What do "fftshift" do exactly for an image ?

What do "fftshift" do exactly for an image ?

Answer 1

[Image Analyst]

Normally the "center" or the 0 frequency location of the spectrum is at the first and last element, and the "ends" of the spectrum (highest frequencies) are at the center of the array. That's just how the fft works. But that's confusing for display because the low frequencies are appearing in two disconnected locations. So fftshift() moves the zero frequency location to the middle of the array so that when you display it with imshow() or plot(), it looks like you'd expect.

[bh dhouha]

can you explain it more with an image please?

[Image Analyst]

See my attached demos.

[Youssef Khmou]

addition : the first element is the sum of signal components.

[bh dhouha]

Image analyst you said: "Normally the "center" or the 0 frequency location of the spectrum is at the first and last element, and the "ends" of the spectrum (highest frequencies) are at the center of the array. That's just how the fft works", i would like to understand the location of each frequency in the spectrum, is there a document on this or something that explains this point?

[Image Analyst]

Try these links:

http://cns-alumni.bu.edu/~slehar/fourier/fourier.html

http://www.mathworks.com/matlabcentral/fileexchange/28810-fourier-transform-demonstration/content/fourier_demo/html/fourier_demo.html

http://blogs.mathworks.com/steve/category/fourier-transforms/

http://www.cv.nrao.edu/course/astr534/FourierTransforms.html