(c) 
  
(b) 
  
(d) 
Fig. 5: Defocusing algorithm is applied to aerial interlaced image. (a) Interlaced image, (b) Interlaced image after 
matching, (c) contours of Q over the search area drawn together with the epipolar line, (d) surface produced from the Q 
factor values (inverted for better visualization). As it is expected, lower values of the Q factor appear closer to the 
epipolar line. 
using the cross correlation algorithm (table 2). In the case of 
noisy images the error was almost the same as it was calculated 
on the original images, which means that the algorithm is not 
affected by the presence of gaussian noise. On the other hand, 
the mean error in the noisy images is multiplied by three, when 
the cross correlation technique is used (table 3). 
The proposed algorithm may be used for a variety of 
application. It can be used for close range applications and also 
for aerial cases, during the restitution process and the DTM 
points' collection. 
One disadvantage of the algorithm is, that it delivers "false 
alarms" when the image texture is low, or when multiple 
patterns appear. However, these problems are also typically 
encountered in all the classical approaches. 
7. REFERENCES 
Liu, Y.F. "4 Unified Approach to Image Focus and Defocus 
Analysis," Ph.D. Thesis, Dept. of Electrical Engg., SUNY at 
Stony Brook, 1998 
Sechidis, L,, V. Tsioukas, P. Patias, *Geo-referenced 3D Video 
as visualization and measurement tool for Cultural Heritage", 
"Surveying and Documentation of HISTORIC BUILDINGS, 
MONUMENTS, SITES -Traditional and Modern Methods", 
International Symposium of CIPA, Potsdam Germany, 18-21 
Sep 2001 
Subbarao, M.,T. Wei, "Depth from Defocus and Rapid 
Autofocusing : A practical Approach", Proceedings of the IEEE 
Computer Society Conference on Computer Vision and Pattern 
Recognition, Champaign, Illinois, June 1992,pp. 773-776 
Subbarao, M., T. Choi, A. Nikzad, "Focusing Techniques", 
Journal of Optical Engineering, Vol. 32 No. 11, pp. 2824-2836, 
1993 
Xiong, Y., S. A. Shafer, "Depth from focusing and defocusing", 
Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 
(New York, USA), 1993, pp. 68-73. 
Intel, 
http://www.intel.com/software/products/perflib/ipl/iplperfspec. 
htm 
278 — 
Key 
Syne 
mer 
Abs 
This 
artic 
mea 
non: 
ofa 
the : 
infle 
and 
Or o 
hard 
the 
read 
files 
pur 
Struc 
addi 
rese; 
test 
Surf 
pres 
dynz 
the c 
any 
such 
The 
over 
targe 
mod 
plac 
spec 
requ