Hi Margaret,
You could use thresholding to isolate the leaf area; then use the
Particle Analyser to get the smallest rectangle that encompasses
the leaf; then use knowledge of the width, height, and location of that
rectangle to define the quadrants. This is probably best done with a
macro:
setThreshold(low, high); // or other operations to isolate the leaf
run("Make Binary");
run("Set Measurements...", "area bounding redirect=None decimal=2");
run("Analyze Particles...", "size="+ minsize+"-"+maxsize+" circularity=0.00-1.00 show=Nothing display exclude clear");
// where minsize and maxsize are variables with your limits
// then, assuming only one leaf "particle" was isolate in the image
wid= getResult('Width', 1);
x = floor(getResult('BX', 1));
y = floor(getResult('BY', 1));
// do some calculations to get the location of the quadrants, then:
makeRectangle(x1, y1, qwidth, qheight);
makeRectangle(x2, y2, qwidth, qheight);
makeRectangle(x3, y3, qwidth, qheight);
makeRectangle(x4, y4, qwidth, qheight);
You can calculate the quadrants based on your centroid and knowledge of
the leaf-enclosing rectangle dimensions, or based on the location of the
leaf-enclosing rectangle itself; I do not know which one makes more sense
for your application.
Best wishes,
Mark
Mark Chopping, Ph.D.
Department of Earth & Environmental Studies
Montclair State University, Montclair, NJ 07043
It's All Here!
http://www.montclair.edu/csam/remote-sensing-lab/----
> Date: Sun, 2 Oct 2016 13:33:30 -0700
> From: Margaret Johnson <
[hidden email]>
> Reply-To:
[hidden email]
> To:
[hidden email]
> Subject: Creating 4 quadrant ROIs based on centroid
>
> Hello,
> I am new to the community. As such, I appreciate any advice in not abusing
> this mailing list as well as helping the community. For example, the
> question I have may not be appropriate. I tried to find an answer. Please
> let me know if I am abusing this list.
>
> My challenge is to divide an irregular shape (a leaf) into 4 quadrants,
> with the centroid being the mid-division. Then put the pixels from each
> quadrant into it's own ROI.
>
> I can easily get the centroid. It is not clear to me how to make the 4
> ROIs. Advice appreciated.
>
> Thank you.
>
> --
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
