Path: news.mathworks.com!not-for-mail
From: "Valentin Magidson" <valentin@wadsworth.org>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Find gaussian humps in a 3D dataset
Date: Fri, 10 Oct 2008 15:33:01 +0000 (UTC)
Organization: Wadsworth Center
Lines: 14
Message-ID: <gcnsjd$qmj$1@fred.mathworks.com>
References: <galg9a$995$1@fred.mathworks.com> <galiqa$6es$1@fred.mathworks.com>
Reply-To: "Valentin Magidson" <valentin@wadsworth.org>
NNTP-Posting-Host: webapp-02-blr.mathworks.com
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Trace: fred.mathworks.com 1223652781 27347 172.30.248.37 (10 Oct 2008 15:33:01 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Fri, 10 Oct 2008 15:33:01 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 1396617
Xref: news.mathworks.com comp.soft-sys.matlab:494578


Peter Perkins <Peter.PerkinsRemoveThis@mathworks.com> wrote in message <galiqa$6es$1@fred.mathworks.com>...
> Thomas Clark wrote:.....
> > I have a 3D image (scalar intensity field) of particles in a fluid flow. .....
> > Each particle could be approximately represented by a gaussian hump in intensity, surrounding the centre position. .....
> > So the question is, how best to find the positions of the particles? 
> 
> Tom, it's possible that the GMDISTRIBUTION function in the Statistics Toolbox, which fits a Gaussian mixture distribution, might be helpful.  I don't know enough about your application, though.
> 
> - Peter Perkins
>   The MathWorks, Inc.
Hello,
    I have a question similar to one Thomas Clark asked (above): I have 3D scalar (intensity) field and I have to find particle centers by fitting mixed Gaussian distribution. 
    gmdistribution.fit you suggested seems like good idea except for one problem: the input data are &#8220;events&#8221; or &#8220;hits&#8221; in XYZ space, which have no value. To convert each Xi,Yi,Zi voxel with intensity Ni to this form, I should  add Ni identical [Xi Yi Zi ] rows to the input, to result in a huge matrix. Is there a better workaround or maybe other function similar to gmdistribution.fit but taking scalar data field as input?