Posted by
Piotr Wendykier on
Feb 06, 2009; 1:36pm
URL: http://imagej.273.s1.nabble.com/Help-to-use-the-Parallel-Spectral-Deconvolution-Piotr-Wndykier-tp3693826p3693827.html
Dear Dani,
> I am a PhD student at the University of Barcelona, at the Department of
> Ecology. I am new in ImageJ, and I am interested in restoring 3D widefield
> images of bacteria and algae for intensity quantification purposes.
>
> I found the plugin Parallel Spectral Deconvolution (Piotr Wndykier) very
> interesting.
You will find answers to some of your questions here:
http://n2.nabble.com/Parallel-Spectral-Deconvolution-plugin-update-td635552.html#a635552> I would need real restoration algorithms (constrained iterative or
> statistical image restoration, p.e, adn NOT no neighbour, closest neighbour
> or linear algorithms). Are the implemented algorithms (Generalized Tikhonov,
> Tikhonov and TSVD image deblurring) real 3D restorative?
The algorithms implemented in Parallel Spectral Deconvolution are fairly simple.
I have also implemented another plugin - Parallel Iterative
Deconvolution, available at
http://piotr.wendykier.googlepages.com/iterativedeconvolutionThis plugin has four iterative algorithms to choose from and two of
them are nonnegatively
constrained. The description of the algorithms (links to the articles)
is on the plugin's website.
> What kind of PSF is used? experimental or theoretical? If experimental, what
> are the requirements? Must it be dyadic-sized? Must it be the same size as
> the blurred image? Must it be an image with just a single microbead
> recorded? Must the microbead be centered?
There are no limitations and Parallel Iterative Deconvolution supports
spatially variant PSFs.
> Finally, is there any User's Guide?
There is no user's guide, since there is no single recipe to do the
deconvolution.
You just need to try all the methods that I provide with different parameters
and hopefully one of them will work for you.
Regards,
Piotr Wendykier