each set which
"he homographic
the double set of
re then projected
parameters. For
nearest point of
described in the
ITAL IMAGE
E
been performed
f the statue of
[useum — Turin,
ition parameters
image has been
a provided by a
the automatic
d on the statue
1e laser internal
[m]
-0,771
-0,832
-0,094
0,819
1,057
-0,682
-0,665
-0,440
-0,074
0,113
0,239
0,412
0,553
th the camera
cquisition; the
assumed as X
p and k angles
d by means of
ind y are listed
the reflecting
bed before, the
centres of the
| the fiducial
founded using
Target x
# [mm] [mm]
20 -7,271 -16,983
21 3,298 -18,512
22 -5,853 -8,145
24 2,714 15,970
25 0,619 20,658
60 -7,208 -14,748
61 4,176 -14,409
62 4,678 -7,838
63 -0,606 0,468
65 0,584 3,577
66 -4,736 5,823
67 0,440 9.212
68 4,474 10,880
Table 11
Target € n
# [mm] [mm]
20 -8,862 -19,854
21 3,198 -21,823
22 -6,476 -9,024
24 4,037 17,469
25 1,924 22,248
60 -8,724 -17,253
61 4,214 -17,050
62 5,165 -8,881
63 -0,282 0,617
65 1.522 4,578
66 -4,093 7,056
67 1,444 10,471
68 5,856 12.329
Table 12
The last step of the software was that of looking for the
homologous of the obtained reflecting targets. Following the
described procedure, points 24, 60, 61 and 66 were selected by
the software for the estimation of the homographic
transformation between the two planes (one of the virtual image
[xy]of the scanner and the other of the digital image [£n]).
Using these parameters, all the obtained reflecting targets were
transferred into the [$n] reference system .
Table 13 lists the “point set 1" coordinates of table 12
trasformed in [En] reference system and table 14 shows the
resulting distances between the transformed points and the
image points, wich were chosen by the software as homologous.
The average distance was of 0.330 mm; considering that the
average scale of the image was of 1:35, the homographic
transformation placed “point set 1" with an approximation of
about 1 cm on the object.
Finally, table 15 compare the results of the orientation of the
image using the calibration certificate parameters and manual
collimation of the targets (row “Photogr.”) with those obtained
using the data coming from the described procedure (row
“Laser calibr.”).
Target E
# [mm] [mm]
20 -9.050 -20.198
21 2.990 -22.511
22 -6.525 -8.940
24 4.037 17.469
25 2.202 21.995
60 -8.724 -17.253
61 4.214 -17.050
62 5.075 -8.730
63 -0.185 1.121
65 1.256 4.578
66 -4.093 7.056
67 1.425 10.617
68 5.597 12.333
Table 13
Target AE An Distance
# mm mm mm
20 -0.188 -0.344 0.392
21 -0.207 -0.688 0.718
22 -0.049 0.083 0.097
24
25 0.278 -0.253 0.376
60
61
62 -0.090 0.151
63 0.097 0.504
65 -0.266 0.000
66
67 -0.018 0.146
68 -0.259 0.004
Table 14
Xc Yc Zc © © K
[m] [m] [m] [gon] [gon] [gon]
Photogr. |20,564]100,206/21,424! 0,2546. 0,3820 0,9549
Laser calibr.[20,562/100,200{21,430| 0.2689| 0.4012] 1,0342
Table 15
The differences are negligible in all cases, than the calibration
of a digital image using laser scanner data and the proposed
automatic procedures can be considered as a new opportunity
for the laser scanner integration with photogrammetry.
6. CONCLUSIONS
The implemented software allows the automatic orientation of a
digital image using only the information which can be extracted
from the DTM produced by a laser scanner device.
The described procedure makes it possible to orientate the entire
—325—