In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Voi. XXXVIII, Part 7B 
A = 
&n + 2a n • x, + ■ x, a 4j + 2a s . ■ x, + 3a 61 • y t 1 
a , n + 2i T„-* +3a 3 „-*/ a 4n +2a Sn -x, + 3a 6n ■ y* 1 
(7), 
0 = 
S,(x, + Ax’,y, + Ay>z GPS (x ] ,y,)-Az‘ 
s „( x . + Ax ‘.y„ + Ay* ) - z GPS (x,,y„) - Az‘ 
(8) 
and n the number of GPS points used for the fusion process. 
The vector \ contains the adjustments 
5 = 
dx 
dy 
v dz y 
(9). 
In Figure 5 the functioning of the algorithm is demonstrated. 
profile 
3. APPLYING THE ALGORITHM FOR THE THREE 
TASKS 
The algorithm described in the preceding section is 
implemented in the software which carries out the fusing tasks 
defined in the introduction. 
3.1 Fusing GPS-RTK points with ALS-TIN-Model 
This case is graphically shown in Figure 3 and the algorithm 
can be used directly. The key preparing task is here to find the 
closest ALS points for each given GPS point in order to model 
the surface in the vicinity of the GPS points. Due to the 
triangulation process, which was carried out before, this 
information can be easily drawn out of the TIN model internally 
stored during the runtime. 
3.2 Fusing profiles with ALS-TIN-Model 
In this case profile lines are available which are surveyed either 
by RTK-GPS or tacheometrie measurements. These profile 
models are composed of a number of lines. These lines are 
defined by precise measurement points which are linked 
together by straight lines. As shown in Figure 6, the points 
along the lines exhibit a much lower density than the ALS 
points. In a first step the shift parameters Ax, Ay and Az can be 
determined only on the basis of the measured profile points. In 
order to put more weight into the profile model it is advisable to 
interpolate additional profile points along the profile lines and 
use those for fusing. Here a linear interpolation is sufficient. 
641