Deconvolution Lab "Batch" Module

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
25 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: Deconvolution Lab "Batch" Module

Fairy
if you can, please share me your mail id...

On Wed, Jun 28, 2017 at 6:06 PM, bnorthan [via ImageJ] <[hidden email]> wrote:
Hi Ash

Without seeing the images, it is hard for me to understand what is
happening.  A while back someone had a similar problem, and it turned out
it was caused by a scale bar that had been drawn on the image.  You can
send images to my personal e-mail, or better yet post them somewhere
publicly.

In general the deconvolved image will appear "dimmer" because of the small
bright emitters will have relatively high intensities after deconvolution.
A few beads or bead like objects in the images often causes this affect.
You can reduce the "max" of the display to help visualize the image.

You can also look at line profiles of the objects you are interested in.
And take a look at the statistics of the image.  The total sum should be
similar before and after deconvolution.

Brian

On Wed, Jun 28, 2017 at 8:00 AM, Ash <[hidden email]> wrote:

> yes...it has become more sharper...and i have given normalize psf too!!
>
> but the intensity of deconvolved image is totally different...deconvolved
> images are more dark...and when i am trying to adjust the brightness and
> intensity of true and deconvolved images, they never match...
>
> the true image appears to have more brightness...the intensity is between 0
> and 255... for the other it is just 0.
>
>
> so how can i compare it...
> i wanted to check their snr too..
> is there any other parameter that i can check to acess the quality
> improvement??
>
> On Wed, Jun 28, 2017 at 5:05 PM, bnorthan [via ImageJ] <
> [hidden email]> wrote:

>
> > Hi Ash
> >
> > The dynamic range of a deconvolved image will be larger.  So the output
> > probably will not scale to 8 bit.
> >
> > Also make sure you click the option 'normalize PSF'.  This will make the
> > PSF is normalized to unit energy, and that will mean you can directly
> > compare input and output values.
> >
> > The deconvolved image should have sharper and higher peaks.  This is
> > because blurry "wide" features, should now be narrower and more intense.
> >
> > Brian
> >
> > On Wed, Jun 28, 2017 at 7:06 AM, Ash <[hidden email]
> > <http:///user/SendEmail.jtp?type=node&node=5018980&i=0>> wrote:

> >
> > > i got results..thanks..one more doubt!!
> > > my input image was 8 bits..after deconvolution size is becoming 32
> > > bits...how do i convert output images intensity to that of input
> image??
> > >
> > > i want to compare the images...but since intensities are varying i m
> not
> > > able to compare!!
> > >
> > > On Sun, Jun 25, 2017 at 11:46 PM, Ash Fairy <[hidden email]
> > <http:///user/SendEmail.jtp?type=node&node=5018980&i=1>> wrote:

