Integrated density along thick line

[Francis Burton]

Hi,

I measure integrated intensity along a freehand line of interest.
It would be useful also to measure this in the region defined by a
thickened line, but setting line width > 1 produces IntDen = 0. It
appears to work for single straight lines, but not freehand. If it
is relatively easy to implement, I would very much appreciate this
to be considered as a future feature.

I also noticed some strange behaviour in the Line Width dialog
(ImageJ 1.43n under Windows). If width is changed, then the dialog
is closed, then reopened, clicking on the scrollbar to decrease the
width doesn't change the value displayed. Also the scrollbar does
not reflect the displayed value - e.g. when it is 150, the scrollbar
is positioned to the left rather than the middle - until the value
is changed by editing.

Francis

[Rasband, Wayne (NIH/NIMH) [E]]

On Jan 14, 2010, at 6:49 AM, Francis Burton wrote:

> Hi,
>
> I measure integrated intensity along a freehand line of
> interest. It would be useful also to measure this in the
> region defined by a thickened line, but setting line width >
> 1 produces IntDen = 0. It appears to work for single
> straight lines, but not freehand. If it is relatively easy
> to implement, I would very much appreciate this to be
> considered as a future feature.

Enable "Mean" in Analyze>Set Measurements and it will work. Integrated density is just the product of area and mean.

> I also noticed some strange behaviour in the Line Width
> dialog (ImageJ 1.43n under Windows). If width is changed,
> then the dialog is closed, then reopened, clicking on the
> scrollbar to decrease the width doesn't change the value
> displayed. Also the scrollbar does not reflect the displayed
> value - e.g. when it is 150, the scrollbar is positioned to
> the left rather than the middle - until the value is changed
> by editing.

This bug is fixed in the 1.43o daily build.

-wayne

[imageJ_user]

Hi,

The default 'Plot profile' for freehand lines with thickness X plots the vertically (or horizontally) averaged pixel value. Is it possible to instead plot the integrated values?

Thanks