N-Dimensional ROIs

[Michael Doube]

As a suggestion for the ImageJDev team:

Recently I've been frustrated by ROI's being limited to 2D. With the
emerging utility of the 3D viewer and the proposal that ImageJ 2.0
handles N-dimensional data, it makes sense that ROIs should keep up
with this development.

In other words, in an N-dimensional image, one should be able to
specify and visualise an N-dimensional ROI. So you can have a 3D VOI,
and a 4D VOI with time limits (or even changing shape over time), or
limit the ROI to a channel (5D).

Cheers,

Michael

[Curtis Rueden]

Hi Michael,

Thanks, we are very interested in multidimensional ROIs as well. They were recently discussed on the bioimagexd-users mailing list:

https://sourceforge.net/mailarchive/forum.php?thread_name=c077d2021002030859i4a62529at312e35e90d5c63fc%40mail.gmail.com&forum_name=bioimagexd-users

The simplest thing for ImageJ would be to add the ability to group (2D) ROIs together to form larger structures, though there is a question of mask-based versus vector-based regions, as well as storing 3D surfaces and similar computed structures.

-Curtis

--
You received this message because you are subscribed to the Google Groups "ImageJX" group.
To post to this group, send email to ima...@googlegroups.com.
To unsubscribe from this group, send email to imagejx+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/imagejx?hl=en.

[Curtis Rueden]

Hi Michael,

Thanks, we are very interested in multidimensional ROIs as well. They were recently discussed on the bioimagexd-users mailing list:

https://sourceforge.net/mailarchive/forum.php?thread_name=c077d2021002030859i4a62529at312e35e90d5c63fc%40mail.gmail.com&forum_name=bioimagexd-users

The simplest thing for ImageJ would be to add the ability to group (2D) ROIs together to form larger structures, though there is a question of mask-based versus vector-based regions, as well as storing 3D surfaces and similar computed structures.

-Curtis

On Fri, Feb 26, 2010 at 6:02 AM, Michael Doube <michae...@gmail.com> wrote: