A method and system for improving signal-to-noise ratio of prestack seismic data

Abstract

The application provides a method and system for improving signal-to-noise ratio of pre-stack seismic data, and belongs to the field of seismic data processing. The method for improving signal-to-noise ratio of pre-stack seismic data constructs a data volume gather related to a center surface element gather according to offset distance and azimuth information in seismic data, and sorts based on the azimuth and offset distance of the center surface element gather, thereby effectively suppressing noise interference and improving energy of weak effective signals. The application improves energy of weak effective signals while effectively suppressing noise interference, protects effective information in complex structures from being lost, and improves signal-to-noise ratio of pre-stack seismic data as a whole on the basis of relative fidelity.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of seismic data processing, and particularly relates to a method and system for improving signal-to-noise ratio of pre-stack seismic data. BACKGROUND

[0002] With the deepening of exploration, the exploration range has changed from simple structure oil and gas reservoirs to complex structure oil and gas reservoirs and subtle oil and gas reservoirs. The quality of collected seismic data is uneven, and it is necessary to finely pretreat the seismic data, find useful information from the seismic data, suppress interference information, and obtain high signal-to-noise ratio, high resolution, and high fidelity seismic data as much as possible, so as to improve the accuracy of seismic interpretation. Since the characteristics of complex structure oil and gas reservoirs are not clearly displayed on the seismic profile, it is urgent to develop a technology capable of suppressing noise interference of seismic data, recovering weak reflection signals in seismic data under the premise of protecting effective signals in seismic data, so as to obtain high signal-to-noise ratio and high resolution seismic migration results, which is helpful for subsequent seismic processing and interpretation work.

[0003] Conventional denoising technology is based on the difference in the distribution of signals and noise in the frequency-wavenumber domain or the time-space domain. For example, the conventional F-X domain predictive filtering has a shielding effect on the amplitude of inclined and curved phase axes, especially when the original signal-to-noise ratio is relatively low, the amplitude is severely lost. Some scholars have proposed K-L filtering, which uses the least square method to extract the orthogonal linear transformation filter of the correlation signal (or noise) from the multi-channel record, controls the strength of the extracted signal correlation, and concentrates the coherent energy in several main components, and the incoherent random noise is distributed in all main components, but it requires a higher original signal-to-noise ratio. The vector decomposition method introduces the concept of unit correlation vector according to the strong correlation between adjacent seismic channels of seismic data, decomposes the amplitude value of the seismic data into a correlation part and a non-correlation part, suppresses the non-correlation part decomposed, and improves the signal-to-noise ratio. The Hilbert-Huang transform is a signal analysis method proposed by Norden Huang et al. in 1998, which is suitable for nonlinear and non-stationary signals, and has the characteristics of strong adaptability and good time-frequency locality. The basic idea of the wavelet threshold denoising proposed by Donoho is that after the signal is transformed by the wavelet, the wavelet coefficients of the signal are larger and the wavelet coefficients of the noise are smaller. By selecting a suitable threshold, the wavelet coefficients greater than the threshold are considered to be generated by the signal and should be retained, and the wavelet coefficients less than the threshold are considered to be generated by the noise and are set to zero to achieve the purpose of denoising. In addition, there are Fourier transform, wavelet transform, curvelet transform and other methods.

