[Insight-users] Mattes Mutual Information implementation

Tim Allman dr.tim.allman at gmail.com
Wed Mar 13 11:45:49 EDT 2013


Ok, then I'll edit the comment again to reflect this.

Tim

On 13-03-13 11:43 AM, brian avants wrote:
> all of the v4 metrics are thread safe.
>
> the difference is that the jhmi is based on the unser paper, not the 
> mattes paper and is less well validated than the mattes metric which 
> has been around for a long time (in itk)
>
>
> brian
>
>
>
>
> On Wed, Mar 13, 2013 at 11:39 AM, Tim Allman <dr.tim.allman at gmail.com 
> <mailto:dr.tim.allman at gmail.com>> wrote:
>
>     Now I'm confused. What is the difference between the two?
>
>     The comment at the top of
>     itkJointHistogramMutualInformationImageToImageMetricv4.h indicates
>     that this is the Mattes et al method. However, the reference
>     predates the Mattes paper by three years. (By the way the class
>     name is wrong in the comment so it doesn't show up in the doxygen
>     document. I have edited the comment.)
>
>     /** \class JointHistogramMutualInformationImageToImageMetric
>      * \brief Computes the mutual information between two images to be
>      * registered using the method of Mattes et al.
>      *
>      * References:
>      * [1] "Optimization of Mutual Information for MultiResolution Image
>      *      Registration"
>      *      P. Thevenaz and M. Unser
>      *      IEEE Transactions in Image Processing, 9(12) December 2000.
>      *
>      * \ingroup ITKMetricsv4
>      */
>
>     So, does this class implement the referenced paper?
>
>     Is the v4 Mattes class thread safe?
>
>
>     Thanks,
>
>     Tim
>
>
>
>     On 13-03-13 10:10 AM, brian avants wrote:
>>     http://www.itk.org/Doxygen43/html/search.php?query=mattes
>>
>>     gives you
>>
>>     http://www.itk.org/Doxygen43/html/classitk_1_1MattesMutualInformationImageToImageMetricv4.html
>>
>>
>>     brian
>>
>>
>>
>>
>>     On Wed, Mar 13, 2013 at 9:37 AM, Tim Allman
>>     <dr.tim.allman at gmail.com <mailto:dr.tim.allman at gmail.com>> wrote:
>>
>>         My apologies for not reading the documentation. :-[
>>
>>         However, the v4 version,
>>         JointHistogramMutualInformationImageToImageMetricv4, truly is
>>         undocumented. What is its thread safety?
>>
>>         Thanks,
>>         Tim
>>
>>
>>         On 13-03-12 05:33 PM, Christopher Mullins wrote:
>>>         /"This class in not thread safe due the private data
>>>         structures used to the store the sampled points and the
>>>         marginal and joint pdfs." [1]/
>>>         /
>>>         /
>>>         /[1]
>>>         /http://www.itk.org/Doxygen314/html/classitk_1_1MattesMutualInformationImageToImageMetric.html
>>>
>>>
>>>         On Tue, Mar 12, 2013 at 4:50 PM, Rick Frank
>>>         <rfrank at dominionsw.com <mailto:rfrank at dominionsw.com>> wrote:
>>>
>>>             What is the reason it is not thread safe?
>>>
>>>             Rick Frank
>>>
>>>             On Mar 12, 2013, at 2:10 PM, "Tim Allman"
>>>             <dr.tim.allman at gmail.com
>>>             <mailto:dr.tim.allman at gmail.com>> wrote:
>>>
>>>>             I have been living with mutual information day and
>>>>             night for  a while. You can get more information about
>>>>             Normalised MI in D. Rueckert et al. IEEE Trans. Med.
>>>>             Imag., 18 712 (1999) and Mattes MI in Mattes et al IEEE
>>>>             Trans. Med. Imag., 22 120-128 (2003) and references
>>>>             contained therein.
>>>>
>>>>             If you understand the methods in the papers, the ITK
>>>>             classes start to make sense in spite of the poor
>>>>             documentation. It is undocumented but true that Mattes
>>>>             MI is not thread safe so you must set the number of
>>>>             threads it uses to 1. I don't know about the other.
>>>>
>>>>             Tim
>>>>
>>>>             On 13-03-12 01:26 PM, Pedro Morais wrote:
>>>>>             Good afternoon,
>>>>>
>>>>>             I'm working with Mattes Mutual Information to register
>>>>>             a MRI sequence.
>>>>>             Someone can send me some information about Mattes
>>>>>             Mutual Information implementation, there are a lot of
