2004 
  
with 
hese 
mote 
‘ibed 
wing 
ation 
1 the 
) the 
grid 
pose 
grey 
used 
iade. 
es of 
for 
sing 
tion 
n of 
CESS 
ages 
onal 
) are 
arate 
>d to 
sin 
rical 
not 
n in 
ages 
S-of 
1 for 
ager 
ires 
s for 
For 
It of 
3D. 
>s to 
ging 
and 
s fit 
pace 
line 
The 
er of 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B6. Istanbul 2004 
  
3.2 Creation of vector model 
The vector model was created by processing of scanned 
photographs, which were taken by photogrammetric cameras. 
For these purposes were used UMK 10/1318 of Zeiss-Jena and 
stereometric camera SMK 5,5/0808 with 40 cm baselength, 
which is Zeiss-Jena production. The usage of colour pictures 
was not very convenient because of small distances for taking 
the photograps that did not allow capturing whole wall in one 
photo. By these reason the traditional photogrammetric cameras 
were used. For mapping of photos was used software product 
DVP (Digital Video Plotter) of DVP Geomatics Inc. (Leica, 
1995). The used version of this product is (working in DOS 
mode) is developed for processing of separate stereo images. It 
is convenient for mapping of separate facades of the object. The 
vector models of facades are produced. The control points, 
measured with surveying methods, were used to connect models 
of separate facades. All models of facades were transformed to 
local coordinate system of grid model. 
3.3 Binding Photo-textures with Grid model 
Binding of photo-textures with vector model requires assigning 
of textures to model surfaces and geometrical connection 
between the photo-textures and corresponding surfaces. 
Different software products could be used for these purposes. 
The requirements that these products must satisfy include are 
possibilities for such binding, easy creation of final model and 
possibilities for export in format compatible Internet browsers. 
Possible candidates are AutoCAD Land Development with 
AutoCAD Overlay, 3D Studio MAX and Micro Station. 
Highest flexibility shows 3D Studio MAX but this product 
requires very long period of training and is not very convenient 
for education in Geodesy. By these reasons in educational 
process are included initially Micro Station of Bentley and 
AutoCAD Overlay. The choice of the product depends on the 
environment in which the vector model is produced and 
processed. DVP product does not export directly *.dgn format 
but creates *.dxf file. Micro Station allows importing of *.dxf 
format and all following processing is made in the environment 
of Micro Station. 
To create adequate model it is necessary to crop the photo- 
texture by the boundaries of surface. For educational purposes 
is used Photoshop. The boundaries are overlapped over the 
raster layer by converting the vector model to raster layer. The 
results for the west wall of the church “The Assumption of the 
Virgin Mary" are shown on the Figure |. 
  
  
  
  
  
Figure 1. Transformed image of west wall 
4. TRAINING FOR DIGITAL ORTHOPHOTO 
PRODUCTION 
Digital orthorectification produces the most precise results for 
orthophoto maps. The main importance for the accuracy of this 
process is the quality of digital terrain model. The Ortho Engine 
of PCI Geomatics was used for education purposes in 
preparation of orthophoto for cadastral purposes 
(Marinov, B.D., Draganova, N.P., 2003). 
49 
Cut image of west wall 
4.1 Processing Steps for Aerial Triangulation 
The inner orientation with Ortho Engine allows examination the 
influence of different number of points over the residuals (PCI 
— Geomatics, 2001). The accurate measurements of fiducial 
marks decrease the errors of transformation. Disadvantage of 
inner orientation stage is that it is not possible to select type of 
transformation and operator could not estimate the difference in 
errors for different transformations.