[0004] Chinese patent publication CN110031899A discloses a weak signal extraction algorithm based on compressed sensing, which includes: step 1, inputting noisy seismic data dobs, measurement matrix Ф, and dictionary base matrix Ψ; step 2, CEEMD decomposition, i.e. collection empirical mode decomposition, decomposing dobs into K IMF components; step 3, performing IMF component autocorrelation analysis; step 4, respectively applying compressed sensing CS denoising method to process IMFk; step 5, reconstructing the IMF components after CS denoising and the remaining components to obtain the denoised seismic signal; Chinese patent publication CN111175811A discloses a seabed seismograph device capable of improving signal-to-noise ratio, which comprises a seismograph, a protective shell for protecting the seismograph, the seismograph being fixedly installed in the protective shell, a ballast anchor, a seismograph mounting structure comprising an upward-opening positioning groove arranged on the ballast anchor, a positioning block arranged at the bottom of the protective shell for cooperating with the positioning groove, a sealed cavity arranged in the positioning block, an electromagnet arranged in the sealed cavity and a battery for providing power for the electromagnet, and an iron plate embedded in the bottom of the positioning groove, the positioning block being inserted into the positioning groove, and the positioning block being tightly attached to the iron plate, the electromagnet and the iron plate being attracted to each other to tightly connect the seismograph, the protective shell and the ballast anchor into one body; Chinese patent publication CN104732493A discloses a SAR image denoising algorithm based on Primal Sketch classification and SVD domain improved MMSE estimation, which mainly includes: first, in the Primal Sketch algorithm, the energy image is improved by using the double-neighborhood contrast enhancement method, then the SAR image is divided into edge class and non-edge class by using the Primal Sketch algorithm; the pixel points of the two classes are respectively decomposed by NLSVD, the singular value matrix is estimated by using the minimum mean square error criterion containing the shrinkage factor, and the estimated values of the edge class and the non-edge class are obtained by inverse transformation; finally, the edge coefficients are calculated, and the boundaries of the edge class and the non-edge class are fused by the Butterworth fusion method to obtain the denoising result.

[0005] In complex areas, the interlayer wave impedance difference of the target layer in actual seismic data is small, the reflection ability is weak, the continuity of reflection wave group is poor, the signal-to-noise ratio is low, the geological structure movement is strong, the stratum folding is serious, the dip angle changes greatly, the overthrust is strong, and the wave field is complex, which causes the low signal-to-noise ratio of the original seismic data and the weak energy of the effective reflection signal in the prestack record. SUMMARY

[0006] The present application aims at solving the problems in the prior art, and provides a method and system for improving signal-to-noise ratio of pre-stack seismic data, which utilizes shot distance and azimuth information of a trace to improve correlation of a data volume in space, sorts the azimuth and offset distance of a center bin trace set, effectively suppresses noise interference while improving energy of weak effective signals, protects effective information in complex structures from being lost, and improves signal-to-noise ratio of pre-stack seismic data as a whole on the basis of relative fidelity.

[0007] The present application is implemented by the following technical solutions.

[0008] In a first aspect, the present application provides a method for improving signal-to-noise ratio of pre-stack seismic data, which constructs a data volume trace set related to a center bin trace set according to offset distance and azimuth information in seismic data, and sorts the azimuth and offset distance of the center bin trace set to effectively suppress noise interference while improving energy of weak effective signals.

[0009] The present application is further improved in that the method comprises:

[0010] Step one: input a pre-stack seismic trace set to obtain a data volume trace set G(l0,c0,x j ,η,t);

[0011] Step two: construct a center bin trace set by using the data volume trace set G(l0,c0,x j ,η,t);

[0012] Step three: perform azimuth sorting and offset distance sorting on the center bin trace set to obtain a processed pre-stack trace set;

[0013] Step four: sequentially perform steps two to three on each bin trace set, and finally obtain a processed pre-stack trace set of all bin trace sets, which constitutes a processed pre-stack three-dimensional seismic data result.

[0014] The present application is further improved in that the operation of step two comprises:

[0015] The center bin trace set is constructed by using formula (2):

[0016] |f(l0,c0,x i ,η,t)-G(l0,c0,x j ,η,t)|→min (2)

[0017] wherein f(l0,c0,x i ,η,t) is the center bin trace set, l0 and c0 are line number and point number of the center bin trace set respectively, and x, t and η are offset distance, time and azimuth respectively.

[0018] The further improvement of the present application is that the operation of step three comprises:

[0019] The central bin gather f(l0, c0, x i ,η,t) is azimuthally sorted and offset sorted by formula (3) to obtain the sorted central bin gather:

[0020]

[0021] Wherein, G(l0, c0, t) is the sorted central bin gather, i.e. the processed pre-stack gather.

[0022] The second aspect of the present application provides a system for improving the signal-to-noise ratio of pre-stack seismic data, which comprises:

[0023] The division unit is used for inputting the pre-stack gather to obtain the data volume gather G(l0, c0, x j ,η,t);

[0024] The central bin gather construction unit is connected with the division unit and is used for constructing the central bin gather by using the data volume gather G(l0, c0, x j ,η,t);

[0025] The sorting unit is connected with the central bin gather construction unit and is used for azimuthally sorting and offset sorting the central bin gather to obtain the processed pre-stack gather;

[0026] The output unit is connected with the sorting unit and is used for outputting the processing result.

