sfn_correlators¶
This module defines the spatial correlator used by SurfaceFeatureNavigation
for subpixel locations of features
in an image.
The function provided by this module, sfn_correlator()
, provides a spatial correlation routine using pearson
product moments to generate a correlation surface between the template and the image. It does this about a predicted
location of the center of the template in the image, and takes into account 2 masks that specify when to include pixels
in the computation of a pixel score. The computation is parallelized for speed, though, with it being a spatial
correlator, it still is slower than FFT based correlation for large seach distances.
Typically you will not use this function for anything besides doing surface feature navigation, however, if you have another application full documentation is provided. Even when doing surface feature navigation the user typically doesn’t interact with this class directly.
Function
This function performs normalized cross correlation in the spatial domain over a given search distance about a center between an image and a template using masks. |