slice associated ROIs

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

slice associated ROIs

Tony Lien
Hi,

I think there is a problem with slice-associated ROIs when the total number
of ROIs is larger than 9999.  Once the number of ROIs is greater than 9999,
the slice-association for the "show all" button becomes messed up.  When you
hit "show all", it displays all of the ROIs whose number is greater than
9999, regardless of the slice number.  I think this might be because the
slice information is coded in the name of the ROI like so: "0001-2323-0323"
where the first 4-digit number is the slice number, the second is the ROI
number, and the third is the y-coordinate.  When you have more than 9999
ROIs, the ROI names change to 5-digit identifiers, ie "00001-02323-00323".
I guess the ROI manager can't figure out which slice these ROIs are
associated with and so it just displays them in all of the slices
something.  Does anyone know how to fix this?

Thanks,

Tony Lien
Reply | Threaded
Open this post in threaded view
|

Re: slice associated ROIs

Wayne Rasband
> I think there is a problem with slice-associated ROIs when the total
> number
> of ROIs is larger than 9999.  Once the number of ROIs is greater than
> 9999,
> the slice-association for the "show all" button becomes messed up.  
> When
> you hit "show all", it displays all of the ROIs whose number is
> greater than
> 9999, regardless of the slice number.  I think this might be because
> the
> slice information is coded in the name of the ROI like so:
> "0001-2323-0323"
> where the first 4-digit number is the slice number, the second is the
> ROI
> number, and the third is the y-coordinate.  When you have more than
> 9999
> ROIs, the ROI names change to 5-digit identifiers, ie
> "00001-02323-00323".
> I guess the ROI manager can't figure out which slice these ROIs are
> associated with and so it just displays them in all of the slices
> something.  Does anyone know how to fix this?

This bug is fixed in ImageJ 1.41c. The bug that caused incorrect
display of rotated 48-bit TIFFs is also fixed.

-wayne