Hi,
I have two ideas which might improve your code.
-
allowing users to provide a mask
E.g. frequently large part of the field of view is irrelevant, e.g. in CT the image has frequently
a circular outer edge, and outside this edge there is no meaningful value.
It could speed up the code if you would ignore these parts.
-
it think it could be quite helpful if the users could provide their own weight function.
(instead of the standard exp(- pixel value difference))
E.g. I have a vectorial image (dual energy CT), and I would like to use a special weight.
(I can modify the code on my own, but I think it would be a nice improvement.)
Hi,
I have two ideas which might improve your code.
allowing users to provide a mask
E.g. frequently large part of the field of view is irrelevant, e.g. in CT the image has frequently
a circular outer edge, and outside this edge there is no meaningful value.
It could speed up the code if you would ignore these parts.
it think it could be quite helpful if the users could provide their own weight function.
(instead of the standard exp(- pixel value difference))
E.g. I have a vectorial image (dual energy CT), and I would like to use a special weight.
(I can modify the code on my own, but I think it would be a nice improvement.)