Hardware manual
Pinpoint Pattern Find™ Impact Reference Guide
Datalogic Automation Inc. 3-84
Pinpoint Pattern Find™
In the Locating Drawer
The Pinpoint Pattern Find tool uses a train region to create a pattern (or model) from part of an image. When
the tool runs, it uses correlation and contour matching routines to find one or more matches to the trained
pattern. It then calculates best-match scores and provides those scores and the x-y coordinates of the found
patterns as one or more origin outputs.
This tool does not automatically find pattern matches in "mirrored" images.
A Setup is available for this tool, which simplifies basic configuration. See “How to use the Pinpoint Pattern
Find tool” on page 3-88.
Input Name What it is
Input Image The image being inspected
Tool Origin The tool’s origin relative to the entire image space
Train Shape List The region of the image used to train the pattern. You can define multiple
shapes and exclude parts of a shape. The train shape must be smaller
than the Find Shape.
Find Shape List The region of the image to search for the trained pattern.You can define
multiple shapes and exclude parts of a shape. The Find shape must be
larger than the Train Shape. If no Find Shape List is defined, the entire
image is searched.
Edge Detection
Sensitivity When
Training
Controls the edge gradient threshold, which indicates sensitivity to noise
in the image, during training. Higher settings detect more weak edges
but cause the tool to run slower.
If the tool fails, the pattern may include edges that are not reliably seen
in all images of the object.
Fixed Edge Thresh-
old When Training
The grey level threshold value to use when Fixed Threshold is selected
for Edge Detection Sensitivity When Training. A larger number rejects
edges that are not well defined.
Percent Of Train
Edge Threshold To
Use When Finding
This setting is used to calculate the gradient threshold for edge finding
when the tool searches for the pattern. To save time, the threshold is cal-
culated once for each image using the Edge Detection Sensitivity When
Training property.
The tool then takes a percentage of that property value and uses that
value for the edge finding threshold when searching.
For example, if this property is 75%, the edges on the pattern edges will
be found if their contrast is 75% of the contrast of the trained pattern.
A smaller number causes more edges to be detected, because the cal-
culated gradient threshold is lower.