Seismic random noise suppression method and system based on reduced-rank FX prediction filtering

Abstract

The present invention discloses a seismic random noise suppression method and system based on reduced-rank F-X predictive filtering, which uses frequency slices and prediction steps obtained by Fourier transform to construct a signal vector of forward predictive filtering, and a data matrix to solve a linear equation group of predictive filter coefficients; uses a reduced-rank matrix to solve the forward filtering signal vector; uses frequency slices and prediction steps to perform backward predictive filtering to obtain a backward filtering signal vector; synthesizes the predicted forward filtering signal vector and the backward filtering signal vector to obtain a denoised signal expression in the space-frequency domain; after all frequency slices are processed, the denoised signal expression in the frequency-space domain is Fourier inverse transformed back to the space-time domain to obtain a denoised signal. The present invention efficiently suppresses random noise in noisy data, and faithfully restores effective signals with spatial correlation, has high computational efficiency, is easily extended to high dimensions, and can be used for random noise suppression of seismic data.

Description

Technical Field

[0001] The present invention belongs to the technical field of exploration geophysics, and in particular relates to a seismic random noise suppression method and system based on reduced-rank FX prediction filtering. Background Art

[0002] Seismic exploration is a very effective geophysical method in the process of finding oil, gas and coal fields. It often includes three steps: seismic data acquisition, seismic data processing and seismic data interpretation. Seismic denoising runs through the entire process of seismic signal processing. At the same time, after years of exploration, with the continuous deepening of oil and gas field exploration and development, the exploration targets have gradually shifted from shallow and medium layers to deep layers and even ultra-deep layers, resulting in extremely low signal-to-noise ratios. The large amount of data caused by the new acquisition method and the low signal-to-noise ratio caused by the more difficult oil and gas reservoir exploration targets have posed new challenges to the suppression of seismic data noise.

[0003] The existing methods for seismic signal denoising mainly include:

[0004] Prior art 1: Filter-based denoising method: Directly use Fourier transform to transform the signal into frequency domain / spatial frequency domain (fx prediction filtering method) / wave number frequency domain (fk filtering method), and then set filters to filter out noise according to the different characteristics of the signal and noise. The disadvantage is that it can usually only remove a single type of noise, for example, fk filtering is for surface wave interference, while fx prediction filtering is often used to remove random noise.

[0005] Prior art 2: Methods based on matrix rank reduction: including rank reduction methods in the time domain (SVD denoising) and rank reduction methods in the frequency domain (Cadzow filtering), which use the fact that after the data matrix is ​​composed of the Hankel matrix, the rank of the matrix is ​​not greater than the number of data inclinations, and the presence of noise will increase the matrix rank. The noise is removed by reducing the rank of the Hankel matrix. The disadvantages are: it requires that the signal to be processed is composed of linear symmetric axes, and when calculating high-dimensional data, the rank of the Hankel matrix is ​​reduced, and the amount of calculation and storage is not small.

[0006] Prior art 3: Method based on sparse transform: Utilize the difference between the amplitude of effective signal and noise in the transform domain, and then denoise by setting the threshold function in the transform domain, such as wavelet denoising, Laden transform, curvelet transform, etc. In general, the higher the sparsity of the signal, the better the denoising. The disadvantages are: there are requirements for the sparsity of the signal; the selection of basis function and threshold function has a great influence on the denoising result; some characteristics of the basis function will introduce new noise in the denoising process, and the calculation efficiency is not high.

[0007] Prior art 4: Filtering method based on deep learning: Apply existing relatively mature neural network structures, such as convolutional neural networks and generative adversarial networks, to the field of seismic denoising. By building a large neural network structure and training it with a large number of samples, the mapping of the target task is learned. The disadvantage is that a large amount of data is required to train the network, and the amount of calculation is large. Summary of the invention

[0008] The technical problem to be solved by the present invention is to provide a seismic random noise suppression method and system based on reduced-rank FX predictive filtering in view of the deficiencies in the above-mentioned prior art, so as to solve the technical problem that the existing seismic data denoising methods cannot achieve both computational efficiency and denoising effect.

