[ITK-users] itkMeanImageFilter on RGB (how to adjust example MedianFilteringOfAnRGBImage?)

[Dr. Roman Grothausmann]

Many thanks Bill, both hints are very helpful.

On 12/07/14 15:18, Bill Lorensen wrote:
> You are better off using vector pixels versus rgb pixels. No
> overriding necessary. You should be able to modify your example by
> looking at this one:
>
> On Fri, Jul 11, 2014 at 11:26 AM, Bill Lorensen <bill.lorensen at gmail.com> wrote:
>> You need to define a cast operator. SOmething like:
>>    operator TComponent()
>>    {
>>      return this->GetLuminance();
>>    }
>>
>> But I don't think you will get the result you want.
>>
>> Bill
>>
>> On Fri, Jul 11, 2014 at 7:53 AM, Dr. Roman Grothausmann
>> <grothausmann.roman at mh-hannover.de> wrote:
>>> Dear mailing list members,
>>>
>>>
>>> How would I need to extend/modify the example for
>>> MedianFilteringOfAnRGBImage
>>>
>>> http://itk.org/ITKExamples/src/Filtering/Smoothing/MedianFilteringOfAnRGBImage/Documentation.html
>>>
>>> to work with itkMeanImageFilter?
>>>
>>> I modified myRGBPixel like this:
>>>
>>> namespace itk{
>>>    template<typename TComponent>
>>>    class myRGBPixel : public RGBPixel<TComponent>{
>>>    public:
>>>      typedef myRGBPixel     Self;
>>>      typedef RGBPixel<TComponent> Superclass;
>>>
>>>      myRGBPixel():Superclass() {} //avoid candidate expects 1 argument, 0
>>> provided
>>>      myRGBPixel(int i):Superclass(i) {} //avoid candidate expects 0
>>> arguments, 1 provided
>>>
>>>      using RGBPixel<TComponent>::operator=;
>>>
>>>      bool operator<=(const Self & r) const{
>>>        return (this->GetLuminance() <= r.GetLuminance());
>>>      }
>>>      bool operator>=(const Self & r) const{
>>>        return (this->GetLuminance() >= r.GetLuminance());
>>>      }
>>>      bool operator<(const Self & r) const{
>>>        return (this->GetLuminance() < r.GetLuminance());
>>>      }
>>>      bool operator>(const Self & r) const{
>>>        return (this->GetLuminance() > r.GetLuminance());
>>>      }
>>>    };
>>> }
>>>
>>>
>>> However gcc still yields this error:
>>>
>>>
>>> In file included from
>>> /opt/itk-4.5.1/include/ITK-4.5/itkMeanImageFilter.h:115:0,
>>>                   from /net/home/ghky/itk/simple/mean-CPU_02.cxx:8:
>>> /opt/itk-4.5.1/include/ITK-4.5/itkMeanImageFilter.hxx: In instantiation of
>>> ?void itk::MeanImageFilter<TInputImage,
>>> TOutputImage>::ThreadedGenerateData(const OutputImageRegionType&,
>>> itk::ThreadIdType) [with TInputImage = itk::Image<itk::myRGBPixel<unsigned
>>> char>, 3u>; TOutputImage = itk::Image<itk::myRGBPixel<unsigned char>, 3u>;
>>> itk::MeanImageFilter<TInputImage, TOutputImage>::OutputImageRegionType =
>>> itk::ImageRegion<3u>; itk::ThreadIdType = unsigned int]?:
>>> /net/home/ghky/itk/simple/mean-CPU_02.cxx:335:5:   required from here
>>> /opt/itk-4.5.1/include/ITK-4.5/itkMeanImageFilter.hxx:81:9: error: invalid
>>> static_cast from type
>>> ?itk::ConstNeighborhoodIterator<itk::Image<itk::myRGBPixel<unsigned char>,
>>> 3u>,
>>> itk::ZeroFluxNeumannBoundaryCondition<itk::Image<itk::myRGBPixel<unsigned
>>> char>, 3u>, itk::Image<itk::myRGBPixel<unsigned char>, 3u> > >::PixelType
>>> {aka itk::myRGBPixel<unsigned char>}? to type
>>> ?itk::MeanImageFilter<itk::Image<itk::myRGBPixel<unsigned char>, 3u>,
>>> itk::Image<itk::myRGBPixel<unsigned char>, 3u> >::InputRealType {aka
>>> double}?
>>> make[3]: *** [CMakeFiles/mean-CPU_02.dir/mean-CPU_02.cxx.o] Error 1
>>> make[2]: *** [CMakeFiles/mean-CPU_02.dir/all] Error 2
>>> make[1]: *** [CMakeFiles/mean-CPU_02.dir/rule] Error 2
>>> make: *** [mean-CPU_02] Error 2
>>>
>>>
>>> What am I missing in myRGBPixel?
>>>
>>> Any help or hints are very much appreciated
>>> Roman
>>>
>>> --
>>> Dr. Roman Grothausmann
>>>
>>> Tomographie und Digitale Bildverarbeitung
>>> Tomography and Digital Image Analysis
>>>
>>> Institut für Funktionelle und Angewandte Anatomie, OE 4120
>>> Medizinische Hochschule Hannover
>>> Carl-Neuberg-Str. 1
>>> D-30625 Hannover
>>>
>>> Tel. +49 511 532-9574
>>> _____________________________________
>>> 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://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://public.kitware.com/mailman/listinfo/insight-users
>>
>>
>>
>> --
>> Unpaid intern in BillsBasement at noware dot com
>
>
>

-- 
Dr. Roman Grothausmann

Tomographie und Digitale Bildverarbeitung
Tomography and Digital Image Analysis

Institut für Funktionelle und Angewandte Anatomie, OE 4120
Medizinische Hochschule Hannover
Carl-Neuberg-Str. 1
D-30625 Hannover

Tel. +49 511 532-9574