[Insight-developers] how to provide A2D2 denoising code?

[Bill Lorensen]

For the time being, I suggest placing it in Filtering/Denoising. It
has a lot of dependencies for such a small module. As Xiaoxiao
mentioned, you will get more feedback once it is in gerrit.

On Thu, Sep 22, 2011 at 5:44 PM, Xiaoxiao Liu <xiaoxiao.liu at kitware.com> wrote:
> Hi Kris,
> Please see my comments below.
> Once you put your patch on gerrit, I am sure you will  receive more detailed
> and  insightful suggestions from everyone.
> Thanks.
>
> On Thu, Sep 22, 2011 at 5:14 PM, Kris Zygmunt <krismz at sci.utah.edu> wrote:
>>
>> Hello!
>>    I am (finally!) ready to provide the A2D2 denoising code and just
>> wanted to discuss the best way to do so.  I will create a gerrit topic
>> called a2d2-denoising and put my changes there.  I think it makes the most
>> sense to create a new Denoising module in Filtering, but please let me know
>> if I should put the code elsewhere.  I have the following files:
>>
>> denoising filter hierarchy that will be dependent on
>> ITKStatistics, ITKCommon, vnl, ITKImageStatistics, ITKImageGrid,
>> ITKTransform, ITKImageFunction, ITKImageAdaptors
>> This hierarchy should be in Filtering somewhere, but does it go into a new
>> module (called Denoising?) or an existing one?
>> itkPatchBasedDenoisingBaseImageFilter.{h,hxx}
>> itkPatchBasedDenoisingImageFilter.{h,hxx}
>> <associated test code will go in the associated test directory>
>>
> How many new classes you are going to introduce into ITK? Are there
> potential growth or further development if we make it
> as a module called ITKDenoising?
>
>>
>> Image to sample adaptor to be put in the Modules/Numerics/Statistics area
>> itkImageToNeighborhoodSampleAdaptor.{h,hxx}
>>
>>
>> Generic subsampler hierarchy that I think fits in the
>> Modules/Numerics/Statistics area
>> itkSubsamplerBase.{h,hxx}
>> itkRegionConstrainedSubsampler.{h,hxx}
>> itkSpatialNeighborSubsampler.{h,hxx}
>> itkUniformRandomSpatialNeighborSubsampler.{h,hxx}
>> itkGaussianRandomSpatialNeighborSubsampler.{h,hxx}
>>
>> It seems like Modules/Numerics/Statistics is rather large, are there any
>> plans to split this into smaller modules?  What is the recommended way to
>> include vnl in CMake for ITKv4?
>>
>>>  There are 64 classes right now in ITKStatistics. A couple of more
>>> classes won't hurt.
> If you have suggestions about how we should divide it up to meaningful
> smaller modules, please advice.
> We have a ITKVNL module distributed inside ITK as a third party library
> (have you tried to use this module yet?).
> Uses could choose to use their own system installed version of  VXL by
> turning on ITK_USE_SYSTEM_VXL.
>
>>
>> Thanks!
>> Kris
>> _______________________________________________
>> Powered by 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://kitware.com/products/protraining.html
>>
>> 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-developers
>
>
>
> --
>
>
> ---------------------------------------------
> Xiaoxiao Liu, Ph.D.
> R & D Engineer
> Kitware Inc.
> Clifton Park, NY
> Phone: (518) 881-4924  or  (518) 371-3971 x124
>
>
> _______________________________________________
> Powered by 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://kitware.com/products/protraining.html
>
> 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-developers
>
>