[0027] The further improvement of the present application is that the central bin gather construction unit constructs the central bin gather by using formula (2):

[0028] |f(l0, c0, x i ,η,t)-G(l0, c0, x j ,η,t)|→min (2)

[0029] Wherein, f(l0, c0, x i ,η,t) is the central bin gather, l0 and c0 are the line number and point number of the central bin gather respectively, and x, t and η are offset, time and azimuth respectively.

[0030] The further improvement of the present application is that the sorting unit sorts the central bin gather f(l0, c0, x i ,η,t) by formula (3) to obtain the sorted central bin gather:

[0031]

[0032] G (l0, c0, t) is the center bin gather after sorting, i.e.

[0033] The further improvement of the present application is that the output unit forms the processed pre-stack 3D seismic data result with the processed pre-stack gathers of all bin gathers and outputs the processed pre-stack 3D seismic data result.

[0034] The three aspects of the present application provide a device for improving the signal-to-noise ratio of pre-stack seismic data, which comprises a memory, a processor and a computer program stored in the memory, wherein the computer program is executed by the processor to perform the following steps:

[0035] Step one: input the pre-stack seismic gather to obtain the data volume gather G (l0, c0, x j ,η,t) ;

[0036] Step two: construct the center bin gather by using the data volume gather G (l0, c0, x j ,η,t) ;

[0037] Step three: perform azimuth sorting and offset sorting on the center bin gather to obtain the processed pre-stack gather;

[0038] Step four: sequentially perform steps two to three on each bin gather, and finally the processed pre-stack gathers of all bin gathers form the processed pre-stack 3D seismic data result.

[0039] The fourth aspect of the present application provides a computer readable storage medium, which stores at least one computer executable program, wherein the at least one program is executed by the computer to make the computer perform the steps in the method for improving the signal-to-noise ratio of pre-stack seismic data.

[0040] Compared with the prior art, the present application has the beneficial effects that the energy of weak effective signals is improved while effectively interfering with noise, the effective information in complex structures is protected from loss, and the signal-to-noise ratio of pre-stack seismic data is improved as a whole on the basis of relative fidelity. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1-1 Large bin sorting schematic diagram;

[0042] Figure 1-2 Center bin sorting schematic diagram;

[0043] Figure 2 Schematic diagram of constructing center gather;

[0044] Figure 3Step block diagram of the method of the present application;

[0045] Figure 4-1 Theoretical ten-layer model without noise;

[0046] Figure 4-2 Add random noise with amplitude value of 1.0;

[0047] Figure 4-3 Recognition result of signal recognition on the noise-added model by using the present application;

[0048] Figure 5-1 Velocity spectrum before processing by using the present application;

[0049] Figure 5-2 Gather before processing by using the present application;

[0050] Figure 5-3 Velocity spectrum after processing by using the present application;

[0051] Figure 5-4 Gather after processing by using the present application;

[0052] Figure 6-1 Gather before processing by using the present application;

[0053] Figure 6-2 Gather after processing by using the present application;

[0054] Figure 6-3 Difference gather;

[0055] Figure 7-1 Stack profile before processing by using the present application;

[0056] Figure 7-2 Stack profile after processing by using the present application;

[0057] Figure 8-1 Signal-to-noise ratio model before processing by using the present application;

[0058] Figure 8-2 Signal-to-noise ratio model after processing by using the present application;

[0059] Figure 9-1 Imaging profile before processing by using the present application;

[0060] Figure 9-2 Imaging profile after processing by using the present application;

[0061] Figure 10 Schematic structural diagram of the system of the present application. DETAILED DESCRIPTION

[0062] The present application will be further described in detail below with reference to the accompanying drawings:

[0063] To improve the seismic imaging accuracy, the weak effective signal in seismic data needs to be fully utilized to improve the seismic data quality. In view of the problems of small interlayer wave impedance difference, weak reflection ability, poor continuity of reflection wave group, low signal-to-noise ratio and complex wave field in a complex exploration area, the present application develops a method for improving the signal-to-noise ratio of seismic data, which is based on the sorting of the azimuth and offset of seismic data, and based on the sorted seismic data to realize the overall improvement of the signal-to-noise ratio and restore the weak effective signal submerged by noise. The present application fully utilizes the effective signal in seismic data to improve the imaging signal-to-noise ratio as much as possible, and protects the weak signal in seismic data from loss.

