Richardson Lucy Deconvolution

236 views
Skip to first unread message

Luiz Tizei

unread,
May 11, 2015, 8:01:24 AM5/11/15
to hypersp...@googlegroups.com
Hello,

I am trying to use the Richardson Lucy deconvolution algorithm in Hyperspy (richardson_lucy_deconvolution(psf, iterations=15, mask=None, show_progressbar=None)) but I am unable to.
I usually use Hyperspy for PCA only and this is the first time trying deconvolution. One thing that intrigues me is that I cannot auto-complete the name of the function using tab (as it works for everything else). Has this function been removed in one of the updates? I have just re-installed version 0.8 to try to solve this issue.
Thanks for any input.

Best wishes,

Luiz

Tomas Ostasevicius

unread,
May 11, 2015, 8:43:31 AM5/11/15
to hypersp...@googlegroups.com
Hi Luiz,

To the best of my knowledge, the function is still there. The most likely problem that you might have is not setting the signal type correctly, as the deconvolution is only accessible for EELSSpectrum.
The quickest way to change it would be to call `s.set_signal_type('EELS') ` somewhere before the deconvolution call. If you want to check what is the current type, it is specified in the metadata.

As a general note though, if the autocompletion does not find the function, it's either not there or "not allowed" for that particular signal type. For example, the deconvolution, ZLP aligning and other EELS functionalities are only available for EELS spectra, while, `get_lines_intensity` - only for EDS spectra. Taking that even deeper, EDS quantification is only available for TEM and not SEM EDS spectra.

I hope that helps,
Best wishes,

Tomas
--
You received this message because you are subscribed to the Google Groups "hyperspy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hyperspy-user...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Luiz Tizei

unread,
May 11, 2015, 11:09:40 AM5/11/15
to hypersp...@googlegroups.com
Hi Tomas,

Thank you very much. It was indeed the problem.
Best wishes,

Luiz
> You received this message because you are subscribed to a topic in the
> Google Groups "hyperspy-users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/hyperspy-users/bU7vqDMsGu4/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
Reply all
Reply to author
Forward
0 new messages