Arabidopsis auto segmentation

[Andrew Defries]

Dear ImageJ  users,

I have a number of images similar to shown below:

http://biocluster.ucr.edu/~adefries/Test.JPG

I essentially want ROIs similar to the outline of the contrasted seedlings:

http://biocluster.ucr.edu/~adefries/Test2.jpg

I have had good success using auto local threshold and edge detection, but
encounter problems when seedlings are twisted together. I would like to
develop a method to outline each plant, if possible (with or without seed),
and then outline the boundaries of each plant such as cotyledon (green
part), stem (hypocotyl), and root.

At present I have been collecting only two variables derived from polyline
rois hand clicked through the hypocotyl and root, so any advice to improve
my throughput would be greatly appreciated.


Thank you,

Andrew


--
------------------------------------------------
Andrew Defries
Lab of Michael Pirrung
University of California Riverside
Botany and Plant Sciences
2150 Batchelor Hall
Riverside, CA 92521

(Phone)        951-827-6991

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

[Ignacio Arganda-Carreras]

Hello Andrew,

You can try using Advanced Weka Segmentation to get better segmentation:
http://fiji.sc/wiki/index.php/Advanced_Weka_Segmentation

Let me know if you need help with it!

ignacio

On Sun, Jan 13, 2013 at 4:12 AM, Andrew Defries <[hidden email]>wrote:

> Dear ImageJ  users,
>
> I have a number of images similar to shown below:
>
> http://biocluster.ucr.edu/~adefries/Test.JPG
>
> I essentially want ROIs similar to the outline of the contrasted seedlings:
>
> http://biocluster.ucr.edu/~adefries/Test2.jpg
>
> I have had good success using auto local threshold and edge detection, but
> encounter problems when seedlings are twisted together. I would like to
> develop a method to outline each plant, if possible (with or without seed),
> and then outline the boundaries of each plant such as cotyledon (green
> part), stem (hypocotyl), and root.
>
> At present I have been collecting only two variables derived from polyline
> rois hand clicked through the hypocotyl and root, so any advice to improve
> my throughput would be greatly appreciated.
>
>
> Thank you,
>
> Andrew
>
>
> --
> ------------------------------------------------
> Andrew Defries
> Lab of Michael Pirrung
> University of California Riverside
> Botany and Plant Sciences
> 2150 Batchelor Hall
> Riverside, CA 92521
>
> (Phone)        951-827-6991
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

... [show rest of quote]

--
Ignacio Arganda-Carreras, Ph.D.
Seung's lab, 46-5065
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
43 Vassar St.
Cambridge, MA 02139
USA

Phone: [phone]
Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html

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