Table of Contents
Advertisement
(7) The expansion/contraction method of eliminating noise in binary images
When converting an image to binary values, it is often the case that dots of noise will show up in the
converted image. This noise may be eliminated during pre-processing by using the smoothing
function. However, it can also be dealt with through the expansion/contraction noise elimination
method.
-1 Expansion
A single white dot, or small groups of white dots, in the middle of a black background can be
eliminated from the image.
2 Contraction
A single black dot, or small groups of black dots, in the middle of a white background can be
eliminated from the image.
-1 Expansion → contraction
When isolated dots of noise are eliminated by expansion, the areas enlarged by the expansion
are returned to their original size by contracting them again.
2 Contraction → expansion
When isolated white noise is eliminated by contraction, the areas shrunk by contraction are
returned to their original size by expanding them again.
The IV-S30 has both "contraction → expansion" and "expansion → contraction" functions for
eliminating binary noise.
- Number of times of expansion and number of times of contraction can be set independenly.
When the number of contraction cycles is set to 0, and the number of expansion cycles is one or
greater, then only the expansion function will be used.
On the other hand, if the number of expansion cycles is set to 0 and the number of contraction
cycles is one or greater, then only the contraction function will be used.
Ë Ë Ë Ë Ë Measurement programs which are affected by these settings
Area measurement by binary conversion, object counting by binary conversion, object identification
by binary conversion
Setting the Conditions for Each Object Type
3-11
3
Advertisement
Table of Contents
