| 2004
1. The
indard
Z4201
lution
icated
dslide
gether
Islide,
y and
oe for
that
9 map
luring
er the
DSMs
f each
JTMs
ice of
ware
odule
data
ut it’s
ccho
own
ity to
nce is
Irvey,
se of
errain
(data
Data
f the
owing
nding
utines
und".
ors on
iginal
arting
tified
of the
basis
ed on
irface
mally
rating
of the
ın be
1 also
data
s and
ferent
ce is
ng. In
from
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV. Part B5. Istanbul 2004
an area to another, with strong changes in slope and the
presence. of rock blocks that affect the geometric "surface
uniformity" characteristics.
In any case the results of the automatic procedures required a
successive interactive editing and this operation has been
assisted by the exploration of digital images acquired together
with laser data.
A = A £
(A)
A
(B)
N
pde
9 25 50 75 100 125
Meters
(C)
Figure 4. Example of laser scanning data filtering: A) Cross-
section profile A-A' before data filtering; B) The same profile
after removing vegetation. C) original cloud of points (2004 and
2001 surveys) with location of profiles.
3.3 Alignment of the multitemporal DTMs
To investigate the multitemporal evolutions of the landslide, the
results of the two surveys were reduced in a unique reference
system using PolyWorks software by InnovMetric. After
manual selection of some homologous points, Poly Works aligns
simultaneously all the data by minimizing the total sum of
positional errors between corresponding points among
overlapping point clouds.
The homologous points were individuated outside the landslide
body, thanks to the long range operation characteristic of the
adopted laser instruments (350 m for the Riegl LMS-Z210; 800
m for the Riegl LMS-Z4201).
Together with the obvious advantage to survey large areas, this
fact could be certainly considered in choosing a class of
instruments for monitoring of changes occurring in territory.
An analysis of the map of the residuals coming from the 2001
and 2000 DTMs comparison doesn't show significant large
mass movements, made exception for very restricted arcas
(figure 5-A). The irregular distribution of high residuals may be
ascribed to the effect of artificial mass movements or to the
presence of vegetation not correctly eliminated during filtering
works. The highest topographic changes are detected in the
differential maps (figure 5 B, C) concerning 2004-2000 and
2004-2001 DTM height differences. The negative values are
located in scarp areas and in the middle part of the landslide
body while the most significant uplifts are localized generally
after the degradation areas and not in the expected position at
the foot of the landslide (figure 5 and 6). The analysis of the
mass movements reveals the effects of human activity to
improve the stability of the landslide.
FEY
E
x
: 9 4105
di : @ 035105 a
po ? 05102 M LAM
i ae
ur aet
an PR A
et
Qe T = en
au, " s NS ; a?
un " er
(A)
»i
i
Me
Ÿ
x
xy ^ @® 271-05
Q 51 05 e
& 0510 2 we
Qe
2^
eS
EN
em Em
SP
25 ps ve
Qm
