Selection of areas of interest.

[Lasse]

Hello Forum!

I'm working with cross-sections of mouse arteries. The purpose is to quantify pixel intensities from the dark areas in between the white curly string-things.

I need an efficient method to select these areas.
Alternatively, I need an efficient method to mask the white strings.

I want to keep my images as 16-bit.

Hope you smart people can help me!

/Lasse

[Kenton Arkill]

Hi
Not the best, and other people are better at defining the shapes etc,
 but a simple way for the short term is the wand (next to the letter A on panel), double click it and increase the tolerance and click between the frilly bits and it fills up most of the areas with each click, then hit "measure". Or threshold it out and multiply the mask by the original to make the wand more efficient. If you are doing >1000 it might be worth doing something cleverer but you can waste more time than you save sometimes.

Kenton Arkill
GB U23 Ladies'
Underwater Hockey Coach



On 25 Jun 2012, at 11:56, Lasse wrote:

> Hello Forum!
>
> I'm working with cross-sections of mouse arteries. The purpose is to
> quantify pixel intensities from the dark areas in between the white curly
> string-things.
>
> I need an efficient method to select these areas.
> Alternatively, I need an efficient method to mask the white strings.
>
> I want to keep my images as 16-bit.
>
> Hope you smart people can help me!
>
> /Lasse
>
> http://imagej.1557.n6.nabble.com/file/n4999184/Screen_Shot_2012-06-25_at_12.47.36_PM.png 
>
>
> --
> View this message in context: http://imagej.1557.n6.nabble.com/Selection-of-areas-of-interest-tp4999184.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

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

[Michael Schmid]

Hi Lasse,

you might try rather strong unsharp masking (weight 0.8, radius 1...2 pixels on the reduced-size 8-bit image; larger radius if you have better resolution) and then use the 'Versatile wand' plugin with gradient detection (approx. 20/pixel with the 8-bit gray level .png image).
The Versatile Wand has preview, so you can easily optimize its parameters.

Your original (larger) image may require an additional step to increase the gradient, e.g. 'thresholded blur' with a threshold about 60% of the difference between the dark background and the darkest parts of the white curly objects.

The plugins I mentioned are at http://imagejdocu.tudor.lu/

Others might have better ideas...

Michael
________________________________________________________________
On Jun 25, 2012, at 12:56, Lasse wrote:

> Hello Forum!
>
> I'm working with cross-sections of mouse arteries. The purpose is to
> quantify pixel intensities from the dark areas in between the white curly
> string-things.
>
> I need an efficient method to select these areas.
> Alternatively, I need an efficient method to mask the white strings.
>
> I want to keep my images as 16-bit.
>
> Hope you smart people can help me!
>
> /Lasse
>
> http://imagej.1557.n6.nabble.com/file/n4999184/Screen_Shot_2012-06-25_at_12.47.36_PM.png 

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