> > >
> > > > Thank you, I will do that. ..
> > > > Well can I send it to your mail id?
> > > >
> > > > On Jun 25, 2017 6:16 PM, "bnorthan [via ImageJ]"
> > <ml+s1557n5018964h66@n6.
> > > > nabble.com> wrote:
> > > >
> > > >> Hi Ash
> > > >>
> > > >> If I understand you correctly, you need to directly deconvolve the
> > > image.
> > > >> The microscope has already approximately convolved the true object
> > with
> > > >> the
> > > >> PSF.
> > > >>
> > > >> If you are able to share the image and bead image, I am happy to try
> > and
> > > >> deconvolve them, and let you know the best approach and settings.
> > > >>
> > > >> Brian
> > > >>
> > > >> On Sun, Jun 25, 2017 at 12:09 AM, Ash <[hidden email]
> > > >> <http:///user/SendEmail.jtp?type=node&node=5018964&i=0>> wrote:
> > > >>
> > > >> > so before going to deconvolutionLab plugin, first do we have to
> > > >> convolve
> > > >> > the image and its PSF?? And then compare results of image
> convolved
> > > >> with
> > > >> > PSF , with the end result of that blurred image with PSF???
> > > >> >
> > > >> > Or are we directly starting to deconvolve image that we have taken
> > and
> > > >> its
> > > >> > PSF???
> > > >> >
> > > >> > On Sat, May 27, 2017 at 5:19 PM, bnorthan [via ImageJ] <
> > > >> > [hidden email] <http:///user/SendEmail.jtp?
> > > type=node&node=5018964&i=1>>
> > > >> wrote:
> > > >> >
> > > >> > > Hi Ash
> > > >> > >
> > > >> > > If you are using deconvolution lab, it should automatically
> > resize
> > > >> the
> > > >> > PSF
> > > >> > > in x and y.  If I recall correctly deconvolution lab does not
> > resize
> > > >> in
> > > >> > z.
> > > >> > > So you should make the PSF x by y by 37 (z same z in image).
> > Make x
> > > >> and
> > > >> > y
> > > >> > > the right size to contain one bead.  Make sure the bead is
> > centered.
> > > >> > >
> > > >> > > I also would subtract the background level (at least subtract
> the
> > > min
> > > >> > > value
> > > >> > > of the cropped bead image).
> > > >> > >
> > > >> > > If you are able to share any images I can give it a try.
> > > >> > >
> > > >> > > On Wed, May 24, 2017 at 4:54 AM, Ash <[hidden email]
> > > >> > > <http:///user/SendEmail.jtp?type=node&node=5018792&i=0>> wrote:
> > > >> > >
> > > >> > > > hii
> > > >> > > > thank you for replying me..yes..i do have multiple bead...the
> > psf
> > > >> has
> > > >> > 99
> > > >> > > > frames and image has 37 frames....if i am cropping a single
> > bead
> > > >> out
> > > >> > > then
> > > >> > > > how should i adjust the size to make it compatible with the
> raw
> > > >> > > > image???because raw images size is 2048*2048 and bead size is
> > > >> > > 512......if i
> > > >> > > > crop out a single bead, then the bead file will become more
> > > >> smaller!!!
> > > >> > > >
> > > >> > > > On Tue, May 23, 2017 at 8:00 PM, bnorthan [via ImageJ] <
> > > >> > > > [hidden email] <http:///user/SendEmail.jtp?
> > > >> > type=node&node=5018792&i=1>>
> > > >> > > wrote:
> > > >> > > >
> > > >> > > > > Hi Ash
> > > >> > > > >
> > > >> > > > > If you can share your image and psf I can try and help you.
> > > From
> > > >> > your
> > > >> > > > > description I suspect your image contains multiple beads.
> To
> > > use
> > > >> it
> > > >> > > as a
> > > >> > > > > PSF you would have to crop out one bead.
> > > >> > > > >
> > > >> > > > > Brian
> > > >> > > > >
> > > >> > > > > On Mon, May 15, 2017 at 5:45 PM, Ash <[hidden email]
> > > >> > > > > <http:///user/SendEmail.jtp?type=node&node=5018778&i=0>>
> > wrote:
> > > >> > > > >
> > > >> > > > > > my input image size is 2048*2048*37 and psf size is
> > > >> > > 512*512*99....how
> > > >> > > > > > should
> > > >> > > > > > i resize it?do i have to do fft or ifft to get
> > deconvolution
> > > >> > > > result..the
> > > >> > > > > > psf
> > > >> > > > > > file that i have contain several beads....should i take
> > > average
> > > >> of
> > > >> > > all
> > > >> > > > > the
> > > >> > > > > > beads in that file to deconvolution.... or just uploading
> > the
> > > >> psf
> > > >> > > file
> > > >> > > > > with
> > > >> > > > > > sveral beads is fine?
> > > >> > > > > > i am doing with deconvolution lab software...how should i
> > > >> proceed?
> > > >> > > > > >
> > > >> > > > > >
> > > >> > > > > >
> > > >> > > > > > --
> > > >> > > > > > View this message in context:
> http://imagej.1557.x6.nabble.
> >
> > > >> > > > > > com/Deconvolution-Lab-Batch-Module-tp5015776p5018724.html
> > > >> > > > > > Sent from the ImageJ mailing list archive at Nabble.com.