[0064] In view of the problem of low signal-to-noise ratio weak signal in a complex structure exploration area, the present application develops a technology for improving the signal-to-noise ratio of prestack seismic data. The present application utilizes the shot distance and azimuth information of the trace gather to improve the correlation of the data body in space, realizes the effective suppression of noise interference and the improvement of the energy of weak effective signal based on the azimuth and offset sorting of the center bin trace gather, protects the effective information in the complex structure from loss, improves the signal-to-noise ratio of prestack seismic data on the basis of relative fidelity, and obviously improves the effect of prestack trace gather, stack and imaging profile.

[0065] (1) Seismic data center trace gather construction technology

[0066] For prestack seismic data, the center bin construction technology based on shot distance and azimuth is developed to improve the correlation of seismic data in space. The center bin trace gather is the development of the traditional large bin. The azimuth sorting is added on the basis of shot distance sorting. The different sorting conditions between the trace gathers make the obtained center bin trace gather have stronger spatial correlation, and improve the continuity of the events while effectively suppressing the interference.

[0067] Figure 1-1 A construction schematic diagram of the traditional 3x3 large bin sorting is shown. A certain trace in the center bin is selected as the center reference trace, the offset of the center trace is selected as the sorting condition of the trace gather, and the trace gather within the allowable error range is selected from the adjacent bins as the relevant trace of the center trace gather. The same processing is performed on all the traces in the center bin, and the large bin trace gather corresponding to the center bin trace gather is obtained. The large bin has the advantage of fast construction speed, but the sorting condition is not very strict, so that the correlation of the center large bin is weak. When the underground structure is complex, the effective signal will be seriously interfered, and even damaged.

[0068] Figure 1-2 A construction schematic diagram of the center bin sorting is shown. Different from the large bin, in addition to the introduction of the azimuth sorting, each center trace gather will form a corresponding bin trace gather. The trace gather is the bin trace gather after the sorting of the data in the spatial data body, and the local correlation between the trace gathers is strengthened.

[0069] Figure 2 To construct a certain center gather corresponding to the face gather, a certain trace is selected from the center face, i.e. a specified offset and azimuth are selected from all offsets and azimuths as the sorting condition, and a trace gather within the allowable error range is extracted from the local spatial data volume to form a new face gather with the relevant traces of the center gather. The mathematical expression is as follows:

[0070]

[0071] The formula (1) is used in the data correlation sorting in the following step one.

[0072] Wherein, l, c are the line number and point number in the local time window respectively, l0, c0 are the line number and point number of the center face respectively; m, n are the line number and point number of the current gather, x, t are the offset and time respectively, and R is the optimization radius of the center face gather, i.e. the given error threshold.

[0073] For the center processing CMP gather X(l0, c0, x, t), optimization of the three parameters l0, c0 and x is needed, and the best line number and CMP number found are m0 and n0.

[0074] (2) Seismic data signal-to-noise ratio improvement technology

[0075] The spatial gather obtained through the center face construction technology has good local correlation characteristics. Due to the different division of azimuth and offset, the sorted face gather data volume is several times of the original pre-stack gather. In order to output the gather after signal-to-noise ratio improvement consistent with the original pre-stack gather, the center face gather based on offset and azimuth sorting is needed for the center face data volume, so as to obtain the sorted center face gather. This process is to restore the azimuth and offset gather to the scalar gather output after signal-to-noise ratio improvement. The objective function expression is as follows:

[0076] |f(l0,c0,x i ,η,t)-G(l0,c0,x j ,η,t)|→min (2)

[0077] x, η are offset and azimuth, f(l0, c0, x i , η, t) is the center face gather, and G(l0, c0, x j , η, t) is the data volume gather constituting the large face.

[0078] The center face gather and the original gather based on the optimization of azimuth and offset will satisfy the best gather of the objective function as the output result, and the expression is as follows:

[0079]

[0080] ηmin, x min represent azimuth minimum value, offset minimum value respectively, x max, ηmax represent azimuth maximum value, offset maximum value respectively.