[0009] The present invention adopts the following technical solutions:

[0010] A seismic random noise suppression method based on reduced-rank FX prediction filtering comprises the following steps:

[0011] S1, using Fourier transform to transform the signal from space-time domain seismic data to space-frequency domain;

[0012] S2, obtaining a frequency slice according to the spatial-frequency domain signal obtained in step S1;

[0013] S3, using the prediction step size p and the frequency slice obtained in step S2 to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients;

[0014] S4, using the data matrix M obtained in step S3 f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix

[0015] S5, using the reduced rank matrix obtained in step S4 Solve for the forward filtered signal vector

[0016] S6, using the prediction step size p and the frequency slice obtained in step S2 to perform backward prediction filtering to obtain a backward filtering signal vector

[0017] S7, the forward filtering signal vector obtained in step S5 The back-filtered signal vector obtained in step S6 After synthesis, the denoised signal is expressed in the space-frequency domain

[0018] S8, repeating steps S2 to S7, after processing all frequency slices, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

[0019] Another technical solution of the present invention is a seismic random noise suppression system based on reduced-rank FX predictive filtering, comprising:

[0020] A transform module, which transforms the signal from the space-time domain seismic data to the space-frequency domain using Fourier transform;

[0021] A processing module obtains a frequency slice according to the spatial-frequency domain signal obtained by the transformation module;

[0022] The construction module uses the prediction step size p and the frequency slices obtained by the processing module to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients;

[0023] The rank reduction module uses the data matrix M obtained by the construction module f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix

[0024] The first prediction module uses the reduced rank matrix obtained by the reduced rank module Solve for the forward filtered signal vector

[0025] The second prediction module uses the prediction step size p and the frequency slice obtained by the processing module to perform backward prediction filtering to obtain the backward filtering signal vector

[0026] The synthesis module converts the forward filtering signal vector obtained by the first prediction module into The backward filtering signal vector obtained by the second prediction module After synthesis, the denoised signal is expressed in the space-time domain

[0027] The suppression module repeats all frequency slices and expresses the denoised signal in the frequency-space domain. Perform inverse Fourier transform back to space-frequency to get the denoised signal Realize random noise suppression.

[0028] Compared with the prior art, the present invention has at least the following beneficial effects:

[0029] The present invention discloses a seismic random noise suppression method based on reduced-rank FX prediction filtering. The idea of ​​reduced rank is introduced on the basis of FX prediction filtering, and at the same time, measures are taken to construct a low-dimensional covariance matrix to reduce the dimension of the matrix to be reduced in rank, thereby improving the ability of random noise suppression while reducing the amount of calculation.

[0030] Furthermore, since the effective signal in the space-time domain (i.e., the linear phase axis) has predictability in the spatial direction after being converted to the space-frequency domain, and random noise does not have this characteristic, Fourier transform is performed on each signal to obtain the predictability of the effective signal in the space domain.

[0031] Furthermore, frequency slices are extracted along the spatial direction at a fixed frequency to provide basic data for subsequent prediction of effective signals.

[0032] Further, construct the forward prediction filter signal vector S f and the data matrix M f , which can provide known information for the subsequent establishment of a predictability model of the signal; at the same time, the construction of a linear equation group in step 03 can be beneficial for extracting the predictability of the signal to achieve the purpose of suppressing noise.

[0033] Furthermore, we construct the covariance matrix Performing feature decomposition and rank reduction processing is helpful to reduce the amount of calculation and improve the ability to suppress noise.

[0034] Furthermore, for the covariance matrix Characteristic decomposition is beneficial for the subsequent extraction of predictable information related to the signal, and is also beneficial for the subsequent Matrix inversion operation.

[0035] Furthermore, retaining the first q larger eigenvalues ​​and corresponding eigenvectors is beneficial to further improve signal predictability and suppress random noise.

[0036] Furthermore, using the reduced rank matrix Solve for the forward filtering signal vector The predictable part (corresponding to the valid signal) on a single frequency slice can be extracted.

[0037] Further, the frequency slice and prediction step size p obtained in step S2 are used to perform backward prediction filtering to obtain a backward filtered signal vector The effective information of the signal can be predicted in another direction, which is conducive to further improving the noise suppression capability.

[0038] Further, the signal vector predicted in step S5 is The signal vector predicted in step S6 The denoised signal is synthesized in the space-frequency domain. The average of forward filtering and backward filtering can help improve the ability to suppress random noise.

[0039] Furthermore, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal This makes the final denoised signal have the same dimension as the original space-time domain signal.

[0040] It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0041] In summary, the present invention improves on the basis of traditional FX prediction filtering, improves the denoising ability, and the improved algorithm has little difference in computational complexity compared to the FX prediction filtering method. It not only retains the advantages of the FX prediction filtering model being simple and easy to calculate, but can also be expanded to higher-dimensional seismic random noise suppression.

