How should I interpret the output of numpy.fft.rff

问题:

Obviously the rfft2 function simply computes the discrete fft of the input matrix. However how do I interpret a given index of the output? Given an index of the output, which Fourier coefficient am I looking at?
I am especially confused by the sizes of the output. For an n by n matrix, the output seems to be an n by (n/2)+1 matrix (for even n). Why does a square matrix ends up with a non-square fourier transform?

回答1:

The output of numpy.fft.rfft2 is simply the left half (plus one column) of a standard two-dimensional FFT, as computed by numpy.fft.fft2. There's no need for rfft2 to supply the right half of the result, because the FFT of a real array has a natural and simple symmetry, and the right half of the full FFT can therefore be derived from the left half using that symmetry.

Here's an example, to illustrate. First, to make it easy to reproduce and easy to see, I'll set NumPy's random state and printing options:

In [1]: import numpy as np

In [2]: np.set_printoptions(precision=3, suppress=True, linewidth=128)

In [3]: random = np.random.RandomState(seed=15206)

Let's create a real input array, with 6 rows and 6 columns:

In [4]: x = random.randn(6, 6)

In [5]: x
Out[5]: 
array([[ 1.577,  0.426,  0.322, -0.891, -0.793,  0.017],
       [ 0.238,  0.603, -0.094, -0.087, -0.936, -1.139],
       [-0.583,  0.394,  0.323, -1.384,  1.255,  0.457],
       [-0.186,  0.687, -0.815, -0.54 ,  0.762, -0.674],
       [-1.604, -0.557,  1.933, -1.122, -0.516, -1.51 ],
       [-1.683, -0.006, -1.648, -0.016,  1.145,  0.809]])

Now take a look at the full FFT (using fft2, not rfft2):

In [6]: fft2_result = np.fft.fft2(x)

In [7]: fft2_result
Out[7]: 
array([[ -5.834+0.j   ,   1.084-2.33j ,  -6.504-3.884j,   3.228-0.j   ,  -6.504+3.884j,   1.084+2.33j ],
       [  1.475-3.311j,   1.865-3.699j,   2.777-0.095j,  -2.570-1.152j,   4.705-3.373j,   4.555-3.657j],
       [  2.758+3.339j,  -3.512+0.398j,   5.824-4.045j,   1.149-3.705j,   0.661-2.127j,  12.368+1.464j],
       [  1.326-0.j   ,   1.191-4.479j,  -3.263+6.19j ,   8.939-0.j   ,  -3.263-6.19j ,   1.191+4.479j],
       [  2.758-3.339j,  12.368-1.464j,   0.661+2.127j,   1.149+3.705j,   5.824+4.045j,  -3.512-0.398j],
       [  1.475+3.311j,   4.555+3.657j,   4.705+3.373j,  -2.570+1.152j,   2.777+0.095j,   1.865+3.699j]])

Notice that there's a symmetry here: for any indices i and j with 0 <= i < 6 and 0 <= j < 6, fft2_result[i, j] is the complex conjugate of fft_result[-i, -j]. For example:

In [8]: fft2_result[2, 4]
Out[8]: (0.66075993512998199-2.127249005984857j)

In [9]: fft2_result[-2, -4].conj()
Out[9]: (0.66075993512998199-2.127249005984857j)

That means that we don't need to include the right-hand half of the output, since it can be derived from the left half. We can save memory and perhaps also a tiny bit of time by only computing the left half of the full FFT. And that's exactly what rfft2 does:

In [10]: rfft2_result = np.fft.rfft2(x)

In [11]: rfft2_result
Out[11]: 
array([[ -5.834+0.j   ,   1.084-2.33j ,  -6.504-3.884j,   3.228+0.j   ],
       [  1.475-3.311j,   1.865-3.699j,   2.777-0.095j,  -2.570-1.152j],
       [  2.758+3.339j,  -3.512+0.398j,   5.824-4.045j,   1.149-3.705j],
       [  1.326-0.j   ,   1.191-4.479j,  -3.263+6.19j ,   8.939-0.j   ],
       [  2.758-3.339j,  12.368-1.464j,   0.661+2.127j,   1.149+3.705j],
       [  1.475+3.311j,   4.555+3.657j,   4.705+3.373j,  -2.570+1.152j]])

Notice that rfft2_result matches fft2_result[:, :4], at least up to numerical error:

In [12]: np.allclose(rfft2_result, fft2_result[:, :4])
Out[12]: True

We could also choose to keep the top half of the output rather than the left half, by using the axes argument to np.fft.rfft2:

In [13]: np.fft.rfft2(x, axes=[1, 0])
Out[13]: 
array([[ -5.834+0.j   ,   1.084-2.33j ,  -6.504-3.884j,   3.228-0.j   ,  -6.504+3.884j,   1.084+2.33j ],
       [  1.475-3.311j,   1.865-3.699j,   2.777-0.095j,  -2.570-1.152j,   4.705-3.373j,   4.555-3.657j],
       [  2.758+3.339j,  -3.512+0.398j,   5.824-4.045j,   1.149-3.705j,   0.661-2.127j,  12.368+1.464j],
       [  1.326+0.j   ,   1.191-4.479j,  -3.263+6.19j ,   8.939-0.j   ,  -3.263-6.19j ,   1.191+4.479j]])

As the documentation for np.fft.rfftn says, NumPy performs a real FFT over the last axis specified, and complex FFTs over the other axes.

