[ITK-dev] description AdaptiveHistogramEqualizationImageFilter

[Johannes Keustermans]

Dear all,

 

 

The description of the AdaptiveHistogramEqualizationImageFilter is not
entirely correct in my (humble) opinion. From the description it seems that
the filter, depending on the parameter alpha, ranges from standard histogram
equalization (alpha = 0) to unsharp masking (alpha = 1). However, from the
paper and the implementation, the parameter alpha is used to steer between
standard histogram equalization and local mean subtraction. The rationale
for this connection is due to the fact that local mean subtraction is an
approximation for histogram equalization with a strongly gaussian smoothed
histogram. The parameter beta controls the level of local mean addition.

 

 

Kind regards,

Johannes Keustermans

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/insight-developers/attachments/20161025/a21a1111/attachment.html>