binary image processing | delete small forks/dots

[Rainer M. Engel]

Hello everyone,

I'm very confident to bring things here a step further by transferring
some processing elements to ImageJ and I came across some points
concerning binary images.

I made a skeletonized version of an image and get lots of small dots or
masks (via Analyse Particles) with small areas of perhaps 1 to 8 px. It
would be relatively easy to filter them by the results table information
and delete corresponding masks from the selection (ROI Manager) i.e. via
macro.

But though this seems to be a very important step like De-speckle or
Median on typical images, I wonder If there are already tools/plug-ins
for that kind of purpose.

Maybe there are workflows/plugins for these szenarios:
- delete small dots/forks from binary/skeletonized images by area-size
- connect lines/forks via distance value


Any thoughts kindly appreciated.

Regards,
Rainer



--
Rainer M. Engel, Dipl. Digital Artist
scientific|Media GbR
Pichelsdorferstr. 143
D-13595 Berlin

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

[Michael Schmid]

Hi Rainer,

you can remove single foreground pixels in a binary image by Process>Binary>Erode, if you set the 'count' in the binary Options to 8.

For larger objects, use 'Analyze Particles' with a size range and 'Show: masks' as output.

For pruning side branches of a skeleton, see Gabriel Landini's page
  http://www.dentistry.bham.ac.uk/landinig/software/software.html

(there you will find several different plugins and macros for it; you have to try which one is best for you)

Hope this helps,

Michael
________________________________________________________________
On Jun 18, 2012, at 11:21, Rainer M. Engel wrote:

> Hello everyone,
>
> I'm very confident to bring things here a step further by transferring
> some processing elements to ImageJ and I came across some points
> concerning binary images.
>
> I made a skeletonized version of an image and get lots of small dots or
> masks (via Analyse Particles) with small areas of perhaps 1 to 8 px. It
> would be relatively easy to filter them by the results table information
> and delete corresponding masks from the selection (ROI Manager) i.e. via
> macro.
>
> But though this seems to be a very important step like De-speckle or
> Median on typical images, I wonder If there are already tools/plug-ins
> for that kind of purpose.
>
> Maybe there are workflows/plugins for these szenarios:
> - delete small dots/forks from binary/skeletonized images by area-size
> - connect lines/forks via distance value
>
>
> Any thoughts kindly appreciated.
>
> Regards,
> Rainer
>
>
>
> --
> Rainer M. Engel, Dipl. Digital Artist
> scientific|Media GbR
> Pichelsdorferstr. 143
> D-13595 Berlin
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

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

[Rainer M. Engel]

Hi Michael and thank you for your tips. I'll have a look at that and see
how far I come with it.

Thanks,
Rainer


Am 19.06.2012 11:22, schrieb Michael Schmid:

> Hi Rainer,
>
> you can remove single foreground pixels in a binary image by Process>Binary>Erode, if you set the 'count' in the binary Options to 8.
>
> For larger objects, use 'Analyze Particles' with a size range and 'Show: masks' as output.
>
> For pruning side branches of a skeleton, see Gabriel Landini's page
>   http://www.dentistry.bham.ac.uk/landinig/software/software.html
>
> (there you will find several different plugins and macros for it; you have to try which one is best for you)
>
> Hope this helps,
>
> Michael
> ________________________________________________________________
> On Jun 18, 2012, at 11:21, Rainer M. Engel wrote:
>
>> Hello everyone,
>>
>> I'm very confident to bring things here a step further by transferring
>> some processing elements to ImageJ and I came across some points
>> concerning binary images.
>>
>> I made a skeletonized version of an image and get lots of small dots or
>> masks (via Analyse Particles) with small areas of perhaps 1 to 8 px. It
>> would be relatively easy to filter them by the results table information
>> and delete corresponding masks from the selection (ROI Manager) i.e. via
>> macro.
>>
>> But though this seems to be a very important step like De-speckle or
>> Median on typical images, I wonder If there are already tools/plug-ins
>> for that kind of purpose.
>>
>> Maybe there are workflows/plugins for these szenarios:
>> - delete small dots/forks from binary/skeletonized images by area-size
>> - connect lines/forks via distance value
>>
>>
>> Any thoughts kindly appreciated.
>>
>> Regards,
>> Rainer
>>
>>
>>
>> --
>> Rainer M. Engel, Dipl. Digital Artist
>> scientific|Media GbR
>> Pichelsdorferstr. 143
>> D-13595 Berlin
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

[Antoinette]

As long as we choose and use a relibale image processing tool, we can easily edit and process image and document file in varibale image mode, such as binary, bitonal image, monochrome image and RGB and CMYK image.