> > > >> > > > > >
> > > >> > > > > > --
> > > >> > > > > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >> > > > > >
> > > >> > > > >
> > > >> > > > > --
> > > >> > > > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >> > > > >
> > > >> > > > >
> > > >> > > > > ------------------------------
> > > >> > > > > If you reply to this email, your message will be added to
> the
> > > >> > > discussion
> > > >> > > > > below:
> > > >> > > > > http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-
> > > >> > > > > Module-tp5015776p5018778.html
> > > >> > > > > To unsubscribe from Deconvolution Lab "Batch" Module, click
> > here
> > > >> > > > > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > >> > > > macro=unsubscribe_by_code&node=5015776&code=
> > > >> > cGhvZWluaXg5M0BnbWFpbC5jb218NT
> > > >> > >
> > > >> > > > AxNTc3NnwtNTkxNTYyNDE=>
> > > >> > > > > .
> > > >> > > > > NAML
> > > >> > > > > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > >> > > > macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&
> > > >> > > > base=nabble.naml.namespaces.BasicNamespace-nabble.view.
> > > >> > > > web.template.NabbleNamespace-nabble.view.web.template.
> > > >> > > > NodeNamespace&breadcrumbs=notify_subscribers%21nabble%
> > > >> > > > 3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_
> > > >> > > > instant_email%21nabble%3Aemail.naml>
> > > >> > > > >
> > > >> > > >
> > > >> > > >
> > > >> > > >
> > > >> > > >
> > > >> > > > --
> > > >> > > > View this message in context: http://imagej.1557.x6.nabble.
> > > >> > > > com/Deconvolution-Lab-Batch-Module-tp5015776p5018779.html
> > > >> > > > Sent from the ImageJ mailing list archive at Nabble.com.
> > > >> > > >
> > > >> > > > --
> > > >> > > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >> > > >
> > > >> > >
> > > >> > > --
> > > >> > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >> > >
> > > >> > >
> > > >> > > ------------------------------
> > > >> > > If you reply to this email, your message will be added to the
> > > >> discussion
> > > >> > > below:
> > > >> > > http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-
> > > >> > > Module-tp5015776p5018792.html
> > > >> > > To unsubscribe from Deconvolution Lab "Batch" Module, click here
> > > >> > > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > >> > macro=unsubscribe_by_code&node=5015776&code=
> > > cGhvZWluaXg5M0BnbWFpbC5jb218NT
> > > >>
> > > >> > AxNTc3NnwtNTkxNTYyNDE=>
> > > >> > > .
> > > >> > > NAML
> > > >> > > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > >> > macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&
> > > >> > base=nabble.naml.namespaces.BasicNamespace-nabble.view.
> > > >> > web.template.NabbleNamespace-nabble.view.web.template.
> > > >> > NodeNamespace&breadcrumbs=notify_subscribers%21nabble%
> > > >> > 3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_
> > > >> > instant_email%21nabble%3Aemail.naml>
> > > >> > >
> > > >> >
> > > >> >
> > > >> >
> > > >> >
> > > >> > --
> > > >> > View this message in context: http://imagej.1557.x6.nabble.
> > > >> > com/Deconvolution-Lab-Batch-Module-tp5015776p5018963.html
> > > >> > Sent from the ImageJ mailing list archive at Nabble.com.
> > > >> >
> > > >> > --
> > > >> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >> >
> > > >>
> > > >> --
> > > >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > > >>
> > > >>
> > > >> ------------------------------
> > > >> If you reply to this email, your message will be added to the
> > discussion
> > > >> below:
> > > >> http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-Mod
> > > >> ule-tp5015776p5018964.html
> > > >> To unsubscribe from Deconvolution Lab "Batch" Module, click here
> > > >> <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > macro=unsubscribe_by_code&node=5015776&code=
> cGhvZWluaXg5M0BnbWFpbC5jb218NT
> >
> > > AxNTc3NnwtNTkxNTYyNDE=>
> > > >> .
> > > >> NAML
> > > >> <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> > > macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&
> > > base=nabble.naml.namespaces.BasicNamespace-nabble.view.
> > > web.template.NabbleNamespace-nabble.view.web.template.
> > > NodeNamespace&breadcrumbs=notify_subscribers%21nabble%
> > > 3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_
> > > instant_email%21nabble%3Aemail.naml>
> > > >>
> > > >
> > >
> > >
> > >
> > >
> > > --
> > > View this message in context: http://imagej.1557.x6.nabble.
> > > com/Deconvolution-Lab-Batch-Module-tp5015776p5018979.html

