Re: remove noise ( Iterative deconvolution)

[Runa Hamid]

Dear Zhengyu,
Thank you for the suggestions!

There is a plugin ' Iterative deconvolution' available in image J. Can I use
it for deconvoluting my time lapse images? I am slightly confused how do I
decide upon the last four options in this window.

 Runa


On Mon, Jun 15, 2009 at 9:39 PM, Pang, Zhengyu (GE, Research) <
[hidden email]> wrote:

> Runa,
>
> I don't know what you mean by photometircaly accurate. There are quite
> many factors will affect your final pixel intensity. Potentially it is
> hard to get to know how many photons getting to each pixel. If you want
> to remove noise that is arising from neighboring pixels, then what you
> need to do is to do a deconvolution. You need to know the point spread
> function (PSF) of your optical system. You need to purchase some
> sub-micron beads to measure it. Confocal microscope should have a small
> PSF compared to traditional epifluorescence microscope.
>
> I coauthored with Lavender at Biomaterials, and I did the experiment on
> point spread function measurement. If you are interested you should be
> able to google search it.
>
> Best regards,
>
> Zhengyu PANG
>
>
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Runa Hamid
> Sent: Monday, June 15, 2009 7:46 AM
> To: [hidden email]
> Subject: remove noise
>
> Dear Image J users,
>
> I have intensity values for each pixel in a given ROI in a time lapse
> series. I need to do some quantitation but before that I want to make it
> photometrically accurate. In other words I want to remove all noise that
> is arising in each pixel from neighbouring pixels, optical defects or
> electronic noises of a confocal system.
>
> I would be grateful if anyone can give me some useful tips/ suggestions
> to do it.
>
> Thanks in advance,
> Runa
>

[Pang, Zhengyu (GE Global Research)]

Runa,

I suppose that you should be able to use it. But I believe that
iterative deconvulution is a blind deconvolution method. It will do you
better if you read a few papers to understand this method and decide how
you want to use it. In addition, you can always ask the authors who
wrote this plugin for more information.

Best,

Zhengyu


Dr. Zhengyu Pang
Biochemistry and Biological Engineering Laboratory
General Electric Company Global Research
 
GE Imagination at work

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Runa Hamid
Sent: Wednesday, June 17, 2009 1:58 AM
To: [hidden email]
Subject: Re: remove noise ( Iterative deconvolution)

Dear Zhengyu,
Thank you for the suggestions!

There is a plugin ' Iterative deconvolution' available in image J. Can I
use it for deconvoluting my time lapse images? I am slightly confused
how do I decide upon the last four options in this window.

 Runa


On Mon, Jun 15, 2009 at 9:39 PM, Pang, Zhengyu (GE, Research) <
[hidden email]> wrote:

> Runa,
>
> I don't know what you mean by photometircaly accurate. There are quite

> many factors will affect your final pixel intensity. Potentially it is

> hard to get to know how many photons getting to each pixel. If you
> want to remove noise that is arising from neighboring pixels, then
> what you need to do is to do a deconvolution. You need to know the
> point spread function (PSF) of your optical system. You need to
> purchase some sub-micron beads to measure it. Confocal microscope
> should have a small PSF compared to traditional epifluorescence
microscope.
>
> I coauthored with Lavender at Biomaterials, and I did the experiment
> on point spread function measurement. If you are interested you should

> be able to google search it.
>
> Best regards,
>
> Zhengyu PANG
>
>
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Runa Hamid
> Sent: Monday, June 15, 2009 7:46 AM
> To: [hidden email]
> Subject: remove noise
>
> Dear Image J users,
>
> I have intensity values for each pixel in a given ROI in a time lapse
> series. I need to do some quantitation but before that I want to make
> it photometrically accurate. In other words I want to remove all noise

> that is arising in each pixel from neighbouring pixels, optical
> defects or electronic noises of a confocal system.
>
> I would be grateful if anyone can give me some useful tips/
> suggestions to do it.
>
> Thanks in advance,
> Runa
>

[John Oreopoulos]

Dear ImageJ list,

I've looked in the listserver archive about this topic but couldn't  
find anything conclusive. I have a 32-bit image on which I'd like to  
set a specific threshold range (0.00 to 0.65) in a macro and then  
have the macro perform automatically the same action as clicking  
"apply". A message checkbox usually appears with the title "NaN  
background" and I'd like to have this option set to true (box  
checked) and performed. I've tried using the standard threshold macro  
functions but they're not behaving the same as when I do this by hand  
in ImageJ. Any thoughts?

John Oreopoulos

[John Oreopoulos]

I just found out there is a "NaN background" command which does the  
trick after setting the threshold. Just in case anyone else needs to  
do this in the future, here's the short macro commands that does what  
I was describing:

selectWindow("Image");
setThreshold(0.00, 0.65);
run("NaN Background");
resetThreshold();

John Oreopoulos


On 17-Jun-09, at 12:42 PM, John Oreopoulos wrote:

> Dear ImageJ list,
>
> I've looked in the listserver archive about this topic but couldn't  
> find anything conclusive. I have a 32-bit image on which I'd like  
> to set a specific threshold range (0.00 to 0.65) in a macro and  
> then have the macro perform automatically the same action as  
> clicking "apply". A message checkbox usually appears with the title  
> "NaN background" and I'd like to have this option set to true (box  
> checked) and performed. I've tried using the standard threshold  
> macro functions but they're not behaving the same as when I do this  
> by hand in ImageJ. Any thoughts?
>
> John Oreopoulos

[Robert Dougherty]

Runa and Zhengyu,

If you are referring to my iterative deconvolution plugin, it is PSF-
based, not blind.  The last few parameters are things you experiment  
with until you get results you like.  One of the parameters is the  
number of iterations.  Several people have pointed out, and I have  
found in my own experience, that the results get worse if you iterate  
too many times.  The suggestion is to try a few values and see what  
works best.  The last checkbox, about dB, is only for acoustics.  Do  
not check it.

Bob

On Jun 17, 2009, at 7:04 AM, Pang, Zhengyu (GE, Research) wrote:

> Runa,
>
> I suppose that you should be able to use it. But I believe that
> iterative deconvulution is a blind deconvolution method. It will do  
> you
> better if you read a few papers to understand this method and  
> decide how
> you want to use it. In addition, you can always ask the authors who
> wrote this plugin for more information.
>
> Best,
>
> Zhengyu
>
>
> Dr. Zhengyu Pang
> Biochemistry and Biological Engineering Laboratory
> General Electric Company Global Research
>
> GE Imagination at work
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Runa Hamid
> Sent: Wednesday, June 17, 2009 1:58 AM
> To: [hidden email]
> Subject: Re: remove noise ( Iterative deconvolution)
>
> Dear Zhengyu,
> Thank you for the suggestions!
>
> There is a plugin ' Iterative deconvolution' available in image J.  
> Can I
> use it for deconvoluting my time lapse images? I am slightly confused
> how do I decide upon the last four options in this window.
>
>  Runa
>
>

Robert Dougherty, Ph.D.
President, OptiNav, Inc.
4176 148th Ave. NE
Redmond, WA 98052
Tel. [phone]
FAX [phone]
www.optinav.com
[hidden email]