ced
(2)
d to
e of
this
igh
sly
ced
s of
sur-
nce
tens
ma-
lis-
ran-
e of
ck -
| we
our
ults
per.
Vittorio Casella
2 THETEST AREA AND THE CHECK-AREAS
In the frame of a national research project, chaired by prof. Galetto of the University of Pavia, a test area has been made
in Pavia. First of all, we created a GPS, high-precision network, consisting of twelve vertices; it is connected with the
geodetic Italian network called IGM95. We conducted spirit levelling measurements so that now we know, for each
point of the network, the ellipsoidal height, as well as the orthometric height. Thus we could characterize geoid undula-
tions in a very detailed way. Our test site also has a good mapping support, indeed we have a 1:500 raster map of down-
town, a 1:2000 vector map of all the town territory and a 1:10000 raster map covering all the region. Finally, our town
has several good photogrammetric coverages.
Our test site has been surveyed with laser scanners twice. The first time was in mid November 1999 and we used the
Toposys sensor, installed on a plane of the Italian Company CGR (Compagnia Generale Ripreseaeree), whose head-
quarters are in Parma. It is remarkable that, thanks to the big and powerful plane used, photogrammetric images were
acquired, together with laser data. The Toposys flight had three stages: one with a height over the terrain of 850 metres
and the last-pulse mode; another with the same height and the first pulse mode; a third one, much smaller, with the last
pulse and only 400 metres above the ground. The German company delivered the raw-data, that is sparse points, subdi-
vided for each acquired stripe, and gridded data, with a step of one metre. They all are in the UTM(ED50) format;
heights are above the sea level. In this paper we will only take into account raw-data.
Other laser data were acquired at the beginning of December, 1999, with an Optech 1210 sensor operated by an Italian
Company called Aquater, whose headquarters are in San Lorenzo in Campo, PS. The Optech instrument is highly con-
figurable and we decided to operate with three different settings: there is a part on the town, with an expected homoge-
neous density of two points per square metre; there is another part, on the countryside, with a lower, homogeneous
density; the third dataset is related to the surveying of suspended cables and it is made of sparse, high density, sections.
It is remarkable that the Optech instrument we used is able to acquire the position and intensity of the first pulse, as well
as of the last pulse, at the same time. Aquater delivered only raw-data in the WGS-84 datum, without any altimetric or
geodetic conversion, or interpolation. They put together data coming from different stripes.
2. Characteristics of the check-areas
We have already measured two check-areas: one tennis court and the ramp of a underground parking. About the first,
we have measured with GPS ten points evenly distributed. The convex polygon enclosing them has a surface of 259
square meters. The ten check points have the following main characteristics
m, =100.869 m
3
s, =0.0098 m 5
Please note that we have used ellipsoidal heights, in order to manipulate data as less as possible. The second check-area
is, as already said, a ramp; we have measured nine points on it, and some others nearby. Its main features have been
estimated by a least squares adjustment. The following table summarizes the results (see the next section for a full clari-
fication of the notation)
Geometrical parameters Least squares adjustment results
0.20798
Width 3.96 m p 20.798 %
Length 12.71 m a 0.21434131 rad
Differenccin |. 549 c -1039298.87 m
height
Surface 50.33 m° 8, 0.013 m
Table 1: Main features of the ramp chosen as check-area
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000. 159