__header__
The correlation tool will
allow the user to train Automation Manager
to find a specific piece of an image by comparing it to a predefined
image (what we will call the model).
To begin drag this tool onto a program and select
it. When the
tool is selected in the program two ROI
boxes
will display in the viewer
window, one green and one blue. The blue box is used to train
the
tool.
Position
the blue ROI over the part of the image you
would like the
correlation tool to locate.
Right-click the Correlation tool in the progam and
choose
train. When the tool is trained its icon will reflect the
trained
area, that is your model.
Now that the tool is trained, whenever the
correlation too is run in
that program it will look for what it has been trained to find and
place a crosshair in the center of that area. By default the
green box is the ROI that scans images to find what the tool has been
trained to locate.
The correlation tool works by doing a low resolution
pass search for
the model (i.e. what you have trained). After this low pass
search is done over the entire image there may be several canidates
that could possibly be the right match. Next, a high
resolution
scan is done on each of the possible canidates. The
caniditate
with the highest percentage of similarity will be selected from the
image.
The Minimum Accpet Score is a percentage based on similarity (of
shape, not brightness or contrast) to the trained model. If
the
score is below the Minimum Accept Score,
nothing will be
selected from that image and it is deemed to
have failed.
The correlation does a low resolution pass search of the entire image
for the model. This
low pass search is
done to select possible regions that may correlate to the model you
have trained. Based on the possible regions found in the low
resolution pass a high resolution pass is done to select the sector
that is most similar to the trained model. The low pass
search
speeds up
the execution time
to find the model.
The subsample factors define how to render the
image for the low
pass search. For example, if the subsample is 3 then only every thrid
pixel
is used in the
low resolution pass. Note: the higher these factors the faster the
search, but if the sub sample factors are too high, then the model may
not
be found. To check
that the factors are not too high, view the "Model-Low" resolution
model and hit "Re-Train". This
low resolution model should be recognizable.
The "Re-Train" button will retrain the Low Pass model only. It will resample the model with a lower resollution for the first pass (see above - Sub Sample). When the "Sub Sample" X and/or Y have been adjusted the "Re-Train" button must be hit in order for the change to be made. You can see how the change has affected the model be selecting "Model-Low Resolution" from the dropdown menu. "Re-Train" will only resample the model that has already been trained, it will not genetrate a completely new model.
Displays the trained model.
- The Model
is the actual
image of the model you have trained.
- The Model-Low
Resolution
shows how the model looks in low resolution
mode.
- Result Image
shows an image
result of the correlation. Points of
correlation are shown as
light spots.
Defines the thoroughness of the high resolution pass. Low
Accuracy will be the fastest but least accurate, Sub Pixel will be the
slowest but most accurate.
Low Accuracy
stops at the low
resolution pass.
Pixel
accuracy stops at the
pixel resolution pass.
Sub pixel
accuracy looks at the
surrounding locations of the found
model to determine more precicely the correct model.
When the correlation too is run it will place a crosshair on the part
of the image that fits the model. The crosshair is placed on
the
center of that image. Using the Center Offsets you can adjust
the
tool so it will place the crosshair either above or below the center
(using the Offset Y) or to the right or left of the image (using the
Offset X).
Train:
This button is equavalent to right-clicking the Correlation tool in the
program and choosing "Train" This button will train as the
new
model whatever is currently inside the blue ROI.
Run:
The model is searched for. Its position, and score
are calculated.
Cycle /Time:
Reports the execution time in msec of the trained model.
Reset
Resets the positions of the ROI.
Number
Points
The number of points/features found. Currently the
correlation
tool will only find one point everytime it is run.
Score
The correlation score. !00% is perfect match.
Contrast Score
The measure of contrast. !00% contrast is the same as the
trained image. The Correlation tool does not take into
consideration contrast, although it does measure it.
X,Y
The found location of the model.