International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004 
    
  
      
tt 
x 
% 
2 _— Q 29 05 ue 
NO EL @ -05 t0 05 x 
NE ES & 05102 " 
serres a 
= wr 
É d we 
um € artt 
SP 
A emt = ge” - 
aid 
t wt 
er^ 
Figure 5. Height differences between DTMs obtained from 
2000 (photogrammetric), 2001 (Riegl LMS-Z210 laser scanner) 
and 2004 (Riegl LMS-Z420i laser scanner) surveys. Differences 
(in meters) are indicated by different grey values. A) 2001 — 
2000; B) 2004 — 2000; C) 2004 — 2001. 
  
mummC — emm un 
Q 5 10 15 20 25 
Meters 
  
  
  
(A) 
  
  
p.5 19 15 20 25 
Meters 
  
  
  
(B) 
Figure 6. Cross-section profiles concerning 2000, 2001, 2004 
surveys (location depicted in figure 4). 
4. EVALUATION OF THE EXPERIENCE 
The experience carried out, although not directly useful for the 
monitoring of the main body of the Cà di Malta landslide 
because of the reasons already mentioned (recent works that 
have artificially altered the surface of the landslide, as well 
confirmed by the profiles and volume difference calculated), 
have shown that the terrestrial laser scanning technology can be 
applied for this kind of application. 
The main advantage in comparison with the traditional 
topographic surveying is the large area surveyed in a very short 
time and with a very high level of detail, without having to 
chose a limited set of points to be monitored; this is of course of 
50 
great interest when the landslide behaviour is unknown. The 
same happens also, in different way, in respect to GPS 
measurements. The achieved precision and accuracy is in many 
cases satisfactory, but for this aspect the quality of 
segmentation operation is fundamental (low vegetation can 
represent a major obstacle in a lot of situations), and also the 
geometrical characteristics of the scan must be taken in account 
for cach case (large distances, laser beam angle in respect to the 
terrain, footprint size, etc. reduce the survey precision). 
In comparison with airborne photogrammetry and airborne 
lasers canning different considerations arise; in front of a 
general lower accuracy, these techniques do not suffer for the 
problems of intervisibility and masking of areas that can affect 
terrestrial measurements, and provide data for larger 
investigations, more casily integrable with other information. 
On the other hand, airborne surveys are expensive and their 
execution highly dependent also on meteorological conditions 
and organizational context. The time needed for production of 
the results is in general longer (photogrammetry), and different 
skills required. 
If terrestrial laser scanning could represent a more flexible 
solution for small/medium size landslides that have to be 
monitored with high frequency to detect small movements, the 
high cost for the acquisition of the hardware and software must 
be considered, together with some problems for on-the-field 
operation: the remarkable weight of the instrument can pose 
some difficulties in movement and displacement on rugged 
areas, and power supply is an important question. For data 
processing, the single pulse files deriving from a terrestrial 
survey do not permit the adoption of established filtering 
procedures, requiring in any case a significant interactive 
editing by an experienced user. 
5. CONCLUSIONS 
The surveys realised over the studied area have demonstrated 
that in many cases Terrestrial Laser Scanners could represent an 
effective and rapid solution to produce economical and accurate 
terrain models for landslides investigation. 
Airborne surveying of landslide areas, traditionally done with 
photogrammetric methods, requires more expensive instruments 
(metric aerial cameras, photogrammetric scanners, digital 
photogrammetric workstations or analytical plotters) and a more 
complex organization framework. In effect there are at least six 
steps to obtain DTMs by means of photogrammetric techniques: 
aerial survey planning and execution; GCPs measurement; 
reproduction of the diapositives with scanner; determination of 
the orientation parameters with respect to an object space 
coordinate system; plotting of DTMs; controls and DTM 
editing. These steps require more time in respect to a simple 
TLS survey and data processing. 
Further more, even if analytic photogrammetry is being 
gradually replaced by digital photogrammetry, thanks to 
improvement of computer power and software, manual work of 
an expert operator is generally indispensable to generate a DTM 
in semi automatic mode or to correct errors coming from image 
matching procedures. Sometime automatic procedures are not 
feasible, in particular for zones characterized by vegetation 
cover, shadows or other obstacles. In theses cases only the 
interpretation by an operator can ‘produce high quality DTM 
data. 
Moreover, concerning laser scanning surveys, interesting is the 
possibility to registry DTMs realized in different periods also in 
the cases where classical ground control points are unavailable 
to reduce in an unique reference system the clouds of points. 
    
    
   
  
    
  
   
   
   
  
  
   
   
  
  
     
   
   
   
   
  
  
   
  
  
  
  
  
   
   
   
  
  
   
  
   
     
  
  
  
  
   
  
  
  
  
  
   
   
  
  
   
  
   
   
   
  
   
  
  
   
   
    
   
  
   
   
   
  
   
Interne 
Even | 
apply 
points, 
algorit 
data re 
freque 
body « 
curren 
scanne 
Acknc 
The au 
(Calen 
Gaisec 
(Horn, 
acquis 
The v 
Nation 
previsi 
emergi 
Refert 
Baldi 
landsli 
Europ: 
Nice, ] 
Barba: 
precisi 
enviro 
ASPR 
visuali 
Octob: 
Bes] P 
3-D s 
Intellis 
Gordo 
Deforr 
Procee 
Defort 
May, ( 
Licht, 
Based 
Geom 
Mora | 
E., Pe 
photog 
applic: 
Italy). 
RIEGI 
http:/A 
Pini G 
Scaglit 
of Am 
Pfeifer 
the S