International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004 
  
  
  
  
  
  
  
  
  
  
  
  
  
Camera 1 valuc o 
Focal Length (mm) 16.006119 0.0093 
Principal Point x 0.001229 0.0046 
Principal Point y 0.000508 0.0016 
Al -0.469369D-03 0.1807D-03 
A2 0.324284D-05 0.2350D-04 
Scale -0.559509D-04 0.5976D-04 
Shear -0.118859D-03 0.5949D-04 
  
Table 2: Results of camera calibration for camera 1 
3. RESULTS 
As a result we get the difference of coordinate's value between 
the beginning and the end of period. Fig 4, 5 illustrate the sur- 
faces of the specimens for either 0 hour of 24 hour measure- 
ments. It is clearly seen from the surfaces that there is a signifi- 
cant settlement on concrete specimen. Colorized scales are also 
given in Fig 4 and Fig 5. The results show unsatisfactory ran- 
dom behaviors in x and y directions. A downward movement 
was determined for the z- coordinates. The maximum value of 
z-coordinate movement is 0.739 mm while the minimum value 
is 0.16 mm and the average is 0.42 mm. 
0 
By 
| 
| 
| 
e 
= 
  
Figure 4: The surface of concrete specimen (Just after casting) 
Settlement of the concrete in early ages (before the setting) is 
an expected phenomenon. Volume of the hydration products is 
lower than the total volume of the reactive materials before hy- 
dration. This phenomenon, which is called chemical shrinkage, 
causes volume reduction on concrete. This volume reduction is 
occurred as settlement in fresh concrete (Aitcin, 1998). Air 
voids could also cause the settlement by leaving from the paste 
in this state. 
1a 
  
- i | 
SE | 
i i 
sS Ten m 
-., pi 
ev Deum m = 
Figure 5: The surface of concrete specimen (1 day later) 
4. CONCLUSIONS 
This paper presented a stereo vision system designed for activi- 
ties of inspection. The system was divided into three stages: 
stereo image acquisition, preprocessing, and reconstruction. 
The precision of this system could be improved by a higher 
camera resolution. The precision and the reliability may be im- 
proved by adding more cameras to the system. A surveying of 
other objects is possible. The size of the object plays no role, 
but must be considered regarding the resolution and the accu- 
racy however. Satisfactory settlement values were obtained 
from this study. Effect of the thickness of the specimens on the 
z-coordinate variation could be studied in further studics. 
ACKNOWLEDGEMENT 
I would like to thank Bahadir Ergün who designed this system. 
REFERENCES 
Aitcin, P.C., 1998, High Performance Concrete E&FN Spon 
Press, London. 
Maas, H.G., Kersten, T., 1994 Digital Close Range Photo- 
grammetry for dimensional checking and control in a shipyard. 
‘ideometrics 111,. 
Neville, A.M., 2000, Properties of Concrete Pearson Ed. Asia 
Press, England. 
Remondio, F., 2002 3-D reconstruction of articulated objects 
from uncalibrated images. SPIE Electronic Imaging , Proc. Of 
SPIE 4661, San Josc, USA. 
Suthau, T., Hemmleb, M., Zuran, D. and P.G. Jost-Brinkmann, 
2000. Photogrammetric measurement of linear objects with 
CCD cameras - super-elastic wires in orthodontics as an exam- 
ple, /APRS, Vol. XXXIII, Amsterdam. 
Suthau, T., More, J., Widemann, A. and Franzen, J., 2001 
Close range photogrammetry with ced cameras and matching 
methods -applied to the fracture surface of an iron bolt. Fourth 
  
   
Intern 
  
Intern 
Days, 
White 
deflec 
gramn 
Applic