Hi Gabriel,
yes, I can confirm it, but my system is also linux (Java 1.8).
In the meanwhile I wonder whether it wouldn't be easier to have the same
strategy for RGB images as for 16 and 32 bits, i.e., a separate copy of
the image data for display (there named 'pixels8'). These days, the
extra memory required should not be problematic any more, and we would
avoid all the trouble with accidental modification of the original image.
If memory is a concern, one could use a separate image only if the B&C
values are different from the default 0-255.
I guess that the main work involved would be finding extra code working
around the idiosyncratic behavior of the RGB images modifying the image
data upon B&C changes.
Happy holidays to all!
Michael
________________________________________________________________
On 22.12.20 18:30, Gabriel Landini wrote:
> Hello,
> In IJ 1.53h14 under Java 14 (linux), when I run the macro and then I adjust
> any of the controls in the B/C applet, the image is "undone" (i.e. the edges
> image reverts to the original).
>
> run("Nile Bend");
> run("Brightness/Contrast...");
> run("Find Edges");
>
> I wonder if this has to do with the undo buffer.
>
> Cheers
>
> Gabriel
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html