ABB PickMaster Twin - PowerPac Applications Manual page 253

Application manual - PickMaster® Twin - PowerPac
3HAC064218-001 Revision: B
Area Overlap defines how much multiple patterns in the image are allowed
to largely overlap each other. PatMax assumes that these patterns actually
represent the same item in the image. When two patterns overlap by a
percentage greater than the area overlap threshold they are treated as a
single pattern.
Uniform Scale is a threshold that accepts hits that differ in size relative to
the taught vision model. A scale value of 1 indicates that there are no
differences between the found item and the taught vision model. A value <1
indicates a smaller model.
Score Using Clutter defines a measure of the extent to which the found item
contains features that are not present in the trained vision model. By default
the PatMax analysis ignores clutter when scoring which means that the
patterns receive the same score regardless of the presence of extra features.
If this checkbox is selected, clutter is included in the calculation of the score.
If the application is an alignment application in which the background does
not change, Score Using Clutter should be selected.
Limit Search Region limits the search area for the PatMax analysis. Only
objects within this area will be found. A smaller search area will decrease
the search time.
Note
When combining Fine/Coarse Granularity and Uniform Scale a slight
difference in the score can appear between design time and Runtime.
Therefore, the model should be tested in Runtime to verify that items are
identified as expected.
7 In the Post search filters part, define the score values for each pattern in
the search image.
Fit Error Limit is a measure of the variance between the shape of the trained
pattern and the shape of the pattern found in the search image. If the found
pattern in the search image is a perfect fit for the trained pattern, the fit error
is 0.
Coverage Limit is a measure of the extent to which all parts of the trained
pattern are also present in the search image. If the entire trained pattern is
also present in the search image, the coverage score is 1. Lower coverage
scores indicate that less of the pattern is present. This parameter can be
used to detect missing features.
Clutter limit is a measure of the extent to which the found pattern contains
features that are not present in the trained pattern. A clutter of 0 indicates
that the found pattern contains no extra features. A clutter score of 1 indicates
that for every feature in the trained pattern there is an additional extra feature
in the found pattern. The clutter can exceed 1.0.
Inspection Levels - Inspection I, this inspection is also called Inspection I
in PickMaster PowerPac. With this function it is possible to classify the found
models into two categories. A model can either be classified as accepted or
rejected. An accepted model has better search results than the rejected
© Copyright 2021 ABB. All rights reserved.
4 Working with PickMaster PowerPac
4.3.9.2 Configuring a geometric model with PatMax
Continued
Continues on next page
253