Of course, there's still some redundancy in rfft2_result: we could discard the bottom half of the first column, and the bottom half of the last column, and still be able to reconstruct them using the same symmetry as before. And the entries at positions [0, 0], [0, 3], [3, 0] and [3, 3] are all going to be real, so we could discard their imaginary parts. But that would leave us with a less convenient array representation.

回答2:

Also note the ordering of the coefficients in the fft output:

According to the doc: by default the 1st element is the coefficient for 0 frequency component (effectively the sum or mean of the array), and starting from the 2nd we have coeffcients for the postive frequencies in increasing order, and starts from n/2+1 they are for negative frequencies in decreasing order. To have a view of the frequencies for a length-10 array:

np.fft.fftfreq(10) the output is: array([ 0. , 0.1, 0.2, 0.3, 0.4, -0.5, -0.4, -0.3, -0.2, -0.1])

Use np.fft.fftshift(cf), where cf=np.fft.fft(array), the output is shifted so that it corresponds to this frequency ordering: array([-0.5, -0.4, -0.3, -0.2, -0.1, 0. , 0.1, 0.2, 0.3, 0.4]) which makes for sense for plotting.

In the 2D case it is the same. And the fft2 and rfft2 difference is as explained by others.

回答3:

from docs:

https://docs.scipy.org/doc/numpy/reference/generated/numpy.fft.rfft2.html

This is really just rfftn with different default behavior. For more details see rfftn.

numpy.fft.rfft2(a, s=None, axes=(-2, -1), norm=None)[source]

vs.

numpy.fft.rfftn(a, s=None, axes=None, norm=None)[source]

https://docs.scipy.org/doc/numpy/reference/generated/numpy.fft.rfftn.html#numpy.fft.rfftn

Notes

The transform for real input is performed over the last transformation axis, as by rfft, then the transform over the remaining axes is performed as by fftn. The order of the output is as for rfft for the final transformation axis, and as for fftn for the remaining transformation axes.

See fft for details, definitions and conventions used.

https://docs.scipy.org/doc/numpy/reference/generated/numpy.fft.fft.html#numpy.fft.fft

I have not worked with 2d fftn but I created this for interpretation of 1d fft, which might give you some insight into the interpretation of your 2d output:

import math
import numpy as np

PERIOD          = 30
SIFT            = 2 # integer from 1 to PERIOD/2

def fourier_series(array, period, sift):

    # Create an array of length data period; then reverse its order
    signal          = (array[-period:])[::-1]  

    # Extract amplitude and phase in (a + bi) complex form
    complex_fft     = np.fft.fft(signal)

    ''' Calculate amplitude, phase, frequency, and velocity '''
    # Define empty lists for later use
    amplitude       = []
    phase           = []
    frequency       = []
    velocity        = []

    # Extract real and imaginary coefficients from complex scipy output
    for n in range(period, 0, -1):   
        amplitude.append(complex_fft.real[-n])
        phase.append(complex_fft.imag[-n])

    # The final equation will need to be divided by period
    # I do it here so that it is calculated once saving cycles     
    amplitude = [(x/period) for x in amplitude]    

    # Extract the carrier
    carrier = max(amplitude)

    # The frequency is a helper function of fft 
    # It only has access to the length of the data set
    frequency.append(np.fft.fftfreq(signal.size, 1))

    # Convert frequency array to list
    frequency       = frequency[-1]

    # Velocity is 2*pi*frequency; I do this here once to save cpu time    
    velocity        = [x*2*math.pi for x in frequency]

    ''' Calculate the Full Spectrum Sinusoid '''
    # Recombine ALL elements in the form An*sin(2*pi(Fn) + Pn) for full spectrum
    full_spectrum   = 0
    for m in range(1, period+1):
        full_spectrum += amplitude[-m]*(1+math.sin(velocity[-m] + phase[-m]))

    ''' Calculate the Filtered Sinusoid '''
    # Normalize user sift input as an integer
    sift = int(sift)

    # If sift is more than half of the period, return full spectrum
    if sift >= period/2:
        filtered_transform = full_spectrum

    # If sift is 0 or 1, return the carrier    
    else:
        filtered_transform = carrier

        # For every whole number of sift over 1, but less than 0.5*period: 
        # Add an 2 elements to the sinusoid respective of 
        # a negative and positive frequency pair
        if sift > 1:
            for m in range(1, sift): 
                p = period - m
                filtered_transform += amplitude[-m]*(1+math.sin(velocity[-m] + phase[-m]))
                filtered_transform += amplitude[-p]*(1+math.sin(velocity[-p] + phase[-p]))

    ''' Print Elements and Return FFT'''
    if 1:
        print('**********************************')
        print('Carrier: %.3f' % amplitude[-period])
        print(['%.2f' % x for x in amplitude])    
        print(['%.2f' % x for x in velocity]) 
        print(['%.2f' % x for x in phase])

    return filtered_transform, carrier, full_spectrum

stochastic = # Your 1d input array
y, y_carrier, y_full = fourier_series(stochastic, PERIOD, SIFT)