155
Smart Camera / Vision-Sensor / 1D-/2D-Code-Scanner / OCR Reader
Characteristic
Optimization In the case of especially large models, it may be advisable to select
the number of model points by setting the optimization parameter to
a value other than “-”. In the case of smaller models, reducing the
number of points does not result in any acceleration.
Auto The number of points is reduced automati-
cally by the algorithm.
- No optimization is conducted. All object
points are saved.
Min Point Reduction There are three different levels for reducing
the number of points of a taught-in model.
Reducing the number of points can be very
helpful for large objects.
Med Point Reduction
Max Point Reduction
Regeneration If this parameter is selected, a new model is
generated each time an image is recorded.
It must be noted that regeneration in the
case of large rotation or scaling values in-
creases memory occupation. Regeneration
also takes a great deal of time.
No Regeneration Regeneration of models is deactivated.
Metric The metrics setting specifies the conditions under which the sample
will still be recognized within the image.
Polarity – Active The object in the image must demonstrate
the same contrast characteristics as the
model. For example, if the model is a bright
object against a dark background, the ob-
ject is only detected within the image if it’s
brighter than the background.
Global Polarity –
Ignore
The model is also detected when the con-
trast characteristics are exactly the oppo-
site of those of the taught-in object.
Local Polarity –
Ignore
If this value is selected, contrast polarity
may only change amongst various parts of
the model, but the polarity of model points
within the same part of the model may not
change The term “Local Polarity – Ignore”
must be correctly understood. It means that
changes in polarity between neighboring
parts of the model don’t influence the score
and are thus ignored.
To Next Page
Loading...