[0042] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the method of the present invention;

[0044] Figure 2 is the noisy tx domain seismic signal;

[0045] Figure 3 Schematic diagram of the results of the fx prediction filtering method, where (a) is the denoising result and (b) is the difference profile;

[0046] Figure 4 Schematic diagram of the Cadzow filtering method results, where (a) is the denoising result and (b) is the difference profile;

[0047] Figure 5 Schematic diagram of the results of the method of the present invention, wherein (a) is the denoising result, and (b) is the difference profile;

[0048] Figure 6 This is a quantitative comparison result diagram of the fx prediction filtering method, the Cadzow filtering method and the method of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0051] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0052] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects are in an "or" relationship.

[0053] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0054] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0055] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0056] The present invention provides a seismic random noise suppression method based on reduced-rank FX prediction filtering, which is improved on the basis of the traditional fx prediction filtering by combining the matrix rank reduction method. First, the signal is transformed from the time-space domain to the frequency-space domain by Fourier transform, and then a reduced-rank model is constructed by utilizing the predictability of the signal to improve the ability to suppress random noise. At the same time, the forward prediction filter and the backward prediction filter are combined to further improve the ability to suppress random noise. The random noise suppression method can be used for various seismic data to provide high signal-to-noise ratio basic data for subsequent links.

[0057] See also Figure 1 The present invention provides a seismic random noise suppression method based on reduced-rank FX prediction filtering, comprising the following steps:

[0058] S1, using Fourier transform to transform the signal from space-time domain seismic data to space-frequency domain;

[0059] Each seismic data in the space-time domain is Fourier transformed and combined into a space-frequency domain signal.

[0060] The result of discrete Fourier transform of the mth signal s[m, n] in the two-dimensional data is:

[0061]

[0062] Where N is the number of time sampling points of the signal, k is the frequency domain index, and S(m, k) is the expression of the signal in the frequency-space domain.

[0063] S2, obtaining a frequency slice according to the spatial-frequency domain signal obtained in step S1;

[0064] Fixed frequency index k, the corresponding frequency slice format is:

[0065] S(1,k),S(2,k),S(3,k),…,S(M,k)

[0066] Where M is the number of channels of 2D seismic data.

[0067] S3, using the frequency slice and prediction step size p obtained in step S2 to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients;

[0068] The forward prediction filter signal vector S f for:

[0069] S f =(S(p+1, k), S( p +2, k), ..., S(M, k)) T

[0070] The data matrix M of the forward prediction filter f for:

[0071]

[0072] And the signal vector S f And the data matrix M f , forming a linear system of equations M f A f =S f , where A f is the forward filter coefficient vector to be solved.

[0073] S4, the covariance matrix formed by the data matrix obtained in step S3 Perform rank reduction processing to obtain the reconstructed matrix

[0074] M f A f =S f Multiply both sides by the transposed matrix of the signal matrix Get the new linear equations Then Perform feature decomposition and rank reduction. First, f Perform eigendecomposition:

[0075]

[0076] Among them, σ i For B f The i-th eigenvalue of and has σ 1 >σ 2 >…>σ p , v i is i The corresponding eigenvector, V is B f The matrix composed of eigenvectors after eigendecomposition.

[0077] Keep the first q larger eigenvalues ​​and the corresponding eigenvectors, and then reconstruct the matrix as follows:

[0078]

[0079] Here, q is the number of eigenvalues ​​retained.

[0080] S5, using the reduced rank matrix obtained in step S4 Solve for the forward filtering signal vector

[0081] First solve for the forward filter coefficient vector

[0082]

[0083] Then use the forward filter coefficient vector Right multiply the original data matrix M f , that is, predict the signal vector as follows:

[0084]

[0085] in,

[0086] S6, using the frequency slice and prediction step size p obtained in step S2 to perform backward prediction filtering to obtain a backward filtering signal vector

[0087] The frequency slices obtained in step S2 are used to construct the backward filtering signal vector S b for:

[0088] S b =(S(1,k),S(2,k),...,S(Mp,k)) T

[0089] The data matrix M for backward prediction filtering b for:

[0090]

[0091] And the signal vector S b And the data matrix M b , forming a linear system of equations M b A b =S b , A b is the unknown backward filter coefficient vector.

[0092] Following the same steps as steps S4 and S5, the signal vector Its form is:

[0093]

[0094] Where M is the number of channels of 2D seismic data, and k is the frequency domain index.

[0095] S7, the signal vector predicted in step S5 The signal vector predicted in step S6 The denoised signal is synthesized in the space-frequency domain.

[0096] The final denoised signal is expressed in the space-frequency domain The signal vector predicted by step S5 The signal vector predicted in step S6 To average:

[0097]

[0098] in, is the backward filtered signal vector, is the forward filtering signal vector, m is the number of signals, and M is the number of channels of 2D seismic data.

