Inverse of padarray()?

I am working on MRI image processing, wherein I work with 3D arrays. To increase the resolution while taking the FFT of the array, I use padarray() and double the size of the image. After the FFT is done, I want to make the size of the resultant array, the same as that of the original array. Is there a way to do it other than looping over and deleting the extra voxels?

For example, if I have 64*64*64 matrix and I increase its size to 128*128*128, and even the resultant will have the same size. I want to change the resultant to a size 64*64*64. To keep it short, is there some kind of inverse padarray?

[Adam]

fft will pad for you if you tell it to use more points than you have. It will always pad at the end so that you can just take the initial portion of the result and leave the trailing zeros behind again afterwards.

[Image Analyst]

Simply crop it. By the way, you're not increasing the spectral resolution, though you may incorrectly think so because the FFT has more elements. Think about it. If you don't realize the truth of what I said, ask.

[jupiter]

@Image Analyst, Thanks for the answer. Ok, I am doing * ifftn(fft(x).*fft(y))* . So if I do the zero padding to x and y before taking fft it does not change anything is what you are saying?

[Adam]

http://www.bitweenie.com/listings/fft-zero-padding/