example of the image of. an edge, the edge is not only shifted
because of image motion but additionally deformed by the
irregular intensity distribution due to the shutter efficiency,
so that the edge becomes steeper in the middle range. Its
lower part drops into an intensity range which lies below
the response threshold of the emulsion. This reduces the
effective amount of blur Ae' compared to the theoretical
amount À th by about 20% in our example. For better under-
standing, Figure 2b shows the temporal intensity behaviour
caused by the shutter as a function of efficiency, with the
magnitude of the total exposure time tyes corresponding to
the theoretical amount of blur Ae, in Fig. 2a because of
the existing relationship between the two parameters.
The size of the image motion amount itself has such an effect
that with increasing image motion the curves presented in Fig.
2a are further flattening, so that their portion lying below
the response threshold of the film increases. Fig. 3 shows
the results of the relevant investigations made on the image
motion simulator (description see below). In the range from
Den = 5 to 130 um the theoretically derived image motion
amounts /Xe' « t are shown as a function of the
th" image
measured blur amounts Ae' in the image being effective at
the test squares and a mean curve (C) derived from this is
recorded.
Here it is noteworthy that unlike Fig. 2a where the relation-
ship between image motion amount Ae', and blur amount
Die) is shown for one edge, the blur amount is further
reduced for an object bounded by two edges, in this case the
test square, since the "ineffective" difference amount De’,
with equal size of the image motion amount, occurs on both
edges.
The seemingly plausible assumption that the reduction is
proportional to the size of the image motion amount, for
example according to the theoretically derived cuve (B), was
not corroborated. In the investigations it was found that up
to a theoretical image motion of about 10 um a blur of the
165
