Systems and methods for de-noising seismic data
Systems and methods are provided for de-noising seismic data recorded by seismic receivers. A first portion of the seismic data having a first signal-to-noise ratio (SNR) to is processed to generate a de-noising operator or function. The de-noising operator is applied to a second portion of seismic data having a second SNR to remove noise from the second portion of the seismic data, where the first SNR is greater than the second SNR.
1. A method for exploring structure of a subsurface of the earth, the method comprising:
receiving a first seismic data acquired from the subsurface;
receiving a second seismic data acquired from the subsurface;
processing with a processor the first seismic data having a first signal-to-noise ratio (SNR) to derive a denoising operator;
applying the de-noising operator to the second seismic data having a second SNR to remove noise from the second seismic data thereby obtaining denoised seismic data; and
generating an image of the subsurface of the earth illustrating layers thereof based on the denoised seismic data to estimate location of a hydrocarbon reservoir in the subsurface,
wherein the first SNR is greater than the second SNR, and
wherein the first and second seismic data are recorded by seismic receivers detecting seismic excitations generated by a seismic source, after the seismic excitations travel through the subsurface of the earth.
2. The method of claim 1 , wherein the first seismic data and the second seismic data come from a same receiver component.
3. The method of claim 1 , wherein the first seismic data and the second seismic data come from different receiver components.
4. The method of claim 3 , wherein the first seismic data comes from a hydrophone and the second seismic data comes from a particle motion component.
5. The method of claim 4 , wherein the particle motion component is a geophone, particle velocity, or particle acceleration component.
6. The method of claim 1 , wherein at least one of the first seismic data and the second seismic data is processed to have similar characteristics to the other at least one of the second seismic data and first seismic data, respectively.
7. The method of claim 1 , wherein at least one of the first seismic data and the second seismic data is a combination of data from more than one receiver component.
8. The method of claim 1 , wherein the first seismic data is associated with a first frequency bandwidth and the second seismic data is associated with a second frequency bandwidth.
9. The method of claim 8 , wherein the first frequency bandwidth is higher frequency than the second frequency bandwidth.
10. The method of claim 9 , wherein the first frequency bandwidth includes only one or more frequencies of 30 Hz or greater and the second frequency bandwidth includes only one or more frequencies of less than 30 Hz.
11. The method of claim 1 , wherein the step of processing further comprises:
deriving a filter from the first portion of the seismic data as the de-noising operator.
12. The method of claim 11 , wherein the step of processing further comprises:
deriving a prediction filter, as the de-noising operator, based on the first seismic data; and
wherein the step of applying further comprises:
applying the prediction filter to the second seismic data.
13. The method of claim 12 , wherein the step of deriving a prediction filter comprises deriving a prediction filter at a first frequency, and the step of applying the prediction filter comprises applying the prediction filter at a second frequency.
14. The method of claim 1 , wherein the step of processing further comprises:
deriving a model from the first seismic data as the de-noising operator.
15. The method of claim 14 , wherein the model is one of an FK domain, curvelet domain, SVD domain, rank reduction domain, Radon domain, tau-p domain, parabolic domain, hyperbolic domain, contourlet domain or ridgelet domain.
16. The method of claim 15 , wherein a comparison between the first seismic data and the second seismic data is made to derive a masking function to separate signal and noise.
17. The method of claim 14 , wherein the model is a Radon transform model and the step of processing further comprises:
deriving the Radon model based on sparseness weights from a previous iteration;
calculating sparseness weights for a current iteration; and
repeating the deriving and calculating steps using lower frequency values to de-noise the seismic data in a manner which is constrained by higher frequency values.
18. The method of claim 1 , wherein data from the high SNR ratio data is used to replace noisy data in the low SNR ratio data.
19. The method of claim 1 , wherein the step of processing further comprises:
deriving one or more common reflection surface (CRS) operators from the first seismic data as the de-noising operator.
20. The method of claim 19 , wherein the step of applying further comprises:
applying the one or more CRS operators to perform CRS processing of the second seismic data.
21. The method of claim 1 , wherein the first SNR is greater than 1.
22. The method of claim 1 , wherein the first seismic data is deghosted hydrophone data, and the second seismic data is deghosted particle velocity data.
23. The method of claim 1 , wherein the first seismic data is synthesized particle velocity data generated from hydrophone data, and the second seismic data is recorded particle velocity data.
24. The method of claim 1 , wherein the first seismic data is recorded hydrophone data and the second seismic data is combined hydrophone and particle velocity data.
25. A system for seismic exploration of structure of a subsurface of the earth, the system comprising:
an interface configured to receive a first seismic data and a second seismic data acquired from the subsurface; and
at least one processor connected to the interface and configured to process the first seismic data having a first signal-to-noise ratio (SNR) to derive a denoising operator, and also configured to apply the de-noising operator to the second seismic data having a second SNR to remove noise from the second seismic data thereby obtaining denoised seismic data, and also configured to generate an image of the subsurface illustrating layers thereof based on the denoised seismic data, to estimate location of a hydrocarbon reservoir in the subsurface,
wherein the first SNR is greater than the second SNR,
wherein the first and second seismic data are recorded by seismic receivers detecting seismic excitations generated by a seismic source, after the seismic excitations travel through the subsurface.