e XXXIX-B3, 2012 
   
  
   
  
   
B] 25.05 Pnts/m' 
0.03 Pnts/m? 
E] 26.24 Pnts/m' 
0.004 Pnts/m’ 
(f) 
iginal LiDAR data, 
pproximate method, 
ilue analysis relative 
t density map based 
oint in question, 
'e cylinder method, 
on adaptive cylinder 
of LiDAR data while 
lensity indices, the 
using an adaptive 
. The segmentation 
isidering local point 
sult of the airborne 
idering local point 
s the result of the 
ng the local point 
n of the derived 
tion of Figures 8.a 
oint density indices 
n results, the most 
>m — as highlighted 
  
(b) 
>ntation results: 
y variations and 
variations 
has been obtained 
of Calgary campus 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B3, 2012 
XXII ISPRS Congress, 25 August - 01 September 2012, Melbourne, Australia 
using Trimble GS200 3D laser scanner. The local point density 
map of this dataset, estimated by the approximate method, is 
shown in Figure 9.b. The results of the planarity check for the 
individual points using the eigen-value analysis and adaptive 
cylinder methods are presented in Figures 9.c and 9.e, 
respectively. The point density maps for the points belonging to 
planar surfaces are then generated using estimated local point 
density indices (Figures 9.d and 9.1). 
  
  
  
(a) 
  
       
   
55919 Pnts/m* 
0.45 Pnts/m* 
  
  
BEEN Planar 
  
  
  
0.11 Pnts/m? 
  
  
E] 53220 Pnts/m* 
779] BESSS Non-Planar 
129 
  
t 
E Planar 
E | Non-Planar 
  
  
  
52641 Pnts/m* 
0.04 Pnts/m? 
  
  
  
Figure 9. Terrestrial LiDAR dataset: (a) original LiDAR data, 
(b) generated point density map using the approximate method, 
(c) planarity check result using the eigen-value analysis relative 
to the point in question, (d) generated point density map based 
on eigen-value analysis relative to the point in question, 
(e) planarity check result using the adaptive cylinder method, 
and (f) generated point density map based on adaptive cylinder 
method 
To assess the impact of considering the estimated local point 
density indices on the quality of terrestrial LiDAR data 
segmentation results, this dataset is segmented using the cited 
segmentation approach. The segmentation process is performed 
with and without considering local point density variations. 
Figure 10.a shows the result of the terrestrial LiDAR data 
segmentation without considering local point density variations 
while Figure 10.b shows the result of the segmentation of the 
same data while considering the local point density variations. 
The comparison of the derived segmentation results with and 
without considering the estimated local point indices 
demonstrates that considering the local point density indices 
avoids the over-segmentation problems in the segmentation 
results.