ate system 
> X-axis is 
he normal 
(NOWN CO- 
riginal and 
lled basic 
en the two 
s makes it 
olved with 
he camera 
ose of the 
ocess. The 
rtophotos, 
] picture is 
vels in the 
1al picture 
ures 2 and 
ir original 
ets. 
  
important 
| we made 
pair photo 
s of known 
'eometrical 
parallax of 
le is Zero. 
conditions 
iple. 
  
  
a) original right photograph 
  
b) normalized right photograph 
Figure 3. Right test photograph 
  
Figure 4. Optimized and edited photograph 
Another possible solution for representing the point mass 
resulting from 3-D object reconstruction is 3-D surface 
modelling, which was solved by implementing animation 
modelling. But from medical practise point of view, the split 
monitor solution remains more useful. In the case of one picture 
we defined the coordinates for the DLT parameters with the 
help of usual image processing software that was developed and 
tested by our team during previous studies. Figure 4, shows 
optimized and edited photograph of the test area. Figure 5, 
shows an X-ray photographs of the human heart, and figure 6 
shows the measured and plotted diagram of the heart's vessels. 
  
Figure 5. X-ray photographs of the human 
hearth 
  
Figure 6. Plotted diagram of the heart's vessels 
4. CONCLUSIONS 
In conclusion, this paper presents partial results of an ongoing 
research work. The photos used were X-ray photographs taken 
about dead human organs. An adequate technique must be 
worked out for usage in living bodies, where special attention is 
to be paid to accuracy and reliability. However, in some 
medical interventions, where the relative positioning of organs 
is more important than that of the absolute accurate location of 
one specific organ, this technique has the potential of being of 
significant benefit and usefulness. In its present status, the 
technique is still useful in assisting the operating physician in 
knowing the geometry of the vascular system, especially in 
certain cases where it assures higher safety when the physician 
can examine through the medical instrument both the patient 
state and the picture of the vascular system at same time. 
REFERENCES 
Alhusain, O. 2000. Correction of non-linear distortion in digital 
images using a proposed normalization algorithm. In: The 
International Archives of Photogrammetry and Remote Sensing, 
Amsterdam, The Netherlands, Vol. XXXIII, Part B1, pp. 12-18 
Gonzales, R.C. and Woods, R.E. 1993. Digital Image 
Processing. Addison-Wesley, Reading, Massachusetts, etc. pp. 
161-201. 
Kenneth, R. K. 1996. Digital Image Processing. Prentice-Hall, 
New Jersey, pp. 329-336. 
Wong, K. W. 1975. Mathematical formulation and digital 
analysis in close-range Photogrammetry. Photogrammetric 
Engineering and Remote Sensing. 41(1 1), pp. 1355-1373. 
—249— 
M mI OE Un mmm