tries to describe the physical properties of 
the sensor and its image acquisition mode. It is based on 
collinearity equations, which are extended in order to describe 
the specific geometry of pushbroom sensors. The adjustment 
parameters must include the exterior orientation and self- 
calibration parameters to describe the physical imaging process. 
Alternatively, rational function models use a general 
transformation to describe the relationship between image and 
ground coordinates. 
In this work both approaches have been applied. In the next 
paragraph the algorithms used for the orientation will be 
described and the results reported. 
4.1 Procedure 1: Rigorous model 
The aim of rigorous sensor models is to establish a relationship 
between image and ground reference systems according to the 
sensor geometry and the available data. For the georeferencing 
of imagery acquired by pushbroom sensors many different 
geometric models of varying complexity, rigor and accuracy 
have been developed, as described in (Fritsch et al., 2000) and 
(Dowman et al, 2003) A flexible sensor model for the 
georeferencing of a wide class of linear CCD array sensors has 
been developed at IGP and already applied to different linear