Re: Connected component analysis

["Jason P. Meyers" <jpm7934(at)cis.rit.edu>]

Dave,

	This sounds like a 3-D extension of a problem we discussed in our
graduate DIP class.  One of our homework problems involved discussing an
algorithm to do a 2-D analogy of what you are asking.  Our professor,
Dr. Rhody, has a rather extensive website which includes most of our
class notes on the subject.  His library of routines also has a program
called HR_LABEL_REGION which is an improved version of IDL's
LABEL_REGION routine which only works on binary images.

You can check out the website at:  http://www.cis.rit.edu/class/simg782/

The textbook that we use in the class is Digital Image Processing, by
Gonzalez and Wood, ISBN 0-201-50803-6.  I don't know how useful it is
for your problem because we have been relying mostly on the professor's
class notes and only use the text as a reference from time to time.

Please let me know if you think I might be able to help you further.

Jason Meyers
Ph.D. Student, Center for Imaging Science
Rochester Institute of Technology
jpm7934@rit.edu

Dave Brennan wrote:
> 
> Hi,
> 
> I am currently trying to write a program in IDL to analyze brain MRI
> images. This involves using various morphological operations. I was
> wondering if anyone has any experience using connected component
> analysis in IDL.
> I am trying to replicate results that I can produce using Analyze AVW's
> connect operator (this connects 3D binary structures and gives them a
> grey scale value relating to their size).
> 
> If anyone has any code or knows of a good way to go about this (or any
> good references) please could you let me know.
> 
> Thanks for your help
> 
> Dave Brennan