Hi, no ready solution, only a first step:
Detection of the membranes (and inside of the membranes the cells):
You have to download the Morphology plugins from Gabriel Landini and the watershed plugin from EFL Lausanne
selectWindow("YA3.jpg");
run("8-bit"); or take the green channel only
run("HMinima Transform", "h=4"); # from Gabriel
on the result hmin do
run("Watershed Segmentation");
do Smooth with Radius 3.0 and than Start Watershed with Dark objects, 4-connected, Min/Max 0 255, Display Object/Background binary.
You will get the membranes in white and the cells in black.
Intensities:
Global membrane intensity could be calculated from slightly dilated membranes redirected to the original intensities.
Membrane intensity per cell: select one cell, make a binary image from it, dilate it (one more than the membrane dilation), dilate the membranes and calculate the intensity on the logical AND of the dilated cell and the dilated membrane, again by redirection to the original intensity image. Of course neighboring cells will share membrane intensities.
Hope that helps
Karsten
Am 30.04.2012 um 20:37 schrieb RBotchlett:
> Hi,
>
> I am trying to measure only the membrane intensity on the following
> image:
http://imagej.1557.n6.nabble.com/file/n4941668/YA3.jpg
>
> I understand that I need to use some kind of threshold but I'm not
> certain how to set the correct threshold for just the membrane. As you can
> see on the image, there is some light green staining on the inside of the
> cells, but the only fluroescence we want to analyze is whats on the membrane
> of each cell.
>
> Also, is there is a specific plugin or macro that can do this? I have
> found a few (i.e. edge_filter) but have either had problems interpretting
> the results or can't figure out how to set all the adjustments.
>
> I appreciate any help or feedback!
>
> Thank you,
> Rachel
>
> --
> View this message in context:
http://imagej.1557.n6.nabble.com/Measuring-Membrane-Fluorescence-tp4941668.html> Sent from the ImageJ mailing list archive at Nabble.com.
Karsten
[hidden email]