ibul 2004
spect that
ysis and
ces. This
in a fifth
hnners can
M.
ance and
‘mine the
he image
ince and
nates for
'€ can be
larkers is
h on the
Lingua
neters of
tem.
first step
onto the
ach point
E T).
cloud is
je radial
ay
—
jection
(4)
ze centre
"nts.
matrices
Illy much
For this
ie. digital
lled: the
els. (this
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004
quantity depends on the density of the points acquired with
the laser scanner and on the image resolution in terms of dpi).
In order to fill the distance matrix, it is necessary to integrate
the missing values with an interpolation process. The
"average weighed method" has been used. The four nearest
pixels, of which the value of distance is known, are
considered. The interpolation of the distance values is carried
out using the formula:
3 = d, 6, d, * d, 6, +d, à,
nL)
d +d, +d; +d, (3)
Where i j= the indexes of the current pixel
Ó,...47 the distances between the centre of the
perspective and each of the 4 object points
d, 47 the distances, on the image, between
the i,j pixel and the pixels used for the
interpolation.
The same process is used for the interpolation of the
reflectivity matrix A.
1.5 Validation of the procedure
A test area was used to calibrate and validate the proposed
method. It has been built in the photogrammetry laboratory at
the Politecnico di Torino. A total of 40 high reflectivity
markers, whose positions were accurately determined using
classical geodetic techniques, were realised in the laboratory.
À laser scanning was carried out, using a RIEGL LMS-Z420
laser scanner, whose accuracy is +5 mm for the 3
coordinates. Simultaneously a series of digital images were
acquired, with a Nikon D1X mounted with a 28 mm lens.
After having carried out the survey, the orientation and
calibration parameters of a digital image were calculated and
a Solid Image of the same image was created.
The image size is 1312 pixels in height and 2000 pixels in
width, the resolution is 2100 dpi, corresponding to a pixel
size of 12,1 uum.
The obtained results are:
Orientation parameters of the digital image
da = -0.071 m w= -83.505 gon
Yo = -0.038 m @= -298.742 gon
du = 0253 : m k= 114.379 gon
. Calibration parameters (see (3) and (4))
&o = -0.160 Mm
No = 0.011 Mm
c= 28.49 Mm
k, = 170376 E-06
il
=
3,98296 E-09
=. 27/153 E-12
?
Fist analysis was performed to investigate the accuracy of the
solid image.
All the possible distances between the markers placed on the
object were calculated and compared with those obtained
from the Solid Image.
The following results were obtained:
Mean and maximum differences
Minimum value -8 mm
Maximum value 10 mm
Mean value 12 mm
2. SOFTWARE FOR THE CREATION AND MANAGEMENT OF A
SOLID IMAGE
2.1 The LSR 2004 software
À specific software package (LSR 2004) has been developed
by the authors in Visual Basic 6.0 to create and manage the
Solid Image and some of its first applications, .
LSR 2004 (Laser Scanner Registration 2004) is a software
made up for terrestrial laser scanner data processing and the
creation and management of the Solid Image.
LSR 2004 allows:
* dense point clouds and high resolution digital images to
be imported and exported;
e point clouds to be filtered to remove outliers and gross
errors;
e the position of the high reflectivity points (markers)
placed on the acquired object to be determined
automatically;
e adjacent scans to be recorded and geo-referenced
automatically using reference markers;
* triangulation to be carried out in the case of different
scans;
e images to be calibrated;
e the Solid Image and its applications (3D colour model
and artificial stereoscopic pair) to be created and
managed;
* direct measurements to be carried out on the Solid
Image.
LSR 2004 allows a Solid Image to be created through:
* an automatic calibration procedure of the digital image
e projection of the point cloud and
» interpolation of the D and A matrices.
It allows the operator to visualise the RGB image or each of
the two D or A matrices, codified in colour scales.
iei Eran
die DEC (CE
& Torch odie sr
siFe laesi [eruit SLSHX à 5 de aitu
RSEN
Figure 4 The LSR 2004 software — Solid Image of the
Tetrarchi statue in Venice
The resulting Solid Image is a high-resolution digital image
integrated with the 3D information of the photographed
object. The spatial position of the image points in terms of
X.Y,Z coordinates can be seen in real time. This is obtained
by simply moving the PC cursor onto the Solid Image: the
