International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanhul 2004 
4. CONCLUSION 
Pre-processing of the aerial image is longer by only 4-5 minutes 
for each aerial image scanned with pixel 21 pum (the time give 
concerns a processing of the image on a computer Pentium IV 
1.5 GHz). Taking into consideration potential of present 
software's (a possibility of execution of required processes in 
batch processing), we can assume that there is the possibility of 
transformation of about 300 aerial images on one computer PIV 
1.5 GHz within one day. 
Using the method of enhancement the original image with 
filtration in object space adapting Prewitt's mask firstly we 
must execute filtration, and then to make association of two 
images by addition them to each other. Such an operation will KEY 
take about 10 minutes for one image, so one may process about 
140 aerial images within one day. ABST 
The nt 
Using filtration in frequency space we must make referer 
transformation to Fourier space, then make there filtration, and which 
then to perform retransformation and to add to original image. DEM. 
The whole procedure takes about 20 minutes for the same 
working parameters, which were presented above. One may 
enhance with this method about only 70 aerial images within produc 
one day. made : 
of exis 
one. T 
In ord 
REFERENCES 
Experimental Test on Digital Aerial Triangulation. OEEPE i 
Official Publication No. 31, March, 1996. Digital 
bottlen 
Kaczynski R., Ziobro J., 1998. Digital Aerial Triangulation for acrial t 
DMT and Orthophoto Generation. International Archives of orthop 
Photogrammetry and Remote Sensing, Vol.32, part 4, Stuttgart, the pr 
pp. 281-283. becom 
model: 
Kasser M. Egels Y., 2002. Digital photogrammetry.. London, stress 
New York, pp. 124-135. On the 
point « 
attracti 
Manua 
Digital 
are stil 
Autom 
proved 
need fc 
valuabl 
automa 
a day, 
probler 
created 
probler 
must cl 
A new 
Labora; 
Univer: 
tuning 
checkir 
became 
most re 
number 
impract 
A simp 
statistic 
measure 
rejected 
* Corr 
cA 
uA 
2