segmentation of colour images

[Girault France]

Dear all,

 

I have series of color images acquired with a microscope (these images
were taken in phase contrast and have a bright (brighter than the
background) halo around the particles of interest. I would like to
segment these images. When I try to work on their resulting gray value
images, the result is really too bad and I was thus wondering if there
is a standard and usual method applied to segment these type of images :
e.g. by splitting the images in different chanel (green, blue, red) or
other method... Any idea, advice?

 

Many thanks,

 

france

[Maria E. Barilla]

Hello France,

What kind of segmentation algorithm did you use? The easiest I am thinking
is a k-means clustering.

You can try different colour spaces. HSB is straight forward.
Image->Type->HSB Stack.

There are another two diffent colour space transformation in the plugins
page: http://rsb.info.nih.gov/ij/plugins/index.html#color
These are Luv and Lab. Try that and see if the results improve.

Another thing is that you could send me an image to see it so I can have a
better idea of the kind of images you are working with.

Kind regards,

Maria

On Mon, 11 Dec 2006, Girault France wrote:

> Dear all,
>
>
>
> I have series of color images acquired with a microscope (these images
> were taken in phase contrast and have a bright (brighter than the
> background) halo around the particles of interest. I would like to
> segment these images. When I try to work on their resulting gray value
> images, the result is really too bad and I was thus wondering if there
> is a standard and usual method applied to segment these type of images :
> e.g. by splitting the images in different chanel (green, blue, red) or
> other method... Any idea, advice?
>
>
>
> Many thanks,
>
>
>
> france
>

[Barthel, Kai Uwe]

Dear France,

you can use the 3D ColorInspector plugin.

http://rsb.info.nih.gov/ij/plugins/color-inspector.html

Try different color segmentation approaches by selecting an  
appropriate color space, select the segmentation mode. Then turn the  
colors you want to discard to the front and set the depth slider so  
that unwanted colors are removed.

Kai

Am 11.12.2006 um 20:37 schrieb Girault France:

> Dear all,
>
>
>
> I have series of color images acquired with a microscope (these images
> were taken in phase contrast and have a bright (brighter than the
> background) halo around the particles of interest. I would like to
> segment these images. When I try to work on their resulting gray value
> images, the result is really too bad and I was thus wondering if there
> is a standard and usual method applied to segment these type of  
> images :
> e.g. by splitting the images in different chanel (green, blue, red) or
> other method... Any idea, advice?
>
>
>
> Many thanks,
>
>
>
> france

[way4thesub]

The perfect solution was just developed at Stanford University - GemIdent. This open-source free pacakge segments cells from miscroscopic images and, more generally, can identify objects of interest and do color image segmentation.

Adam

Girault France wrote

Dear all,

 

I have series of color images acquired with a microscope (these images
were taken in phase contrast and have a bright (brighter than the
background) halo around the particles of interest. I would like to
segment these images. When I try to work on their resulting gray value
images, the result is really too bad and I was thus wondering if there
is a standard and usual method applied to segment these type of images :
e.g. by splitting the images in different chanel (green, blue, red) or
other method... Any idea, advice?

 

Many thanks,

 

france