DBPapers
DOI: 10.5593/SGEM2016/B22/S10.118

LIDAR FILTERING

D. Badea, M. Gheorghe, L. Calin
Thursday 8 September 2016 by Libadmin2016

References: 16th International Multidisciplinary Scientific GeoConference SGEM 2016, www.sgem.org, SGEM2016 Conference Proceedings, ISBN 978-619-7105-59-9 / ISSN 1314-2704, June 28 - July 6, 2016, Book2 Vol. 2, 925-930 pp

ABSTRACT
LIDAR data set contains a great number of points not located on real terrain surface [4].
All this data represent the Digital Model of Reflective Surface (MDSR). In many
studies we need only the Digital Height Model (MDA) which can be obtained after a
filtering process. MDSR filtering is an important research topic and have been studied
by : Petzold et al. (1999) noted that filtering can be done by applying minimum filter
iteratively by changing filter size. Axelsson (2000) uses TIN. Lohman et al. (2000) use
dual-rank-filtering. Vosselman (2000) uses slope as criteria. Briese and Pfeifer (2001)
use hierarchical approach, Karsten Jacobsen is using a combination of geometric tests
together with linear prediction. [3]
Many problems occur when trying to eliminate points not belonging to terrain surface,
but on man made objects and/or vegetation [3]. Points located outside the bare terrain surface can be eliminated through many filtering methods. A special atention should be mantained in order not to eliminate terrain points, which can be easily mistaken as located on man made objects or vegetation.
In this paper I will use a simple math filter and test the results against one of the above
methods.

Keywords: DHM, DSM, filtering, interpolation, Lidar, terrain model.