Re: watershed

Posted by Pamela Young on Aug 28, 2014; 1:00am
URL: http://imagej.273.s1.nabble.com/watershed-tp5009231p5009388.html

Thanks Wayne!  

Dr Pamela A. Young
 | Light and Optical Microscopist
Australian Centre for Microscopy & Microanalysis

THE UNIVERSITY OF SYDNEY
Rm 116A, Madsen Building F09 | The University of Sydney | NSW | 2006 |
Australia
T [phone] | F [phone]
E [hidden email] | W http://sydney.edu.au/acmm

Incorporating:
Australian Microscopy & Microanalysis Research Facility (AMMRF) | W
http://www.ammrf.org.au <http://www.ammrf.org.au/>
ARC Centre of Excellence for Design in Light Metals | W
http://www.arclightmetals.org.au <http://www.arclightmetals.org.au/>

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.



On 16/08/2014 2:41 am, "Rasband, Wayne (NIH/NIMH) [E]"
<[hidden email]> wrote:

>On Aug 14, 2014, at 11:07 PM, Pamela Young wrote:
>
>> I was giving a FIJI demo yesterday and noticed that
>>
>> Process>Binary>Watershed
>>
>> separated joined particles. But if I inverted the LUT, it connects the
>>particles.
>>
>> I was playing with the Blobs file and just threw a quick threshold on
>>to make a mask.  So the blobs were 255 and the background was 0.  But I
>>didn¹t like that the blobs were visually black and the background was
>>visually white.  So I clicked Invert LUT.  The blobs were still 255 and
>>the background was still 0.  But with the new LUT watershed connects all
>>the blobs.  Why does the LUT matter?
>
>Commands in the Process>Binary submenu, such as Watershed,  assume that
>images contain black objects on a white background. Check "Black
>background" in the Process>Binary>Options dialog to have them assume that
>objects are white and background is black. Add
>
>   setOption("BlackBackground", true);
>
>to the Edit>Options>Startup dialog to insure that "Black background" is
>always enabled when ImageJ starts up (requires ImageJ 1.49e or later).
>
>-wayne
>--
>ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html