Re: Finding "objects" in images

[rivers(at)cars3.uchicago.edu (Mark Rivers)]

In article <F1GFxC.Lus@midway.uchicago.edu>, rivers@cars3.uchicago.edu (Mark Rivers) writes:
>
>I am looking for code to detect "objects" in images.  In this specific case I
>am looking for x-ray diffraction peaks in a CCD camera image, but the problem
>is general.  I have used the IDL Erode and Dilate functions to clean up the
>data, so that I have adjacent pixels with value=1 inside the "objects" and all
>other pixels=0.  I can easily make a list of all of the pixels which are
>inside the objects of interest.  There might be hundreds of such pixels.
>But I want to do is to find the objects themselves which contain these adjacent
>pixels.  There might only be a few such objects.  
>
>I am sure there are lots of people doing this: finding cells or stars or
>whatever in images, and getting shape information about them.

To answer my own post, I found a routine which does exactly what I want at:
http://idlastro.gsfc.nasa.gov/contents.html#C2

It is called FIND.PRO, written by Wayne Landsman.  It was designed to find
stars in telescope images, but it works great for finding diffraction peaks
in diffraction images.

____________________________________________________________
Mark Rivers                             (773) 702-2279 (office)
Argonne National Laboratory             (630) 252-0422 (office)
Building 434A                           (630) 252-0405 (lab)
9700 South Cass Avenue                  (630) 252-1713 (beamline)
Argonne, IL 60439                       (630) 252-0443 (FAX)