[Insight-users] Fast binary morphology dilation

Miller, James V (Research) millerjv at crd.ge.com
Mon Apr 4 10:42:40 EDT 2005


I just found my test that I was using (a hacked version of 
MathematicalMorphologyBinaryFilters.cxx). This test passes if 
I replace the BinaryDilateImageFilter with a FastIncrementalBinaryDilateImageFilter.

It also didn't crash when I set it up to run on a subregion
(using the streaming mechanism) and produce the correct subregion
result.

I checked in the versions of the FastIncrementalBinaryDilateImageFilter
that I have.  If Jerome sees this and has newer versions, we can supplant
what I just checked in.

The FastIncrementalBinaryDilateImageFilter is much faster than the 
BinaryDilateImageFilter and is more general than the ObjectMorphology
filters.  However, the FastIncrementalBinaryDilateImageFilter does
support multithreading.  But the speed increase is would justify that
we supplant the BinaryDilateImageFilter with this filter.

We should still contruct a version that will do erosion.

Jim



-----Original Message-----
From: insight-users-bounces at itk.org
[mailto:insight-users-bounces at itk.org]On Behalf Of Miller, James V
(Research)
Sent: Monday, April 04, 2005 10:15 AM
To: Peter Cech; Luis Ibanez; Insight-users (E-mail);
jerome.schmid at ircad.u-strasbg.fr
Subject: RE: [Insight-users] Fast binary morphology dilation


I still have this code.  My last modification is dated September 15th.

I lost a mailbox a few weeks ago, so I lost some of the message thread 
I had with  Jerome Schmid who wrote the code.  I believe I had made some 
changes to Jerome's original code to support streaming.

I think Jerome might have made another set of changes to this code
but I am not sure whether the version I have right now is his latest 
version.

I would be motivated to check this in today since I am about to get a 
new computer and I do not trust our computer deployment process will 
keep my copy around.  I think I lost the test/example that I was using
to test this program.

Jim




-----Original Message-----
From: insight-users-bounces at itk.org
[mailto:insight-users-bounces at itk.org]On Behalf Of Peter Cech
Sent: Sunday, April 03, 2005 2:05 PM
To: Luis Ibanez
Subject: Re: [Insight-users] Fast binary morphology dilation


On Sun, Apr 03, 2005 at 11:12:02 -0400, Luis Ibanez wrote:
> Could you please be more specific regarding the email
> message that you are quoting ?
> 
> 
> If you can provide at least the date of that email (adding
> the author will be even better), that will help us to track
> this issue. The mailing-list receives between 250 and 400
> emails a month, this makes difficult to remember every single
> issue, unless they are reported in the bug tracker. Thanks a
> lot for your understanding in this matter.

Sorry, I deleted too much. Here are the cited headers:

-----Message d'origine-----                                                                                             
De : Miller, James V (Research) [mailto:millerjv at crd.ge.com]                                                            
Envoyé : mercredi 15 septembre 2004 22:52                                                                               
À : 'jerome schmid'                                                                                                     
Objet : RE: [Insight-users] Fast binary morphology dilation                                                             


It was sent to list indirectly with this reply:

http://public.kitware.com/pipermail/insight-users/2004-September/010447.html


There was some early code in that thread, after addin some typename
keywords, it compiled and is working. I'm working with rather big 3D
data sets, so it takes 10-30 minutes with current
BinaryDilateImageFilter (having several objects I cannot use
DilateObjectMorphologyImageFilter). However with the fast dilate filter
it took always under one minute. Quite impressive speedup. It would be
pity if it got forgotten.

Regards,

Peter Cech


> [...]


> -------------------
> Peter Cech wrote:
> 
> >>I think I have a little clean up/testing to do and then I will check it 
> >>into
> >>ITK.  I haven't quite decided whether to supplant the current filters or
> >>not.
> >
> >
> >Any update on this? The last version of BinaryDilateImageFilter in CVS
> >is from April 2004 and I was unable to locate any new filter for binary
> >dilation.
> >
> >Regards,
> >Peter Cech
> >_______________________________________________
> >Insight-users mailing list
> >Insight-users at itk.org
> >http://www.itk.org/mailman/listinfo/insight-users
> >
> >
> 
> 
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
_______________________________________________
Insight-users mailing list
Insight-users at itk.org
http://www.itk.org/mailman/listinfo/insight-users
_______________________________________________
Insight-users mailing list
Insight-users at itk.org
http://www.itk.org/mailman/listinfo/insight-users


More information about the Insight-users mailing list