In: Wagner W., Szgkely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
The data of the 16 TLS stations was automatically co-registered 
using asymmetrically distributed reflectors. The total TLS 
point-cloud was geo-referenced using 28 absolutely determined 
reflectors within the Riegl software RiScanPro (Riegl, 2010). 
As a result both data sets were finally geo-referenced in the 
Austrian co-ordinate system GK-M34. This process of geo- 
referencing could be checked by the comparison of identical 
surfaces that were observed by ALS and TLS (e.g. some roof 
faces). Additionally, any possible presence of errors can be 
inspected by a 3D visualisation of both ALS and TLS point 
clouds. This is, however, impractical: due to the high point 
density, individual ALS and TLS points can hardly be 
recognized. Only profile views with clipped point clouds were 
suitable to visually inspect the resulting merged point cloud. As 
displayed in Figure 3, the ALS and TLS data set fit well 
together (the tree trunks and branches are located at the same 
position in both data sets). No discrepancies could be detected 
visually. However, a quantitative study of the accuracy has to be 
done as a future step in the project. 
4. ANALYSIS OF THE FWF-ALS DATA 
Based on both geo-referenced data sets, the position of each 
individual FWF-echo together with the derived FWF- 
parameters can be studied. This analysis should allow to 
increase the knowledge of the interaction of the laser beam with 
the different surface elements. Based on these studies, we see 
further potential for advanced classification for DTM generation 
and object separation. 
In the first step of the analysis, individual ALS points can be 
viewed together with the recorded respectively digitised 
waveform, while the context of the object is provided from the 
TLS data (see Figure 4). In Figure 4, the ALS-derived terrain 
points below the tree follow the terrain points of the TLS data 
set and have a very narrow echo width. The same is true for 
those ALS points, which have been reflected from the lower 
stem. ALS points, which are visible on the main branches, 
coincide perfectly with the TLS points. This emphasizes the 
high quality geo-referencing of both data sets. When looking on 
the respective waveforms one can typically find one strong echo 
resulting from the extended target at the tree surface. When 
looking at the echo width a slightly broadened echo (caused by 
the locally sloped surface) can be found. 
On the smaller braches on top of the tree typically more than 
one echo (up to five) can be seen in the waveform display. 
Some of the echoes are very close or even overlaid. Some of the 
echo widths are broadened, especially in areas with very dense 
and thin branches. Concerning the vertical ALS point 
distribution one can see that the top canopy layers are very 
densely covered and that many of the ALS echoes are located 
on thin but dense branches. 
All in all the (big) tree is represented well in the ALS point 
cloud, just small branches below the very dense canopy layer 
are not covered by ALS points. Holes in the TLS data are 
caused by shadow or by the profile selection. 
In order to study the sequence of echoes that are the result of a 
single emitted laser pulse, 3D visualisations with connecting 
lines were generated (Figure 5). In the upper part of Figure 5 
ALS echoes that result from the same line of sight are 
connected by a grey line. In the middle part of the figure the 
FWF information (in this case split into two graphs) of the 
selected line of sight can be inspected. It can be seen that the 
echo amplitude of the tree echoes caused by the tree top differs 
significantly. While the second echo is quite weak (just a little 
bit above the detection threshold) the third echo is even stronger 
than the first. This might be explained by the vegetation density. 
It seems that the third echo results from a thicker branch than 
the first echo, while the second echo must be caused from a 
very small branch. In the second graph of the FWF information 
one can see that the first local maximum that is visible in the 
FWF signal was not accepted in the echo detection step due to 
its low amplitude. The amplitude of the detected fourth echo is 
also very low (even lower than the second echo). The last echo 
of this line of sight has the highest amplitude and my result 
from an extended target (compare lower part of Figure 5). 
Figure 5. Inspection of all FWF-ALS echoes from one line of 
sight (red points) and TLS point cloud (green).