International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
Intern« 
5. ORTHOIMAGES ACCURACY EVALUATION features in three different areas of the images: in the plain area, LIDA] 
: both in Quindici village (top image in the figure), and in georef 
Orthoimages obtained from the two different processing have Episcopio (central image), and in the mountainous area (bottom of 4] 
been compared through digitalisation of significant features image). The displacement between the orthoimages is very resolu 
located in areas where both DTM were available. variable over the whole area: in the plain areas it is around 4 - 5 points, 
It was not possible to evaluate accuracy through ground check pixels (around 3 meters), in the mountainous area it decreases to 9). 
points because of the lack of a significant amount of points 
univocally recognisable on the image. In fact the GPS survey 
was performed just after the emergency when QuickBird images 
were not yet available 
  
  
Figure 6. Features digitalisation over the orthoimages. Red 
lines are relative to orthorectification with integrated DTM, the 
cyan lines to orthorectification with IGM DTM. The base image 
is obtained from integrated DTM. 
Figure 6 shows a visual comparison between the orthoimages 
obtained. The evaluation has been carried out by digitizing 
one pixel (around 60 cm).Actually, the lack of recognizable 
features in mountainous zones doesn't allow to have significant 
elements to compare. 
6. INTENSITY LASER DATA INTEGRATION 
As previously said, laser scanning records both the height and 
the intensity data of the hit points. Intensity LIDAR data are 
collected in the near infrared part of the electromagnetic 
spectrum. The value of intensity is normalized so it depends 
only from the reflectance of the object and not from the 
reflection angle. 
LIDAR intensity is expressed in percent value; in the first 
LIDAR survey the intensity values of all points varies about 
from 0% to 45% (Bracigliano), while in the second one about 
from 0% to 25% (Sarno). This is caused by the different height 
of flight, different foot-print size, different scan angle and 
unfavourable atmospheric conditions. 
In the area of landslide of Bracigliano, the intensity of first 
echo, that represents most of all canopy mass, ranges from 0% 
to 25%; while the intensity of last echo, that mostly 
represents the terrain, varies from 0% to 50% as the histograms 
show in figure 7. 
x 10 
   
   
   
œ 
Absolute Frequency 
o 
  
lu. JA atento i 
5 10 15 20 30 35 40 
29 
5 
LIDAR Intensity - First Echo - Bracighano 
25- 
Absolute Frequency 
    
- 
rate 
05r 
   
{ i i À 
0 10 20 30 40 50 60 70 80 90 100 
LIDAR Intensity - Last Echo - Bracigliano 
Figure 7. First and Last Echo Laser Intensity Histograms 
  
Figu 
Raster 
vegetai 
strengt 
reflect 
echo a 
be use 
laser p 
As the 
infrare 
results 
distanc 
sample 
Howey 
detach 
identif 
ground 
second 
rather t 
The 
panchr 
resolut 
raster. 
948