[ITK-users] Masks for pixelwise normalized correlation

Carlos Hernando Ramiro CHernando at telespazio.es
Mon Jun 9 11:22:07 EDT 2014


Dear all,

I am trying to adapt the code of the example "MeanSquaresImageMetric1.cxx"
to calculate the Normalized Correlation for just one pixel in the fixed
image considering different translations, to see which pixel of the moving
image match with it in the best way.

Then I wonder how must I set:
SetFixedImageRegion
SetFixedImageMask
SetMovingImageMask
in the Normalized Correlation metric.

For instance, shall the fixed mask include the metric window? Is it needed a
moving mask? Shall the fixed region also be restricted?


Thank you in advance and best regards
Carlos

--------------------------------------------
Telespazio
A Finmeccanica/Thales Company

Carlos Hernando Ramiro
SAR R&D Junior Engineer

Telespazio Ibérica
O´Donnell, 18 4to I · 28009 Madrid
Tel. +34 91 520 10 98  Fax +34 93 583 02 01

chernando at telespazio.es
http://www.telespazio.es


More information about the Insight-users mailing list