| by
ime
sing
| as
ree
can
tor.
by
uld
ing
on
cal
| of
à
set
the
the
Magnetic Disk GOULD IP6432
eu] 512
pti] - x
NEO VAX" LLL I 512
11/750 . X T
4 MB »* |
ped 5v i
fnew IT
Overlay . | + Stereo Y
Magnetic Tape | i
Stereo
Monitor KERN
a
Floating GP |
Mark
Terminal
o
210 X,Y
Fig. 1: Components of the DSS at the Technical University of Berlin
So far considerable hardware limitations have reduced the applicability of the
system. The main purpose of recent studies is to gain experience in handling imagery
of limited size (512x512 pixels), keeping the processing time as short as possible. Up
to now processing of larger amounts of image data, e.g. digitized aerial photo-
graphy, has been possible but not efficient.
The data for the studies at the Technical University of Berlin are acquired by an
A/D conversion of Scanning Electron Microscope (SEM) signals. It must be stressed
that there is no fundamental difference in processing between SEM data and other
types of stereo imagery. Possibly, a slightly changed mathematical model is valid
and SEM object texture and topography are more difficult to handle. The assumption
seems to be justified that experience made by evaluating SEM imagery may be
useful for the processing of other data, as, for instance, aerial photography.
Furthermore, the evaluation of SEM data shows the versatility of the DSS.
TEST OBJECTS
SEM images from surfaces of broken metallic workpieces are used for first practical
tests of the DSS. These images result from a research project which is sponsored by
the German Research Foundation (Deutsche Forschungsgemeinschaft). In the course
of growing demands for three-dimensional (3D) information from electron micro-
graphs it was established in cooperation with the working group 'Metallography and
Metallic Constructions' at the Federal Board of Material Testing (Bundesanstalt fiir
Materialpriifung - BAM) in Berlin.
131
