[Insight-users] Is FFT in ITK a complex FFT

Martin Kavec kavec at messi.uku.fi
Thu Jun 9 11:44:36 EDT 2005


Hi Hans,

I am actually more interested in iFFT, which is suppose to take a complex
image pair, or in ITK terminology, a std::complex< Type > PixelType
images. Using these I need to reconstruct raw MRI data and get absolute
images (reflecting proton density) and - very importantly - phase images.

What you write is a bit disappointing. :)
I am going to file a bug/feature request.

Thanks for reply,

Martin

On Thu, 9 Jun 2005, Hans Johnson wrote:

> Martin,
>
> I believe that the current FFT implementations will only work with
> taking forward FFT of float and double data types (i.e. not complex,
> integer, RGB, vector, or any other type of data). The current
> implementation also takes advantage of the complex conjugate symmetry
> when working with real data.
>
> Hans
>
>
>
>
> On Jun 09, 2005 07:41 AM, Martin Kavec <kavec at messi.uku.fi> wrote:
>
> > Hi,
> >
> > following code does not compile:
> >
> >   typedef std::complex<float> ComplexPixelType;
> >   const   unsigned int        Dimension = 3;
> >
> >   typedef itk::Image< ComplexPixelType, Dimension > ComplexImageType;
> >
> >   typedef itk::ImageFileReader< ComplexImageType >  ReaderType;
> >   ReaderType::Pointer reader = ReaderType::New();
> >   reader->SetFileName( "foo.hdr" );
> >
> >
> > typedef itk::VnlFFTComplexConjugateToRealImageFilter< ComplexPixelType
> >   ,
> > Dimension >  FFTFilterType;
> >   FFTFilterType::Pointer fftFilter = FFTFilterType::New();
> >
> >   fftFilter->SetInput( reader->GetOutput() );
> >
> > So I am in doubts, if Fourier transform in ITK is actually a complex
> > FFT?
> >
> > Thanks for explanation.
> >
> > Martin
> > _______________________________________________
> > Insight-users mailing list
> > Insight-users at itk.org
> > http://www.itk.org/mailman/listinfo/insight-users
> >
>
>
>
> Hans Johnson
> hans-johnson at uiowa.edu
>


More information about the Insight-users mailing list