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
(b)
Figure 10. Terrestrial LiDAR dataset segmentation results:
(a) without considering local point density variations and
(b) considering local point density variations
5. CONCLUSIONS AND RECOMMENDATIONS FOR
FUTURE RESEARCH WORK
In this paper, alternative methodologies have been presented for
the estimation of local point density indices. These methods try
to overcome the shortcomings of available methods by
considering the 3D relationships among LiDAR points and
physical properties of the enclosing surfaces. In the simplest
approach, the local point density index is estimated while
considering a pre-defined number of neighbouring points to the
point in question in 3D space normalized by by the area of the
circle that comprises these neighbouring points. Although this
approach is simple and computationally efficient, it does not
take the physical properties of the surfaces enclosing the
individual points into account. In the other approaches, the
planarity of the surfaces enclosing the LiDAR points is firstly
checked through eigen-value analysis or adaptive cylinder
definition. Then, the local point density indices are estimated
for the points belonging to planar surfaces. The main advantage
of the eigen-value analysis over the adaptive cylinder definition
is the efficient computation. However, this approach is not able
to filter out the neighbouring points which do not belong to the
planar neighbourhood of the point in question prior to the
estimation of the local point density index. In spite of its
computational burden, the adaptive cylinder definition
technique can provide more accurate point density estimations
by filtering out the outliers that do not belong to the planar
neighbourhoods before the computation of the local point
density indices. The other advantage of the this method is
directly providing the segmentation attributes through the
parameters of the best fitting plane through the points within the
defined adaptive cylinder.
In order to demonstrate the impact of considering the estimated
local point densities on the quality of. LiDAR data processing,
different cases have been discussed in which incorporating
these indices leads to major improvements in the derived results.
Future research work will focus on the quantitative evaluation
of LiDAR data processing outcome (i.e., boundary detection,
segmentation and classification) while considering the local
point density variations and comparative analysis of these
results with the results from other processing techniques.
ACKNOWLEDGEMENTS
This work was supported by the Canadian GEOmatics for
Informed DEcisions (GEOIDE) Network of Centres of
Excellence (NCE) (Project: PIV-SII72), the Natural Sciences
and Engineering Research Council of Canada (Discovery and
Strategic Project Grants), and TECTERRA. The authors would
130
also like to thank Dr. Jan Skaloud, EPFL (École Polytechnique
Fédérale de Lausanne), Switzerland for providing the airborne
LiDAR datasets.
REFERENCES
Bes| P. J. and Jain, R. C., 1988. Segmentation through
Variable-order Surface Fitting, IEEE Transactions on Pattern
Analysis and Machine Intelligence, 10(2), pp.167-192.
County, K., 2003. LiDAR digital ground model point density,
http//wwws.kingcounty. gov/sdc/raster/elevation/LIDAR, Dig
ital. Ground, Model, Point Density.htm/, KGIS Center,
Seattle, WA.
Danilin, I. M. and Medvedev, E. M., 2004. Forest inventory and
biomass assessment by the use of airborne laser scanning
method, example from Siberia, International Archives of
Photogrammetry, Remote Sensing and Spatial Information
Sciences, XXXVI (8/W2), pp. 139 — 144.
Elmqvist, M., Jungert, E., Persson, A., and Soderman, U., 2001.
Terrain modeling and analysis using laser scanner data.
International Archives of Photogrammetry and Remote
Sensing, XXXIV -3/WA, pp. 219-227.
Isenburg, M., Liu, Y., Shewchuk, J., and Snoeyink, J., 2006.
Streaming Computation of Delaunay Triangulations,
Proceedings of ACM SIGGRAPH'06, New York, USA, pp.
1049-1056.
Kim, C., Habib, A., and Chang, Y., 2008. Automatic generation
of digital building models for complex structures from LiDAR
data, The International Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences, XXVII
(B4), pp. 463 — 468.
Lari, Z., Habib A., and Kwak E., 2011. An Adaptive Approach
for Segmentation of 3D Laser Point Cloud, /nternational
Archives of the Photogrammetry, Remote Sensing and Spatial
Information Sciences, XXXVII-5/W12, Calgary, Canada.
Patias, P., Grussenmeyer, P., Hanke, K., 2008. Applications in
cultural heritage documentation, Advances in
Photogrammetry, Remote Sensing and Spatial Information
Sciences, ISPRS Congress Book, 7, pp. 363-384.
Raber, G. T., Jensen, J. R., Hodgson, M. E., Tullis, J. A., Davis
B. A., and Berglund, J., 2007. Impact of Lidar Nominal Post-
spacing on DEM Accuracy and Flood Zone Delineation,
Photogrammetric Engineering & Remote Sensing, 73(7),
793-804.
Shih, P.T, and C. M. Huang, 2006. Airborne Lidar Point Cloud
Density Indices, American Geophysical Union, Fall Meeting
2006, abstract # G53C-0919.
Uddin, W. and Al-Turk, E., 2001. Airborne LIDAR digital
terrain mapping for transportation infrastructure asset
management, in Proceedings of Fifth International
Conference on Managing Pavements, Seattle, Washington.
Vosselman, G. and Maas, H. G., 2010. Airborne and
Terrestrial Laser Scanning. Whittles Publishing, Scotland,
UK, 320 p.
eom rm m AN
^o PA A BY ee mT (8 8 BY AN 08 0 BY T AS 6 (0 FN Pt bed bs fund AN Pd AS o — 0 m oM ONY
th MN m.
