<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7654.12">
<TITLE>RE: [Insight-users] Problem implementing GradientAnisotropicDiffusionImageFilter</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Hello Bill and Luis,<BR>
With conductance = 6 the ITK data output do compare well with MATLAB. Thanks for your help. <BR>
<BR>
Regards,<BR>
Mike<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: Bill Lorensen [<A HREF="mailto:bill.lorensen@gmail.com">mailto:bill.lorensen@gmail.com</A>]<BR>
Sent: Wed 12/2/2009 12:03 AM<BR>
To: O'Connor, Michael<BR>
Cc: Luis Ibanez; insight-users@itk.org<BR>
Subject: Re: [Insight-users] Problem implementing GradientAnisotropicDiffusionImageFilter<BR>
<BR>
The average gradient in your data is about .0005. So the conductance<BR>
of the itk code should be about the matlab conductance / .0005 which<BR>
is about .003 / .0005 = 6.<BR>
<BR>
Try a conductance of 6 without scaling your data and see how it compares.<BR>
<BR>
Bill<BR>
<BR>
<BR>
On Tue, Dec 1, 2009 at 11:36 PM, Bill Lorensen <bill.lorensen@gmail.com> wrote:<BR>
> In itk, the K divisor is average_gradient_squared *<BR>
> conductance_squared. I suspect that accounts for some of the<BR>
> difference.<BR>
><BR>
> On Tue, Dec 1, 2009 at 9:01 PM, O'Connor, Michael<BR>
> <Michael.OConnor@umassmed.edu> wrote:<BR>
>> Hello Luis,<BR>
>> The pixel spacing is .2 mm. I did not SetSpacing in ITK code.<BR>
>> Thanks for the insight. I will check that out.<BR>
>><BR>
>> Regards,<BR>
>> Michael<BR>
>><BR>
>><BR>
>> -----Original Message-----<BR>
>> From: Luis Ibanez [<A HREF="mailto:luis.ibanez@kitware.com">mailto:luis.ibanez@kitware.com</A>]<BR>
>> Sent: Tue 12/1/2009 7:56 PM<BR>
>> To: O'Connor, Michael<BR>
>> Cc: Bill Lorensen; insight-users@itk.org<BR>
>> Subject: Re: [Insight-users] Problem implementing<BR>
>> GradientAnisotropicDiffusionImageFilter<BR>
>><BR>
>> Hi Michael,<BR>
>><BR>
>> What is the pixel spacing of your image ?<BR>
>><BR>
>> I believe that the ITK implementation may<BR>
>> be taking the pixel spacing into account<BR>
>> when using the value of conductance.<BR>
>><BR>
>><BR>
>> Please let us know,<BR>
>><BR>
>><BR>
>> Thanks<BR>
>><BR>
>><BR>
>> Luis<BR>
>><BR>
>><BR>
>> ------------------------------------------------------------------<BR>
>> On Tue, Dec 1, 2009 at 5:17 PM, O'Connor, Michael<BR>
>> <Michael.OConnor@umassmed.edu> wrote:<BR>
>>> Hi Bill,<BR>
>>> I use the exact same file with ITK (the one attached to the previous<BR>
>>> email with MATLAB code).<BR>
>>><BR>
>>> Regards,<BR>
>>> Michael<BR>
>>><BR>
>>><BR>
>>> -----Original Message-----<BR>
>>> From: Bill Lorensen [<A HREF="mailto:bill.lorensen@gmail.com">mailto:bill.lorensen@gmail.com</A>]<BR>
>>> Sent: Tue 12/1/2009 4:17 PM<BR>
>>> To: O'Connor, Michael<BR>
>>> Cc: insight-users@itk.org<BR>
>>> Subject: Re: [Insight-users] Problem implementing<BR>
>>> GradientAnisotropicDiffusionImageFilter<BR>
>>><BR>
>>> Can you provide a png file or meta file, or whatever file you tried with<BR>
>>> itk?<BR>
>>><BR>
>>> On Tue, Dec 1, 2009 at 4:08 PM, O'Connor, Michael<BR>
>>> <Michael.OConnor@umassmed.edu> wrote:<BR>
>>>> Hi Bill,<BR>
>>>> Thanks. Attached is simple MATLAB test code + the MATLAB filter<BR>
>>>> function used (Dr. Guy Gilboa's code). I've also take the liberty of<BR>
>>>> attaching a test image (specifically opened in MATLAB code). The test<BR>
>>>> image is 400x400 float with background = 0.024 + additive Gaussian Noise<BR>
>>>> and<BR>
>>>> foreground (circle) = 0.031 + additive Gaussian Noise. As said<BR>
>>>> previously<BR>
>>>> MATLAB code seems to work as expected as shown in plot. In ITK there is<BR>
>>>> no<BR>
>>>> similar intraregion filtering (i.e. filtered image ~ input )<BR>
>>>><BR>
>>>> Regards,<BR>
>>>> Michael<BR>
>>>><BR>
>><BR>
>><BR>
>><BR>
><BR>
<BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>