fast_match_mg — Searching all good gray value matches in a pyramid.
fast_match_mg is obsolete and is only provided for
reasons of backward compatibility. New applications should use the
shape-based or NCC-based operators instead.
fast_match_mg like the operator
fast_match performs a matching of the template of
In contrast to
fast_match, however, the search for
good matches starts in scaled down images (pyramid).
The number of levels of the pyramid will be determined by
Hereby the value 1 indicates that no pyramid will be used.
In this case the operator
equivalent to the operator
fast_match. The value
2 triggers the search in the image with half the framesize.
The search for all those points showing an error small enough
in the scaled down image (error smaller than
will be refined at the corresponding positions
in the original image (
The runtime of matching dependends on the parameter
the larger the value the longer is the processing time,
because more points of the pattern have to be tested.
MaxError is to low the pattern will not be found.
The value has therefore to be optimized for every
NumLevel indicates the number of levels of the pyramid,
including the original image.
Optionally a second value can be given. This value specifies
the number (0..n) of the lowest level which is used the matching.
The region found up to this level will then be zoomed to the size
of the original level.
This can used to increase the runtime in the case that the accuracy
does not have to be so high.
Input image inside of which the pattern has to be found.
All points which have an error below a certain threshold.
Maximal average difference of the grayvalues.
Default value: 30.0
Suggested values: 0.0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 9.0, 11.0, 15.0, 17.0, 20.0, 30.0, 40.0, 50.0, 60.0, 70.0
Typical range of values:
Minimum increment: 1
Recommended increment: 3
Number of levels in the pyramid.
Default value: 3
Suggested values: 1, 2, 3, 4, 5, 6, 7, 8
If the parameter values are correct, the operator
returns the value 2 (H_MSG_TRUE).
If the input is empty (no input images are available) the behaviour can
be set via
If necessary, an exception is raised.