plugin for crowded-field photometry (fitting overlapping identical objects)?

[Michael Schmid]

Hi everyone,

is there some out there who knows about an ImageJ plugin that does roughly
what astronomers call "crowded-field photometry"?

I.e., imagine that you have stars that appear slightly blurred, and the
images of the individual stars are overlapping.  Apart form different
brightness, the stars are identical (identical point spread function PSF).
 In super-resolution microscopy one might encounter the same problem if
there is an overlap of the image of neighboring fluorescent particles.

Such images cloud be analyzed by deconvolution, but this has some
limitations.  I am looking for a plugin that tries to fit the image with
the sum of a large number of PSF functions (preferably with subpixel
positions).  Probably 2D Gaussians would be sufficient.

If someone is aware of an easy-to-understand algorithm or code that could
be converted into an ImageJ plugin, such hints would be also helpful.
[For the astronomers among you: yes, I am aware that there is quite some
literature on crowded-field photometry, but delving into it and converting
one of these methods into code is beyond what I could do within reasonable
time.]


Thank you,

Michael

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

[lechristophe]

Hi Michael,

I don't know about astronomy, but in single molecule localization
microscopy (PALM/STORM) there are a few algorithms that are available for
high density data.
- ThunderSTORM is an ImageJ plugin that has a "high density" mode:
https://github.com/zitmen/thunderstorm
- DAOSTORM is actually adapted from an astronomy crowded field algorithm I
think. It's Python and can be downloaded here:
https://seamusholden.wordpress.com/tag/daostorm/
- There is a 3D version of DAOSTORM here (also Python):
https://github.com/ZhuangLab/storm-analysis/tree/master/3d_daostorm

Hope this helps, and happy new year to everyone on the list!

Christophe


--
Christophe Leterrier
Researcher
Axonal Domains Architecture Team
CRN2M CNRS UMR 7286
Aix Marseille University, France


On Sun, Jan 3, 2016 at 5:12 PM, Michael Schmid <[hidden email]>
wrote:

> Hi everyone,
>
> is there some out there who knows about an ImageJ plugin that does roughly
> what astronomers call "crowded-field photometry"?
>
> I.e., imagine that you have stars that appear slightly blurred, and the
> images of the individual stars are overlapping.  Apart form different
> brightness, the stars are identical (identical point spread function PSF).
>  In super-resolution microscopy one might encounter the same problem if
> there is an overlap of the image of neighboring fluorescent particles.
>
> Such images cloud be analyzed by deconvolution, but this has some
> limitations.  I am looking for a plugin that tries to fit the image with
> the sum of a large number of PSF functions (preferably with subpixel
> positions).  Probably 2D Gaussians would be sufficient.
>
> If someone is aware of an easy-to-understand algorithm or code that could
> be converted into an ImageJ plugin, such hints would be also helpful.
> [For the astronomers among you: yes, I am aware that there is quite some
> literature on crowded-field photometry, but delving into it and converting
> one of these methods into code is beyond what I could do within reasonable
> time.]
>
>
> Thank you,
>
> Michael
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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