Hi Louis,
probably your roi contains no pixels. You can have such Rois, e.g. with
the following macro command, which creates a roi where the line folds
back on itself, without encircling any pixels:
makePolygon(7,8, 6,7, 7,8);
or (L-shaped):
makePolygon(5,4, 5,3, 6,3, 5,3, 5,4);
You can also have a roi with non-zero area when looking at high
magnification. This can happen if a pixel is cut in half, but the pixel
is not inside the Roi by the ImageJ convention of how border pixels are
handled. Then ImageJ considers it a roi with no pixels enclosed, see
this macro example:
makePolygon(8,6, 9,5, 9,6);
For Rois with zero pixels, a "Measure" command reports an area of 0 and
NaN values for Mean, Min, Max, IntDen, etc.
Michael
________________________________________________________________
On 27.03.19 14:53, lougerard wrote:
> Hi all ! I'm new here, so i'l try to be as clear as possible !
>
> I'm using the FreehandRoi object in my project. I would like to know if it's
> possible to measure the integrated density of a FreehandRoi object ?
>
> I'm doing it well for rectangular Roi's (with the good results) via :
> [...]
> Analyzer.setMeasurements(Measurements.INTEGRATED_DENSITY);
> resultTable = myRoiManager.multiMeasure(MyImagePlus);
> resultTable.show("test");
> [...]
>
> When i tried to do this with an FreehandRoi object, it doesn't work. I only
> receive NaN for both (Inden and RawInDen) for all my frames in my ImagePlus.
>
> Does someone have an idea ?
>
> PS : the debug properties of myFreehand object is :
> bounds : null
> x,y,w,h: 57,32,1,1
> as example, but i don't know if the two 1 and the null bounds are
> normal
>
>
> Thanks in advance !! Have a nice day.
>
> Best regards,
>
> Louis
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