[0081] Where f(l0, c0, x, η, t) is the center bin gather containing offset and azimuth, G(l0, c0, t) is the output gather after improving signal-to-noise ratio.

[0082] Formula (3) sorts the related gathers with different azimuth and offset by integral summation, and the result G obtained finally no longer contains azimuth and offset, and a sorting process is realized through the calculation of formula (3).

[0083] (3) Technical idea and technical implementation

[0084] In view of the low signal-to-noise ratio weak signal problem existing in the complex structure exploration area, the prestack seismic data signal-to-noise ratio improvement technology is researched, the correlation of the data body in space is improved through the shot-receiver distance and azimuth information contained in the gather, the azimuth and offset sorting based on the center bin gather is realized, the effective noise interference is suppressed, the energy of the weak effective signal is improved, the effective information in the complex structure is protected from being lost, and the overall signal-to-noise ratio of the prestack seismic data is improved on the basis of relative fidelity.

[0085] As shown in Figure 3 , the method of the present application comprises:

[0086] Step one: input all prestack seismic gathers (referred to as prestack gathers), and divide the azimuth and offset according to the prestack seismic gathers: the azimuth and offset are generally divided according to experience, in general, the more rich the azimuth and offset information in the seismic data, the smaller the division; on the contrary, the narrow azimuth and short offset should be divided larger; in general, it is necessary to ensure that there are uniform data bodies in each division area, and the specific division method is realized by using the prior art, which will not be described here.

[0087] The prestack seismic gather comprises a plurality of bin gathers, and the data body gather G(l0, c0, x j ,η,t) related to the center bin is obtained through step one.

[0088] Step two: constructing the center bin gather: the center bin gather of the current bin gather is constructed by using formula (2):

[0089] According to the data body gather G(l0, c0, x j ,η,t) obtained in step one, the gather with the smallest error is found by using formula (2), and the gather is the center bin gather f(l0, c0, x i,η,t), wherein G(l0,c0,x j ,η,t) is constructed in step one, and the trace gather f(l0,c0,x j ,η,t) with the minimum error is found from G(l0,c0,x i ,η,t) to construct a new central bin trace gather f(l0,c0,x

[0090] Step three: azimuth sorting and offset sorting of the central bin trace gather:

[0091] Based on the constructed central bin trace gather f(l0,c0,x i ,η,t), the central bin trace gather f(l0,c0,x i ,η,t) is azimuth sorted and offset sorted by using formula (3) to obtain the sorted central bin trace gather.

[0092] The central bin trace gather f(l0,c0,x i ,η,t) is the combination of all data bodies related to the central trace gather, and contains different offsets and azimuths x,η, the trace gather G(l0,c0,t) obtained by sorting the trace gather with consistent offset and azimuth of the central bin trace gather by using formula (3) is the sorted central bin trace gather, that is, the processed pre-stack trace gather, so that the correlation of the data bodies in the local space is used to obtain the calculated trace gather with improved signal-to-noise ratio.

[0093] Step four: steps two to three are sequentially performed on each bin trace gather, and the processed pre-stack trace gathers of all bin trace gathers finally obtained constitute the processed pre-stack 3D seismic data result.

[0094] The seismic data is composed of many bin trace gathers, and is arranged according to the line point number in the seismic data, and a bin trace gather is input according to the line number and point number in the input seismic data, steps one to three are performed on the bin trace gather, the selected central bin is the bin to be processed each time, and the processed pre-stack trace gather of the bin trace gather is obtained after the processing, and the processing of the next bin trace gather is performed after the processing of the bin trace gather is completed, until all bin trace gathers of the entire data body are processed to obtain the processed pre-stack 3D seismic data result.

[0095] The application also provides a system for improving the signal-to-noise ratio of pre-stack seismic data, as shown in Figure 10 , the system comprises:

[0096] The dividing unit 10 is used for inputting the pre-stack trace gather to obtain the data body trace gather G(l0,c0,x j ,η,t);

[0097] A center surface bin gather set unit 20 is connected with the division unit 10, and is used for constructing a center surface bin gather set by using the data volume gather set G(l0,c0,x j ,η,t);

[0098] A sorting unit 30 is connected with the center surface bin gather set unit 20, and is used for azimuth angle sorting and offset distance sorting on the center surface bin gather set to obtain a processed pre-stack gather set;

[0099] An output unit 40 is connected with the sorting unit 30, and is used for outputting a processing result.