>>>>>             parameters but I think that the documentation is not
>>>>>             enough.
>>>>>             I don't understand what is the difference between the
>>>>>             Normalized Mutual Information and Mattes Mutual
>>>>>             Information. I need to decide between the two methods,
>>>>>             but I need to know the advantagens of each technique.
>>>>>
>>>>>             Thanks for the help.
>>>>>             -- 
>>>>>             Kind Regards,
>>>>>
>>>>>             Pedro Morais
>>>>>
>>>>>
>>>>>             _____________________________________
>>>>>             Powered bywww.kitware.com  <http://www.kitware.com>
>>>>>
>>>>>             Visit other Kitware open-source projects at
>>>>>             http://www.kitware.com/opensource/opensource.html
>>>>>
>>>>>             Kitware offers ITK Training Courses, for more information visit:
>>>>>             http://www.kitware.com/products/protraining.php
>>>>>
>>>>>             Please keep messages on-topic and check the ITK FAQ at:
>>>>>             http://www.itk.org/Wiki/ITK_FAQ
>>>>>
>>>>>             Follow this link to subscribe/unsubscribe:
>>>>>             http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>>             -- 
>>>>             Tim Allman, Ph.D.
>>>>             35 Margaret Street,
>>>>             Guelph Ontario N1E 5R6
>>>>             Canada
>>>>             <dr_tim_allman.vcf>
>>>>             _____________________________________
>>>>             Powered by www.kitware.com <http://www.kitware.com>
>>>>
>>>>             Visit other Kitware open-source projects at
>>>>             http://www.kitware.com/opensource/opensource.html
>>>>
>>>>             Kitware offers ITK Training Courses, for more
>>>>             information visit:
>>>>             http://www.kitware.com/products/protraining.php
>>>>
>>>>             Please keep messages on-topic and check the ITK FAQ at:
>>>>             http://www.itk.org/Wiki/ITK_FAQ
>>>>
>>>>             Follow this link to subscribe/unsubscribe:
>>>>             http://www.itk.org/mailman/listinfo/insight-users
>>>
>>>             _____________________________________
>>>             Powered by www.kitware.com <http://www.kitware.com>
>>>
>>>             Visit other Kitware open-source projects at
>>>             http://www.kitware.com/opensource/opensource.html
>>>
>>>             Kitware offers ITK Training Courses, for more
>>>             information visit:
>>>             http://www.kitware.com/products/protraining.php
>>>
>>>             Please keep messages on-topic and check the ITK FAQ at:
>>>             http://www.itk.org/Wiki/ITK_FAQ
>>>
>>>             Follow this link to subscribe/unsubscribe:
>>>             http://www.itk.org/mailman/listinfo/insight-users
>>>
>>>
>>>
>>>
>>>         -- 
>>>         Christopher Mullins
>>>         R&D Engineer
>>>         Kitware Inc.,
>>>         919.869.8871 <tel:919.869.8871>
>>
>>         -- 
>>         Tim Allman, Ph.D.
>>         35 Margaret Street,
>>         Guelph Ontario N1E 5R6
>>         Canada
>>
>>
>>         _____________________________________
>>         Powered by www.kitware.com <http://www.kitware.com>
>>
>>         Visit other Kitware open-source projects at
>>         http://www.kitware.com/opensource/opensource.html
>>
>>         Kitware offers ITK Training Courses, for more information visit:
>>         http://www.kitware.com/products/protraining.php
>>
>>         Please keep messages on-topic and check the ITK FAQ at:
>>         http://www.itk.org/Wiki/ITK_FAQ
>>
>>         Follow this link to subscribe/unsubscribe:
>>         http://www.itk.org/mailman/listinfo/insight-users
>>
>>
>
>     -- 
>     Tim Allman, Ph.D.
>     35 Margaret Street,
>     Guelph Ontario N1E 5R6
>     Canada
>
>

-- 
Tim Allman, Ph.D.
35 Margaret Street,
Guelph Ontario N1E 5R6
Canada

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20130313/1804c5f5/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dr_tim_allman.vcf
Type: text/x-vcard
Size: 159 bytes
Desc: not available
URL: <http://www.itk.org/pipermail/insight-users/attachments/20130313/1804c5f5/attachment.vcf>


More information about the Insight-users mailing list