> > > Sent from the ImageJ mailing list archive at Nabble.com.
> > >
> > > --
> > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> > >
> >
> > --
> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >
> >
> > ------------------------------
> > If you reply to this email, your message will be added to the discussion
> > below:
> > http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-
> > Module-tp5015776p5018980.html
> > To unsubscribe from Deconvolution Lab "Batch" Module, click here

> > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> macro=unsubscribe_by_code&node=5015776&code=cGhvZWluaXg5M0BnbWFpbC5jb218NT
> AxNTc3NnwtNTkxNTYyNDE=>
> > .
> > NAML
> > <http://imagej.1557.x6.nabble.com/template/NamlServlet.jtp?
> macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&
> base=nabble.naml.namespaces.BasicNamespace-nabble.view.
> web.template.NabbleNamespace-nabble.view.web.template.
> NodeNamespace&breadcrumbs=notify_subscribers%21nabble%
> 3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_
> instant_email%21nabble%3Aemail.naml>
> >
>
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.
> com/Deconvolution-Lab-Batch-Module-tp5015776p5018981.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html



If you reply to this email, your message will be added to the discussion below:
http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-Module-tp5015776p5018982.html
To unsubscribe from Deconvolution Lab "Batch" Module, click here.
NAML

Reply | Threaded
Open this post in threaded view
|

Re: Deconvolution Lab "Batch" Module

Fairy
In reply to this post by Fairy
I didn't understand  Normalize Psf..and conversion  of it to unit energy....
And comparing  the 8 bit image or 16 bit image with the output of
deconvolution which is 32 bit by just giving normalize Psf...

 I saw in a paper that the image becomes 3 2  bit since it is successful  in
reassiging  the diffracted lights to its original location.. Is it true??
Please help!!



--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Deconvolution Lab "Batch" Module

Fairy
In reply to this post by Fairy
Suppose if I have averaged the beads and has padded zeroes to make the size
of original image and Psf same,  then is it wrong!? 1??



--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Deconvolution Lab "Batch" Module

bnorthan
Hi Ash

If you cropped out several beads, and managed to align them accurately,
then averaged them, that would be fine.

On Mon, Oct 2, 2017 at 4:57 AM, Fairy <[hidden email]> wrote:

> Suppose if I have averaged the beads and has padded zeroes to make the size
> of original image and Psf same,  then is it wrong!? 1??
>
>
>
> --
> Sent from: http://imagej.1557.x6.nabble.com/
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Deconvolution Lab "Batch" Module

Fairy
This post has NOT been accepted by the mailing list yet.
Great, thanks!.. But right now I have zero padded it and made it as the same size as input image... Is it wrong??? 

I was going through richardson lucy algorithm... I havint completely understood it.. Which step in that derivation makes the reassignment of diffracted lights?? 

How is the averaged PSF deconvolving each pixels... In my case averaged PSF and input image are of same size... I cannot relate how the PSF will be reassigned to a single focal spot

Plz help

On 02-Oct-2017 4:36 PM, "bnorthan [via ImageJ]" <[hidden email]> wrote:
Hi Ash

If you cropped out several beads, and managed to align them accurately,
then averaged them, that would be fine.

On Mon, Oct 2, 2017 at 4:57 AM, Fairy <[hidden email]> wrote:

> Suppose if I have averaged the beads and has padded zeroes to make the size
> of original image and Psf same,  then is it wrong!? 1??
>
>
>
> --
> Sent from: http://imagej.1557.x6.nabble.com/

>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html



If you reply to this email, your message will be added to the discussion below:
http://imagej.1557.x6.nabble.com/Deconvolution-Lab-Batch-Module-tp5015776p5019486.html
To unsubscribe from Deconvolution Lab "Batch" Module, click here.
NAML

12