[0100] The application further provides a device for improving signal-to-noise ratio of pre-stack seismic data, which comprises a memory, a processor and a computer program stored in the memory, and the computer program is executed by the processor to perform the following steps:

[0101] Step one: inputting a pre-stack seismic gather set to obtain a data volume gather set G(l0,c0,x j ,η,t);

[0102] Step two: constructing a center surface bin gather set by using the data volume gather set G(l0,c0,x j ,η,t);

[0103] Step three: performing azimuth angle sorting and offset distance sorting on the center surface bin gather set to obtain a processed pre-stack gather set;

[0104] Step four: sequentially performing the processing of step two and step three on each surface bin gather set, and finally obtaining a processed pre-stack gather set of all surface bin gather sets, which constitutes a processed pre-stack three-dimensional seismic data result.

[0105] The application further provides a computer readable storage medium, which stores at least one computer executable program, and the at least one program is executed by the computer to make the computer perform the steps in the method for improving signal-to-noise ratio of pre-stack seismic data.

[0106] The application is applied as follows:

[0107]

Example one

[0108] In order to verify the effectiveness of the application and the application effect on processing of actual data, the application processing of actual data is performed. Figure 4-1 The ten-layer model is a theoretical model without noise, and the amplitude values of the same phase axis are 1.0, 0.9, 0.8, …, 0.1 from top to bottom, Figure 4-2 The random noise with an amplitude value of 1.0 is added, the signal of the noise-added model is identified by using the method, and the identification result is as follows: Figure 4-3As shown in Figure 4-3 The identification result shows that when the signal-to-noise ratio is lower than 0.2, the identification of the signal will have a judgment error, indicating that the identification and recovery of the method of the application for low signal-to-noise ratio data also needs to meet certain conditions.

[0109]

Example Two

[0110] Figure 5-1 to Figure 5-4 For the pre- and post-processing gathers and the corresponding velocity spectrum, it can be seen from the velocity spectrum that the energy focusing property of the velocity spectrum in the middle-deep layer is good, and in the CMP gather, the shallow layer has weak signal energy and low signal-to-noise ratio, and the energy focusing property of the velocity spectrum in the shallow layer is poor. The gather and the corresponding velocity spectrum after the processing of the application are shown in Figure 5-3 and Figure 5-4 The energy focusing property of the velocity spectrum in the shallow layer is good, the continuity of the data in-phase axis in the CMP gather and the signal-to-noise ratio are improved, and the weak effective signal is enhanced.

[0111]

Example Three

[0112] Figure 6-1 to Figure 6-3 For the pre- and post-processing gathers and the difference gather, the signal-to-noise ratio of the gather is obviously improved, and there is no obvious effective signal on the difference gather, indicating that the method will not damage the original effective signal.

[0113]

Example Four

[0114] Figure 7-1 The CMP data stack profile of all original CMP data of a certain actual work area is shown in Figure 7-2 The new stack profile obtained after the processing of the application is shown in

[0115]

Example Five

[0116] In order to further illustrate the effect before and after the processing, the signal-to-noise ratio model of the stack surface before and after the processing is quantitatively analyzed, and the signal-to-noise ratio attribute Figure 8-1 and Figure 8-2 It is shown that the overall signal-to-noise ratio is improved after the processing of the application, and the weak effective signal submerged by noise is enhanced.

[0117]

Example Six

[0118] Figure 9-1 and Figure 9-2 The migration result comparison profile before and after the processing is shown in

[0119] In summary, as a pre-stack processing method, the application achieves good application effects on pre-stack gathers, stacked sections and imaging sections, and the signal-to-noise ratio of the migration sections is improved to different degrees from shallow to deep.

[0120] Finally, it should be noted that the above technical solutions are only one embodiment of the present application, and for those skilled in the art, on the basis of the application disclosed application method and principle, various types of improvements or modifications can be easily made, and are not limited to the method described in the above embodiment of the present application, therefore, the above described method is only preferred, and does not have the meaning of limitation.

