[Insight-users] Receiving the output of deconvolution filter, pls help!

[Cory Quammen]

Pratik,

I don't know the details about MATLAB's Wiener deconvolution
implementation, but I wouldn't be surprised if it is different from
what I implemented. Looking at the documentation of deconvwnr isn't
much help for figuring out the equation it implements. You have some
choices in how to estimate the power spectral density of the true
signal and noise, so MATLAB may very well use different estimates from
what the ITK filter uses.

Cory

On Sun, May 20, 2012 at 11:35 AM, Pratik Talole <pratiktalole at gmail.com> wrote:
> Ah!! That did the trick. Thanks a ton Cory.
>
> The only thing that now concerns me though is that there is some error
> between the output of matlab and itk.
>
> The error is fluctuating. There is at times even a sign change, but for most
> part, the output of ITK is faithful to the output of matlab upto 1 decimal
> point.
>
> Heres my matlab code :
> x = [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20]
> k = [1 2 3 4 5]
> y = deconvwnr(x,k,1.0)  %%%% output = deconvwnr(input, kernel, SNR)
>
> Is it possible to maybe tweak some part of my current ITK implementation so
> that the output of ITK matches that of Matlab more closely?
>
> A couple of things I am not sure I should have done :
> 1) I set normalize to false
> 2) I set NoiseVariance to 1.0
> 3) I did something about boundary conditions
>
> Regards,
> Pratik
>
> --
> View this message in context: http://itk-insight-users.2283740.n2.nabble.com/Receiving-the-output-of-deconvolution-filter-pls-help-tp7567771p7568011.html
> Sent from the ITK Insight Users mailing list archive at Nabble.com.
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.php
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users



-- 
Cory Quammen
Research Associate
Department of Computer Science
The University of North Carolina at Chapel Hill