International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012 
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia 
4.2 Calibrated amplitude reading 
Beside geometrical information, the VZ-line additionally 
delivers a calibrated amplitude information for each target 
echo, which can be used to improve the object classification. 
The amplitude reading is defined as the ratio of the actual 
detected optical amplitude of the echo pulse versus detection 
threshold (eq. 1). The ratio is stated in decibels (dB). 
D ( 1 ) 
thr 
A, =10-log 
where 
Agp = calibrated amplitude [dB] 
Pecho 7 echo signal power [W] 
Pa, — power detection limit [W] 
4.3 Calibrated reflectance reading 
With the VZ-series of laser scanners, Riegl has introduced 
some measure of the laser radar cross-section by providing a 
reflectivity value for each target echo. The reflectivity value 
gives the ratio of the actual optical amplitude versus the 
optical amplitude of a diffuse white target located at the same 
range (it is furtherly assumed that the white target is larger 
than the laser footprint, 100% reflecting, flat, and its surface 
is orthogonal to the laser beam). The value is again given in 
dB (eq. 2). 
Pre (R7) = Ay (R, Y Ag white (R,) Q) 
where 
Ry = target Trange [m] 
Pra (Rz) = calibrated reflectance of target T at range Ry 
Agp 1 = calibrated amplitude of T at range Ry [dB] 
Agp white - amplitude of white target T at range Rr [dB] 
White diffuse targets would have assigned about 0 dB, while 
black paper with a reflectance of 10% would give about -10 
dB. Reflectivity values above 0 dB indicate that the target 
returns an optical echo amplitude larger than those of a 
diffuse white target, ie, the target is (partially) retro- 
reflecting. By assigning a brightness value to each point of a 
point cloud according to the reflectivity, objects with the 
same reflectivity will have the same brightness regardless of 
the distance to the scanner. Note that this applies only to 
single and first targets, not for the succeedings ones located 
along the laser beam axis. Indeed, except for single and first 
targets, the response of each subsequent target hit by the laser 
beam is a combination of the respective laser radar cross 
section and of the attenuation of the laser pulse due to the 
preceding targets. Since this target-induced attenuation 
cannot be retrieved from the amplitude of the backscattered 
pulse, actually the calibrated reflectance readings are 
representative of the backscattering properties of just the 
single or first targets acquired by a multi-echo digitizing 
system. An example of calibrated amplitude and calibrated 
reflectance is shown in figures 3(a) and 3(b), while a small 
portion of a VZ-400 scan coloured according to the recorded 
multiple echoes is displayed in figure 4. 
514 
  
  
(b) 
Figure 3: Front view of the test area rendered according to the 
calibrated amplitude (a) and the calibrated reflectance (b). 
  
Figure 4: Side view of a small portion of the landslide 
surveyed with the Riegl VZ-400 laser scanner. Green color 
denotes single targets, yellow first targets, light blue 
intermediate targets and blue last targets. 
5. VEGETATION FILTERING 
The data collected in the same subarea of the Brustolé 
landslide with the Riegl LMS-Z620 and the VZ-400 laser 
scanners were properly filtered in order to remove as much as 
possible the dense vegetation. The resulting DTMs were then 
compared to assess the potential of a multi-target TLS system 
with online waveform processing capability, such as the 
Riegl VZ-400, for better discriminate between vegetation and 
terrain. A slight different processing workflow was applied to 
the collected datasets. Firstly, the VZ-400 scan, covering the 
area of interest, was pre-filtered by exploiting the multi-target 
capability and the "calibrated relative reflectance" readings 
provided by this TLS. Out of the four types of echoes 
recorded by the instrument, only single, first and last targets 
were extracted from the original point cloud. Intermediate 
("others") targets were discarded as they were most likely 
originated by non-ground points. Then a subsequent filtering