noise. The presented method complies with theoretical
expectations; only one tree shows a different behaviour
(continuous decrease of the number of filled voxels).
DBH DBH Diff. Height Height Diff.
Tree | contr. estim. [cm] contr. | estim. ie
[m] [m] [m] | [m]
Al 0.41 0.42 - 0.01 15.3 14.8 +05
A2 0.44 0.42 +002} 155 14.9 + 0.6
A3 0.38 0.38 0.00 | 12.6 119 +07
C1 0.25 0.26 - 0.01 14.7 13.4 +1.3
RI 037 0.37 0.00 | 10.4 9.9 OS
R2 0.73 0.65 +0031 187 17.9 +0.8
Bl 0.42 0.42 0.00 --- 14.1 ---
TI 0.42 0.41 * 0.01 11.6 10.0 251.6
T2 0.33 0.32 t 0.01 10.6 10.6 0.0
Table 2. Estimated and control DBH and tree heights
In addition to tree volume, DBH and the height of trees — often
used as variables for biomass functions — are calculated (Table
2). Estimates of DBH correspond to the measured control
values with only marginal deviations (with one exception). For
all trees, height determination is systematically lower than
related manual measurements. Approximately, the differences
lie between 0.5m and 1.5m caused by the uncertain acquisition
of the highest part of crowns from terrestrial position — by TLS
as well as by manual measurement.
6. CONCLUSIONS
The presented method delivers good results in accordance with
the control data. It was found to be more accurate than other
methods such as detailed analysis of aerial images or
measurement of main branch perimeters by specialists climbing
the trees and predicting the connected branch volumes. The
usage of a voxel structure, the noise suppression and the
processing of the data 1n horizontal layers leads to a significant
data reduction and enables the processing of large data sets
consisting of 20 to 60 million points. The applied extended
algorithm for filling hollow surfaces is generally applicable for
stems as well as for complex crowns. The advantages of this
method are its simplicity and robustness, ie. no complex
Skeleton extraction or fitting of cylinders is necessary.
Moreover, volume extraction is independent of branch length,
diameter or orientation. It needs no advanced input information
and can be used for a wide range of applications. The
drawbacks of this approach are the necessity of high resolution
scans and the extensive computing time for determination of a
suitable threshold for noise reduction. In future studies, it is
suggested that the number of analysed trees be increased to
obtain a better statistical basis, to allow extrapolation to larger
areas. Additionally, further investigations have to be carried out
concerning the influence of the number and position of
necessary scans, possibly in dependence on the tree structure.
7. REFERENCES
Bucksch, A., Lindenbergh, R., 2008. CAMPINO - A
skeletonization method for point cloud processing. ISPRS
Journal of Photogrammetry and Remote Sensing, 63, theme
issue: terrestrial laser scanning, pp. 115-127.
Bucksch, A., Lindenbergh, R., Meneti, M., 2010. SkelTre —
Robust skeleton extraction from imperfect point clouds. The
Visual Computer, 26, special issue, pp. 1283-1300.
Côté, J.-F., Widlowski, J.-L., Fournier, R., Verstracte, M.,
2009. The structural and radiative consistency of three-
dimensional tree reconstructions from terrestrial lidar. Remote
Sensing of Environment, 113, pp. 1067-1081.
Dassot, M., Barbacci, A., Colin, A., Fournier, M., Constant, T.,
2010. Tree architecture and biomass assessment from terrestrial
LiDAR measurements: a case study for some Beech trees
(Fagus sylvatica) In: Silvilaser 2010, 10^ International
conference, 14-17 Sept. 2010, Freiburg, Germany.
Good, N., Paterson, M., Brack, C., Mengersen, K., 2001.
Estimating tree component biomass using variable probability
sampling methods. Journal of Agricultural, Biological and
Environmental Statistics, 6, pp. 258-267.
Gatziolis, D., Popescu, S., Sheridan, R., Kub, N.-W., 2010.
Evaluation of terrestrial LIDAR technology for the development
of local tree volume equations. In: Silvilaser 2010, 10^
International conference, 14-17 Sept. 2010, Freiburg, Germany.
Gorte, B., Pfeifer, N., 2004a. 3D image processing to
reconstruct trees from laser scans. In: Proceedings of the 10
annual conference of the advanced school for computing and
imaging (ASCI), 2-4 June 2004, Ouddrop, the Netherlands.
Gorte, B., Pfeifer, N., 2004b. Structuring laser-scanned trees
using 3D mathematical morphology. In: Proceedings of ISPRS
congress Istanbul, Comm. V, Vol. XXXV, Part B5, Istanbul,
Turkey, pp. 929-933.
Kronijnendijk, C., Nilsson, K., Randrup, T., Schipperijn, J.,
2005. Urban Forests Trees: A Reference Book, Springer,
Berlin, Heidelberg.
Lefsky, M., McHale, M., 2008. Volume estimation of trees with
complex architecture from terrestrial laser scanning. Journal of
Applied Remote Sensing, 2, 023521, May 14, 2008.
Norra, S., 2009. The astysphere and urban geochemistry — a
new approach to integrate urban systems into the geoscientific
concept of spheres and a challenging concept of modern
geochemistry supporting the sustainable development of planet
earth. Environmental Science and Pollution Research, 16, pp.
539-545.
Pfeifer, N., Gorte, B., Winterhalder, D., 2004. Automatic
reconstruction of single trees from terrestrial laser scanner data.
In: Proceedings of ISPRS congress Istanbul, Comm. V, Vol.
XXXV, Part B5, Istanbul, Turkey, pp. 114-119.
Pfeifer, N., Winterhalder, D., 2004. Modelling of tree cross
sections from terrestrial laser scanning data with free-form
curves. In: Proceedings of ISPRS working group VIII/2, Vol.
XXXVI, Part 8/W2, 3-6 Oct. 2004, Freiburg, Germany, pp. 76-
81.
Raumonen, P., Kaasalainen, S., Kaasalainen, M., Kaartinen, H.,
2011. Approximation of volume and branch size distribution of
trees from laser scanner data. In: ISPRS workshop
Laserscanning 2011, 29-31 Aug. 2011, Calgary, Canada.
Xu, H., Gossett, N., Boaquan, C., 2007. Knowledge and
heuristic-based modeling of laser-scanned trees. ACM
Transactions on Graphics, 26, 19.
Zianis, D., Mencuccini, M., 2004. On simplifying allometric
analyses of forest biomass. Forest Ecology and Management,
187, pp. 311-332.
ABS
Fores
supp
Conv
COVE
a Vic
repre
cons;
com]
objec
by a
situ «
inten
Sust:
prese
(Gar
indic
impc
1998
fores
whic
capa
Nati
cano
of f
Agri
LAI
canc
defi
isa
iden
mea;
quar
land
‘Car
Cove
et a
Foli
