3. 3D RECONSTRUCTION 
A photogrammetric system for automated teeth arch 3D model 
generation (Fig. 2.) includes two digital high resolution 
cameras, high resolution structured light projector and PC- 
controlled rotation positioning stage. It allows producing high 
resolution 3D mjdels of teeth arch in automated mode. 
  
Figure 2. Photogrammetric system for teeth arch 3D model 
reconstruction 
3.1 Teeth 3D model generation 
The requirements to 3D reconstruction system were formulated 
according to investigation purposes: 
— High accuracy of 3D reconstruction 
— High resolution of tooth 3D model 
— Short time for producing teeth arch 3D models 
— The possibility of real occlusion registration and 
reconstruction 
— The possibility of occlusion investigation 
To meet these requirements the configuration of 
photogrammetric system including two digital IEEE 1394 
cameras of high resolution (2 Mpixel), structured light projector 
with resolution of 1440x1080 pixels and PC-controlled rotation 
positioning stage is chosen. For automated corresponding 
problem solution coded light is applied providing robust and 
fast scanning. 
The system is calibrated using the original technique for 
estimating camera interior and exterior orientation parameters 
and for determining the parameters of positioning stage rotation 
axis (Knyaz, V., 2005). Special calibration test field with coded 
targets is used for system parameters determination in 
automated mode. The calibration technique provides residuals 
of co-linearity conditions for the reference points after least 
mean square estimation at the level of 0.005 mm. This accuracy 
is quite adequate for the concerning problem. 
For whole teeth arch 3D model generation a set of scans at 
various plaster model positions is acquired. The number of 
scans and their orientation are the subject of user choice. All 
scans are transformed to global coordinate system using the 
results of rotation axis parameters estimation (Fig. 3.). This first 
566 
order scans alignment serves as an initial approximation for 
accurate scan registration by iterative closest point algorithm 
(Besl, 1992). 
  
Figure 3. 3D reconstruction process 
After scan merging a single mesh 3D model is generated using 
interpolating mesh algorithm. 
For acquiring teeth arch 3D model using the described 
automated procedure it is required about 5 minutes. The 
resulting 3D model has accuracy of 0.03 mm and consist of 
about 1 000 000 points. Both upper and lower teeth arch 3D 
models are produced for further occlusion analysis. 
3.2 Teeth occlusion registration 
For occlusion registration special mode is used (Knyaz V., 
2007). In occlusion registration mode both upper and lower 
teeth arch 3D model are placed in working space of scanning 
system and a 3D scan of teeth arches in occlusion is captured. 
Then this 3D scan is used for placing 3D models of upper and 
lower teeth arches in corresponding position. 
  
Figure 4. The scan of the teeth front surface for teeth arch 3D 
models positioning 
The first 
arches 1 
orientatic 
installed 
mould 1 
registerir 
dentist ir 
or right s 
The first 
set of re 
virtual | 
accordin, 
points. 
registrati 
method f 
After the 
models 1 
can perf 
he needs 
occlusioi 
4.1 3E 
For plan 
teeth arc 
3D mod 
  
Figt 
The segt 
using 
segment: 
form. TI 
problem 
wrong-d 
42 Pc 
position 
Segment 
required 
calculati 
of sever: