DMOOMu 
phs. 
tion 
of 
aphs 
ent, 
at 
ure. 
in a 
in 
lass 
the 
the 
the 
phs, 
the 
ixed 
onal 
be 
may 
at 
not 
the 
to 
for 
thod 
hich 
the 
ons. 
sted 
can 
nent 
ater 
ject 
able 
used 
ents 
s or 
very 
3-RECORDING THE DISPLACEMENT FIELD 
An accurate and complete photographic 
record of the displacement field should 
fulfill the following conditions: 
-A set of static reference marks should be 
recorded in the photographs. This is 
necessary to determine the datum level for 
height measurement in the stereo image. In 
the present investigation an orthogonal 
grid was scored for this purpose on the 
inside face of the glass side of the box. 
This grid was formed of equal squares 25mm 
X 25 mm. 
-The camera used should have a good 
quality lens capable of producing sharp 
negatives so that the objects in the pair 
of photos are easily fused when viewed 
stereoscopically. In displacement field 
work the overlap is 100% and the movement 
of particles between photos will be 
recorded through the same part of the 
lens. Any distortion by the lens will 
produce similar errors in succesive photos 
and the measurement of the parallaxes 
between points will not be significantly 
in error. For a high degree of accuracy a 
plate camera capable of taking glass 
plates or large size negatives is 
preferable. In the present study a 
Hasselbad 500 LM camera fitted with an 
80mm lens, capable of producing a SS5mm by 
55mm negative, was used. A clear relief 
image is enhanced if the photos contain 
contrasts between adjacent objects. 
Careful illumination of the model can 
improve the texture of the surface and the 
quality of the photos. Two floodlights 
were positioned at an angle of 45° with 
respect to the camera sand bed axis. The 
camera had a black screen (with only the 
lens protruding) to minimize unwanted 
reflections in the glass window. The 
camera position was approximately 0.5m 
from the edge of the model. 
-The camera should be fixed in position 
relative to the stationary parts of the 
model so that the optical axis is 
perpendicular to the plane of 
displacements. If the camera moves 
between photos this will become apparent, 
when a pair of photos is viewed 
stereoscopically, by differences in 
parallax between points in the model which 
are known not to have moved. Andrawes 
(1976) stated that the error introduced by 
a tilt of the camera of 2? from the 
perpendicular will introduce an average 
error of about t1% in the value of the 
contours. In this study the camera has 
been mounted on a rigid pedestal and the 
pictures were taken with the help of a 
remote control device so that the camera 
was not touched during the test. 
4-MEASUREMENT OF DISPLACEMENT 
The major movements in the anchor model 
occur parallel to the vertical axis and 
therefore the negatives were set up in the 
instrument so that the vertical axis shown 
in the negatives was parallel to the x 
axix of the instrument. This dramatized 
the apparent relief when the pair was 
viewed stereoscopically. In the negatives 
    
the grid appears as a distinct white 
colour "datum" surface from where the 
measuring mark can be driven towards the 
elevated sand particles. The measurements 
were made point by point along steep 
slopes, peaks, depression and breaks in 
the relief. If the whole displacement 
field is to be reproduced there must be an 
interpolation between the measured points. 
It was found that measurements at 500 
points gave good reproduction of the 
images viewed. 
S-COMPUTER PROCESSING OF DATA 
The author has developed a computer 
program which processed all the data given 
by the stereocomparator and plotted all 
the required diagrams using the  UNIRAS 
package. The measurements made in the 
stecometer are from points which are 
randomly spaced throughout the field. 
These measurements are used by the program 
to calculate the values of the vertical 
and horizontal displacements which occur 
in the field at the position of the nodes 
of a regular grid which covers the whole 
field. The fields of the vertical and 
horizontal displacement components are 
then plotted. The displacement component 
values are combined to give the resultant 
displacements at the nodes of the grid and 
they are also used to derive the strains 
throughout the field. 
  
    
    
    
Input Data from 
Stecometer 
Pre-process Data 
Transtorm Coordinates 
Calculate Vertical 
& Horizontal Displacement 
| 
Calculate Displacement 
Resultant 
& Strain Values 
  
  
  
  
  
  
  
  
Plotting of Contours 
of Vertical and Horizontal 
Displacement 
Plotting of Displacement 
Field 
Plotting of Contours of 
Maximum Shear Strains 
  
  
  
  
Fig. 1 Flow Chart