Method for interpolating seismic data by anti-alias, anti-leakage Fourier transform

1. A method for interpolating traces in seismic data, comprising:

generating an estimated frequency-wavenumber spectrum by applying a first Anti-leakage Fourier transform method to unaliased frequency components of temporal-transformed seismic data and applying a second Anti-leakage Fourier transform method to aliased frequency components of the temporal-transformed seismic data, wherein the second Anti-leakage Fourier transform method applies an absolute frequency-wavenumber spectrum extrapolated from unaliased frequencies to aliased frequencies to weight frequency-wavenumber components of the aliased frequencies; and

applying an inverse temporal and spatial Fourier transform to the estimated frequency-wavenumber spectrum thereby generating trace interpolation of the seismic data.

2. The method of claim 1 , wherein the first Anti-leakage Fourier transform method is a standard Anti-leakage Fourier transform method.

3. The method of claim 1 , wherein the unaliased frequencies are substantially the lower frequencies in the seismic data and the aliased frequencies are substantially the higher frequencies in the seismic data.

4. The method of claim 2 , initially comprising:

obtaining input seismic data in the time-space domain;

applying a temporal Fourier transform to the input seismic data, generating transformed seismic data;

determining which frequency components in the transformed seismic data correspond to unaliased frequencies and which frequency components in the transformed seismic data correspond to aliased frequencies; and

applying the first Anti-leakage Fourier transform method to each of the unaliased frequency components from the transformed seismic data.

5. The method of claim 4 , wherein the temporal Fourier transform is a Fast Fourier Transform.

6. The method of claim 4 , further comprising:

determining a number of iterations N 1 for the first Anti-leakage Fourier transform method.

7. The method of claim 6 , wherein the determining a number of iterations N 1 comprises:

determining a threshold ε 1 for the first Anti-leakage Fourier transform method.

8. The method of claim 7 , further comprising:

initializing an estimated frequency-wavenumber spectrum to zero for the unaliased frequency component;

applying a spatial Fourier transform to the selected unaliased frequency component; and

performing the following for N 1 iterations:

selecting a largest wavenumber component in the transformed unaliased frequency component;

adding the largest wavenumber component to the estimated frequency-wavenumber spectrum for the unaliased frequency component;

applying an inverse spatial Fourier transform to the selected largest wavenumber component; and

subtracting the inverse-transformed largest component from the selected unaliased frequency component, generating a corrected frequency component.

9. The method of claim 8 , wherein the spatial Fourier transform is a Discrete Fourier Transform.

10. The method of claim 8 , wherein the inverse spatial Fourier transform is an inverse Discrete Fourier Transform.

11. The method of claim 8 , wherein the spatial Fourier transform is a Nonuniform Fast Fourier Transform.

12. The method of claim 8 , wherein the inverse spatial Fourier transform is an inverse Nonuniform Fast Fourier Transform.

13. The method of claim 8 , further comprising:

combining the estimated frequency-wavenumber spectra for the unaliased frequency components, generating the unaliased estimated frequency-wavenumber spectrum;

taking an absolute value of the unaliased estimated frequency-wavenumber spectrum, generating an absolute frequency-wavenumber spectrum;

extrapolating the absolute frequency-wavenumber spectrum to unaliased frequencies, generating an extrapolated absolute frequency-wavenumber spectrum; and

applying the second Anti-leakage Fourier transform method to each of the aliased frequency components from the transformed seismic data.

14. The method of claim 13 , further comprising:

smoothing the absolute frequency-wavenumber spectrum before extrapolating the absolute frequency-wavenumber spectrum to unaliased frequencies.

15. The method of claim 13 , further comprising:

determining a number of iterations N 2 for the second Anti-leakage Fourier transform method.

16. The method of claim 15 , wherein the determining a number of iterations N 2 comprises:

determining a threshold ε 2 for the second Anti-leakage Fourier transform method.

17. The method of claim 8 , further comprising:

initializing an estimated frequency-wavenumber spectrum to zero for the aliased frequency component;

applying a spatial Fourier transform at the selected aliased frequency component; and

performing the following for N 2 iterations:

applying the extrapolated absolute frequency-wavenumber spectrum to the transformed aliased frequency component, generating a weighted aliased frequency-wavenumber spectrum;

selecting the largest wavenumber component in the weighted aliased frequency-wavenumber spectrum;

obtaining the unweighted wavenumber component that corresponds to the largest weighted wavenumber component;

adding the corresponding unweighted wavenumber component to the estimated frequency-wavenumber spectrum for the aliased frequency component;

computing an inverse spatial Fourier transform of the largest unweighted wavenumber component; and

subtracting the computed inverse-transformed largest wavenumber component from the aliased frequency component, generating a corrected aliased frequency component.

18. The method of claim 17 , further comprising:

applying an inverse spatial Fourier transform to the selected aliased frequency component, generating interpolated seismic data back in the frequency-space domain.

19. The method of claim 18 , wherein the inverse spatial Fourier transform is an inverse Discrete Fourier Transform.

20. The method of claim 19 , wherein the inverse spatial Fourier transform is an inverse Nonuniform Fast Fourier Transform.

21. The method of claim 18 , further comprising:

applying an inverse temporal Fourier transform to the selected aliased frequency component, generating interpolated seismic data back in the time-space domain.

22. The method of claim 21 , wherein the inverse Fourier transform is an inverse Fast Fourier Transform.