[ImageJ-devel] last release of HistThresh

Gabriel Landini G.Landini at bham.ac.uk
Wed Mar 19 11:02:44 CDT 2014

On Tuesday 18 Mar 2014 18:31:58 Barry DeZonia wrote:
> I am no longer working with the lab that worked with Antti. I have looked
> at the backups I made when I left and I do not have a copy of Antti's work.
> Some people in my old lab might have it and they check this mailing list.
> However I have also CCed Gabriel Landini (affiliated with Fiji) as he
> worked with me on this a bit and may have made a copy of the code.
> On Tue, Mar 18, 2014 at 5:04 PM, Carnë Draug <carandraug at octave.org> wrote:
> > Hi Barry
> > 
> > early last year you have contacted Antti Niemistö about his HistThresh
> > toolbox [1]. This was a Matlab toolbox for image thresholding which he
> > released under GPLv2+. In the mailing list, he agreed to release it
> > under BSD for ImageJ2 development.
> > 
> > Unfortunately, it appears that he no longer works at the Tampere
> > University of Technology and the website for the project has been
> > taken down. His previous email address has also been cancelled and I
> > couldn't find his new contact. I tried to use wayback machine to
> > salvage his project, but the last backup [2] was before the last
> > release (he made one release prompted by your email which fixed a
> > couple of bugs in the minimum and maximum likelihood method). I have
> > made a github repository with the version I found [3]
> > 
> > I made an Octave port which works fine for me [4] but I know some
> > Matlab users who would like to use it. Would you happen to have made a
> > backup of the last release he made?
> > 
> > Thanks in advance,
> > Carnë
> > 
> > [1] http://imagej.net/pipermail/imagej-devel/2013-February/001404.html
> > [2]
> > http://web.archive.org/web/20120628113512/http://www.cs.tut.fi/~ant/histth
> > resh/ [3] https://github.com/carandraug/histthresh
> > [4]
> > https://sourceforge.net/p/octave/image/ci/default/tree/inst/graythresh.m

Hi Barry, Carnë,
This is the only version I have of the maximum likelihood estimator 
(th_maxlik.m) code from Anti's website. The date of the file in my computer is 
His method uses the th_minimum.m procedure as a first approximation, which in 
turn uses the bimodtest.m and others (also attached), but in my endeavours of 
porting the th_maxlik method in Java for IJ I failed miserably in reproducing 
the results that the code gave for matlab (shown in the pdf attached).
I spent some time with this particular method trying to resolve what was the 
issue, but I never succeeded, so rather than having a broken method, I left it 
our from the autothreshold plugin.

I hope it is useful.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: th_maxlik.m
Type: text/x-modelica
Size: 2341 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0006.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: th_minimum.m
Type: text/x-modelica
Size: 1226 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0007.bin>
-------------- next part --------------
HistThresh toolbox for MATLAB
Copyright (C) 2004 Antti Niemist?his program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA


The HistThresh toolbox contains selected histogram-based thresholding
algorithms for MATLAB. The toolbox has been tested with MATLAB 6.5
(R13), but it may work with some older versions of MATLAB as well.


v1.02 (2 Nov 2004)
- minor update: fixed the perpetual loop bug in th_minimum.m and


Extract the contents of the zip file "HistThresh.zip" into a directory
from which MATLAB can find the m-files.


The thresholding algorithms are named with the prefix "th_". For
example, "th_minimum.m" is the function for the algorithm by Prewitt
and Mendelsohn that chooses the threshold to be in the valley of a
bimodal histogram. The threshold for an 8-bit image can be found by
typing "T = th_minimum(I)". Further help can be obtained by typing
"help <m-file>" in MATLAB, where "<m-file>" is replaced by the name of
an m-file.


Antti Niemist?stitute of Signal Processing
Tampere University of Technology
P.O. Box 553

E-mail: antti.niemisto at tut.fi
Tel: +358-3-3115 4507
Fax: +358-3-3115 4989



-------------- next part --------------
A non-text attachment was scrubbed...
Name: ThreshComp.pdf
Type: application/pdf
Size: 1454978 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0001.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bimodtest.m
Type: text/x-modelica
Size: 618 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0008.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: A.m
Type: text/x-modelica
Size: 408 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0009.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: B.m
Type: text/x-modelica
Size: 419 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0010.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: C.m
Type: text/x-modelica
Size: 422 bytes
Desc: not available
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20140319/3e7a4b3b/attachment-0011.bin>

More information about the ImageJ-devel mailing list