In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
case ALL-iV 2 and scanner overlap. In the latter case the offset 
of 0,2 mm from the expected zero mean could indicate the scan 
ners calibration error, which is in any case very small. In case 
of drive-line overlap the average of height differences AZ as ex 
pected equals to zero, thus it can be assumed there is no offset 
between drive-lines. In other words there is no systematic error 
in GPS/INS positioning. The standard deviations (std) are equal 
or smaller than 3.5 mm, which denotes the relative precision of 
LMMS laser points. Analyses of the correlation between height 
ALL-N Z Scanner Drive-line 
overlap overlap 
No. of ident. point pairs 
17 754 608 5 473 
tt • i min 
Height 
max 
difference 
AZ [mm] 
-47 -20 -47 
46 36 46 
0.1 0.2 0.0 
3.1 2.5 3.5 
Table 1: Statistics of the height differences of identical points. 
differences of identical points lying on a horizontal surface and 
the geometric attributes, i.e. range and incidence angle, do not 
show a clear trend. 
3.3 Results of DTM interpolation and precision estimation 
Within the test area the terrain laser points, as classified by Geo- 
maat, are used in the following DTM analysis. In Fig. 5 a 3D sur 
face of the interpolated DTM is shown. In this raster image each 
pixel represents an lxl m grid cell and the pixel color shows 
the corresponding grid point height. The grid point elevation 
is changing from -0.19 m at the coastline to up to 22 m in the 
dunes. The white holes in the DTM are results of the shadow- 
effect (white holes in green area) and most probably of the pres 
ence of water-bodies on the beach (white holes in blue area). 
Digital terrain model [m] 
Figure 5: Raster image of interpolated DTM grid points visual 
ized in 3D. 
The height precision ctdtm of the grid points, as computed by 
Eq. 11, varies between 0.0018 and 2.9 m. The average precision 
of grid points crdtm equals to 4.7 mm. For comparison, the 
precision of the observations azi is on average 2.4 cm. 
In Fig. 6 the relation between the height precision of grid points 
ctdtm(y-axis), the number of points n (x-axis) and data quality 
component cr a o (colorbar), is presented. Comparison of the col- 
orbar and the y-axis scale shows, that the size of the grid point 
height precision ctdtm depends mainly on the data quality com 
ponent a a o- Besides, one can observe that, if approximately 50 or 
more points are included in the grid point computation, the stan 
dard deviation of grid point heights ctdtm drops below 1 cm. 
In Fig. 7 the spatial variation of standard deviation ctdtm over 
the test area is shown. Green color shows grid points having a 
DTM height precision vs. no.of points 
No. of points / grid cell 
Figure 6: Correlation between the grid point height precision 
ctdtm and the number n of terrain laser points; color-coded by 
the data quality component a a o- 
Std of grid points - o 0TM 
Figure 7: Grid point height precision ctdtm ■ 
height precision ctdtm smaller than 1 cm. Most of the beach area 
has good DTM quality, which decreases with the distance from 
a trajectory. For example, the precision at the edges of the drive 
line DL11 (most left one) decreases and is at some point worse 
than 2.56 cm, mostly due to the lower point density. The DTM 
quality gets worse also in the dune area, due to the low point den 
sity, low theoretical precision of terrain laser point heights and 
high terrain roughness. 
4 CONCLUSIONS AND FUTURE WORK 
In this article most important results are the empirical laser point 
height precision assessed by the QC of identical points and the 
precision of the grid point heights estimated my a mathematical 
model employing results of MLS adjustment. Both values are 
surprisingly small. 
Firstly, the empirical relative precision of laser points is around 
3 mm. Besides, it was concluded there is almost no bias in the 
StreetMapper system. However, it is recommended to analyze 
bigger number of identical points. Within the QC of identical 
points an attempt was made to show the influence of the scanning 
geometry on the laser point quality. Results show that height dif 
ferences between identical points do not depend neither on the 
range neither on the incidence angle. To verify the influence 
of the scanning geometry on the laser point quality it is recom 
mended to additionally measure control points across the drive- 
line and compare them with the point cloud. 
Secondly, the average precision of grid point height equaled to 
4.7 mm. The computation was performed within the weighted 
MLS adjustment, using the theoretical height precision of terrain 
laser point as weights. It was found that the main influencing fac 
tor on the grid point height precision is the density of terrain laser 
points. That is, because a higher number of observations (i.e. 
terrain laser points) enabled partly elimination of the observation 
noise. The consequence is, that the precision of grid point height 
improved with an increasing number of terrain laser points and