filter particles

[Christian Tischer]

hello,

i am using "Analyze Particles" for finding particles and adding them to the
ROI manager.

afterward i want to filter the particles according to certain criteria
(e.g. mean intensity of the ROIs in a certain image a.s.o.)

one way is writing a for-next loop over the ROIs, compute for each ROI what
i want and then kick out the ROIs that are "not good".

i was just wondering whether there exists already a plugin for this
workflow, something like "Filter Particles".

otherwise i would write one :-)

cheers, Tischi

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

[Gabriel Landini]

On Thursday 16 Aug 2012 10:38:07 you wrote:
> i am using "Analyze Particles" for finding particles and adding them to the
> ROI manager.
> afterward i want to filter the particles according to certain criteria
> (e.g. mean intensity of the ROIs in a certain image a.s.o.)
>
> one way is writing a for-next loop over the ROIs, compute for each ROI what
> i want and then kick out the ROIs that are "not good".

I do this using the Particles8 plugin on a binary image with the "morphology"
and "redirect" options. This collects shape, size and geyscale/colour
information that can then accessed from the Results Table. This is available
in the Morphology zip in my page.
Some macros there show how to delete particles, the Delete_ZeroArea uses the
flood fill of the area selected with the wand on the original image, while the
KeepParticlesInRange macro uses binary reconstruction.
Hope it helps

Gabriel

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

[PEARSON Matthew]

Hi all,

I have some images of embryos captured using a colour camera  
outputting RGB images.  I have a batch of images of mouse embryos  
captured in a petri dish without agar which gave the images a nice  
black background almost darkfield in appearance.  I then captured some  
images later of different embryos but then the petri dish had agar in  
and this has given a grey/blue background with the embryo not so well  
contrasted and i want to put them all in a figure and have a uniform  
background.

So in essence i want to change the background of the later images  
(grey/blue) background to black without affecting the colour of the  
embryo.  I didn't think this would be a straight forward process on a  
colour image.  Does anyone have any suggestions?  I have looked at the  
background subtraction commands and this might be able to get me there  
but thought this was more for fluorescence images and not sure its the  
best way..  I also wondered about doing a colour threshold to identify  
the embryo then setting the pixel value outside of the thresholded  
embryo to zero if you can do that but i think this will will give a  
jagged outline to the embryo.

Any ideas would be greatly appreciated.

Thanks,

Matt

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

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

[Anderson, Charles (DNR)]

Consider the SIOX plugin (Simple Interactive Object Extraction).

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Matthew Pearson
Sent: Thursday, August 16, 2012 5:25 AM
To: [hidden email]
Subject: Changing background colour on colour RGB images

Hi all,

I have some images of embryos captured using a colour camera outputting RGB images.  I have a batch of images of mouse embryos captured in a petri dish without agar which gave the images a nice black background almost darkfield in appearance.  I then captured some images later of different embryos but then the petri dish had agar in and this has given a grey/blue background with the embryo not so well contrasted and i want to put them all in a figure and have a uniform background.

So in essence i want to change the background of the later images
(grey/blue) background to black without affecting the colour of the embryo.  I didn't think this would be a straight forward process on a colour image.  Does anyone have any suggestions?  I have looked at the background subtraction commands and this might be able to get me there but thought this was more for fluorescence images and not sure its the best way..  I also wondered about doing a colour threshold to identify the embryo then setting the pixel value outside of the thresholded embryo to zero if you can do that but i think this will will give a jagged outline to the embryo.

Any ideas would be greatly appreciated.

Thanks,

Matt

--
The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

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

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

[Glen MacDonald-2]

Dear Tischi,
If you are interested in editing  only the results, Run Analyze Particles then step through the results table and remove rows with intensities that are out of range with
'IJ.deleteRows(index1,index2)'

The shape descriptors and fraction area options in Set Measurements will provide additional parameters for discriminating your particles.

Regards,
Glen


Glen MacDonald
Cellular Morphology Core
Center for Human Development and Disability
Box 357920
University of Washington
Seattle, WA 98195-7920  USA
[phone]
[hidden email]


On Aug 16, 2012, at 2:38 AM, Christian Tischer wrote:

> hello,
>
> i am using "Analyze Particles" for finding particles and adding them to the
> ROI manager.
>
> afterward i want to filter the particles according to certain criteria
> (e.g. mean intensity of the ROIs in a certain image a.s.o.)
>
> one way is writing a for-next loop over the ROIs, compute for each ROI what
> i want and then kick out the ROIs that are "not good".
>
> i was just wondering whether there exists already a plugin for this
> workflow, something like "Filter Particles".
>
> otherwise i would write one :-)
>
> cheers, Tischi
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

... [show rest of quote]

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

[Christian Tischer]

Dear Glen,

yes, good point, in fact, that's how i implemented it now.
running through the results table seems efficient and one can use all the
nice "Measurement" features of ImageJ.
if i wanted i could then probably also delete the "not good" ROIs from the
ROI manager.

Tischi




On Thu, Aug 16, 2012 at 5:26 PM, Glen MacDonald <[hidden email]>wrote:

> Dear Tischi,
> If you are interested in editing  only the results, Run Analyze Particles
> then step through the results table and remove rows with intensities that
> are out of range with
> 'IJ.deleteRows(index1,index2)'
>
> The shape descriptors and fraction area options in Set Measurements will
> provide additional parameters for discriminating your particles.
>
> Regards,
> Glen
>
>
> Glen MacDonald
> Cellular Morphology Core
> Center for Human Development and Disability
> Box 357920
> University of Washington
> Seattle, WA 98195-7920  USA
> [phone]
> [hidden email]
>
>
> On Aug 16, 2012, at 2:38 AM, Christian Tischer wrote:
>
> > hello,
> >
> > i am using "Analyze Particles" for finding particles and adding them to
> the
> > ROI manager.
> >
> > afterward i want to filter the particles according to certain criteria
> > (e.g. mean intensity of the ROIs in a certain image a.s.o.)
> >
> > one way is writing a for-next loop over the ROIs, compute for each ROI
> what
> > i want and then kick out the ROIs that are "not good".
> >
> > i was just wondering whether there exists already a plugin for this
> > workflow, something like "Filter Particles".
> >
> > otherwise i would write one :-)
> >
> > cheers, Tischi
> >
> > --
> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

... [show rest of quote]

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