All
ate
15
ng
ire
Cóltekin, Arzu
To continue the process the images are read from files into WorldToolkit, then a simple box is created; the size is
visually adjusted to start with. Navigation of the wire frame model on 2 dimensional background to supply the visual
matching is the main process to create the visual model. Then every object is symbolized with box-shaped wire frames.
Most of the cases one object would have more than one box to be completed.
In this environment, rather hard to navigate, but concerning the time usage and complexity still more reasonable then
measuring all the details, the images are processed to get the functional model. Note that the viewpoint of the model is
moved, not the model itself. The idea in simplest words was to find the same viewpoint as the camera was standing. The
complete model can be seen below:
Figure 2. The completed model: 3" Floor of the institute, notice the level of detail.
After completing the functionally correct model, control points on the image and corresponding points on the model are
measured. These control points are used to “rectify” the model onto the image (photogrammetric bundle block
adjustment). The process is repeated several times from several different viewpoints iteratively to get the best possible
correction. This is needed because of using single images. Accuracy values for X,Y,Z coordinates ranged a maximum
of s, 20.62 cm and a minimum of s, — 0.25 cm, please note the proportional accuracy.[Haggrén, Mattila]:
N N
POINT 2 X ooi y... m s OY ed. A ois NR
E iss 1. 83142. UU ANE 6.000. 1 0272205 1 0.272366 : 0300964 :
s à: MESSE = AIS | 201062 0277134 : 0.257210 : 0.577030 :
4... 2105 ; … 2494 321 470 : | 0.312741 : 0.280385 © | 0.559316 :
4 5 2 55 AR T 345.401 : | 0310138 = 0.265001 : 0.572219
eae 22.343: - 26.322 : . 3431 : 0341388 : 0.296909 : 0.612104
La Baad acis i 228658 : . AUS 0.348306 © 0.273489 : 0.623333 :
dl i 88505 io $5001 369 311 : 0.238465 : 0.310697 : ! 0.600461 :
M UE T8345 Sao : — AN DS 0 0.309054 : 0.309336 : 0.603973
EH -60.187 „51.852 306.018 : 0.205445 : 0.325232 0.603706 :
ese ere aspi waa REN NE Ne TE EEE ERA ER RU TA NES ST UC TA TAN RE EUR AUTRE NTM SACS EEE CAM S
Table 1. Model Coordinates (X,Y,Z) and their accuracy
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000. 145