Claims

1. A method of improving signal-to-noise ratio of prestack seismic data, characterized by: The method constructs a data volume gather related to a center bin gather according to offset and azimuth information in seismic data, and sorts the center bin gather based on azimuth and offset, thereby improving the energy of weak effective signals while effectively suppressing noise interference. The method comprises: Step one: input pre-stack seismic trace set, obtain data volume trace set G(l0, c0, x j , η, t); Step two: Construct the center bin gather using the data volume gather G(l0, c0, x j , η, t), comprising: The center bin gather is constructed by using formula (2): |f(l0, c0, x i , η, t) - G(l0, c0, x j , η, t) |→ min (2) where f(l0, c0, x i , η, t) is the central bin gather, l0and c0are the line number and point number of the central bin gather, respectively, t and η are the time and azimuth, respectively, and x i is the offset of the central bin gather, and x j is the offset of the bin in the data volume gather. Step three: azimuth sorting and offset sorting of the center bin gather to obtain the processed pre-stack gather; comprising: using formula (3) to perform azimuth sorting and offset sorting on the center bin gather f(l0, c0, x i , η, t) to obtain the sorted center bin gather: Wherein, G(l0, c0, t) is the sorted center bin gather, i.e. the processed pre-stack gather; x is offset; Step four: sequentially performing steps two to three on each bin gather, and the processed pre-stack gathers of all bin gathers constitute the processed pre-stack three-dimensional seismic data result.

2. A system for improving signal-to-noise ratio of prestack seismic data, characterized by: The system comprises: The division unit is configured to input a pre-stack seismic trace set, obtain a data volume trace set G(l0, c0, x j , η, t); A center bin gather unit is configured to connect with the division unit, and configured to construct a center bin gather by using the data volume gather G(l0, c0, x j , η, t), including: The center bin gather is constructed by using formula (2): |f(l0, c0, x i , η, t) - G(l0, c0, x j , η, t) |→ min (2) wherein f(l0, c0, x i , η, t) is a central bin gather, l0, c0 are the line number and point number of the central bin gather, respectively, t, η are time and azimuth, respectively, x i is the offset of the bin within the central bin gather, and x j is the offset of the bin within the data volume gather. The sorting unit is connected with the center trace gather set unit, and is used for azimuth angle sorting and offset distance sorting of the center trace gather set to obtain a processed pre-stack gather set; the sorting unit sorts the center trace gather set f(l0, c0, x i , η, t) by using formula (3) to obtain a sorted center trace gather set: Wherein, G(l0, c0, t) is the sorted center bin gather, i.e. the processed pre-stack gather; x is offset; An output unit is connected with the sorting unit, and is configured to output the processing result.

3. The system for improving signal-to-noise ratio of prestack seismic data of claim 2, wherein: The output unit outputs the processed pre-stack three-dimensional seismic data result constituted by the processed pre-stack gathers of all bin gathers.

4. An apparatus for improving signal-to-noise ratio of prestack seismic data, the apparatus comprising: The device comprises a memory, a processor, and a computer program stored in the memory, and the computer program is executed by the processor to perform the following steps: Step one: input pre-stack seismic trace set, obtain data volume trace set G(l0, c0, x j , η, t); Step two: Construct the center bin gather using the data volume gather G(l0, c0, x j , η, t), comprising: The center bin gather is constructed by using formula (2): |f(l0, c0, x i , η, t) - G(l0, c0, x j , η, t) |→ min (2) where f(l0, c0, x i , η, t) is the central bin gather, l0and c0are the line number and point number of the central bin gather, respectively, t and η are the time and azimuth, respectively, and x i is the offset of the central bin gather, and x j is the offset of the bin in the data volume gather. Step three: azimuth sorting and offset sorting of the center bin gather to obtain the processed pre-stack gather; including: using formula (3) to perform azimuth sorting and offset sorting on the center bin gather f(l0, c0, x i , η, t) to obtain the sorted center bin gather: Wherein, G(l0, c0, t) is the sorted center bin gather, i.e. the processed pre-stack gather; x is offset; Step four: sequentially performing steps two to three on each bin gather, and the processed pre-stack gathers of all bin gathers constitute the processed pre-stack three-dimensional seismic data result.

5. A computer-readable storage medium, characterized in that: The computer readable storage medium stores at least one computer executable program, and the at least one program is executed by the computer to make the computer perform the steps in the method for improving the signal-to-noise ratio of pre-stack seismic data according to claim 1.

Images