[0099] S8, repeating steps S2 to S7, after processing all frequency slices, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal

[0100]

[0101] Among them, N is the number of time sampling points of the signal, k is the frequency domain index, and n is the time domain index.

[0102] The present invention performs Fourier transform on each signal in the space-time signal, and then extracts the same frequency of each time signal to form a frequency slice, and performs the processing of steps S2-S7 respectively. After each frequency slice is filtered, the filtering results of each frequency slice are combined together to obtain the final denoised signal in the space-frequency domain, and each signal is subjected to inverse Fourier transform to return to the space-time domain, thus obtaining the space-time signal after noise suppression. Since the effective signal in the space-frequency domain is predictable in the spatial direction and the noise is not predictable in the spatial direction, by optimizing the solution of the linear equation group M f A f =S f and M b A b =S bThe predictable part can be extracted and the unpredictable part can be ignored, and random noise suppression can be finally achieved. In addition, the seismic random noise suppression method based on the reduced-rank FX prediction filter mentioned in the present invention solves the linear equation system M f A f =S f and M b A b =S b In the process and Optimization can further suppress random noise.

[0103] In another embodiment of the present invention, a seismic random noise suppression system based on reduced-rank FX prediction filtering is provided. The system can be used to implement the above-mentioned seismic random noise suppression method based on reduced-rank FX prediction filtering. Specifically, the seismic random noise suppression system based on reduced-rank FX prediction filtering includes a transformation module, a processing module, a construction module, a reduced-rank module, a first prediction module, a second prediction module, an integrated module and a suppression module.

[0104] Among them, the transformation module uses Fourier transform to transform the signal from the space-time domain seismic data to the space-frequency domain;

[0105] A processing module obtains a frequency slice according to the spatial-frequency domain signal obtained by the transformation module;

[0106] The construction module uses the prediction step size p and the frequency slices obtained by the processing module to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients;

[0107] The rank reduction module uses the data matrix M obtained by the construction module f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix

[0108] The first prediction module uses the reduced rank matrix obtained by the reduced rank module Solve for the forward filtered signal vector

[0109] The second prediction module uses the prediction step size p and the frequency slice obtained by the processing module to perform backward prediction filtering to obtain the backward filtering signal vector

[0110] The synthesis module converts the forward filtering signal vector obtained by the first prediction module into The backward filtering signal vector obtained by the second prediction module After synthesis, the denoised signal is expressed in the space-frequency domain

[0111] The suppression module repeats all frequency slices and expresses the denoised signal in the frequency-space domain. Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

[0112] In another embodiment of the present invention, a terminal device is provided, the terminal device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the seismic random noise suppression method based on the reduced-rank FX prediction filter, including:

[0113] The signal is transformed from the space-time domain seismic data to the space-frequency domain using Fourier transform; a frequency slice is obtained based on the signal in the space-frequency domain; the signal vector S of the forward prediction filter is constructed using the prediction step size p and the frequency slice. f , data matrix M f And solve the linear equations for predicting filter coefficients; use the data matrix Mf to construct the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix Using reduced rank matrix Solve for the forward filtered signal vector The backward prediction filtering is performed using the prediction step size p and frequency slice to obtain the backward filtering signal vector The forward filtered signal vector and the back-filtered signal vector After synthesis, the denoised signal is expressed in the space-frequency domain After processing all frequency slices, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

[0114] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory, or a non-volatile memory (Non-Volatile Memory), such as at least one disk memory.

[0115] The processor may load and execute one or more instructions stored in a computer-readable storage medium to implement the corresponding steps of the seismic random noise suppression method based on reduced-rank FX prediction filtering in the above embodiment; the processor may load and execute the following steps:

[0116] The signal is transformed from the space-time domain seismic data to the space-frequency domain using Fourier transform; a frequency slice is obtained based on the signal in the space-frequency domain; the signal vector S of the forward prediction filter is constructed using the prediction step size p and the frequency slice. f , data matrix M f And solve the linear equations for the prediction filter coefficients; using the data matrix M f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix Using reduced rank matrix Solve for the forward filtered signal vector The backward prediction filtering is performed using the prediction step size p and frequency slice to obtain the backward filtering signal vector The forward filtered signal vector and the back-filtered signal vector After synthesis, the denoised signal is expressed in the space-frequency domain After processing all frequency slices, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

[0117] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0118] See also Figure 2 , is a noisy spatial-temporal seismic signal (signal-to-noise ratio is -1dB), the horizontal axis is the number of seismic traces, and the vertical axis is the number of time points. The sampling interval is 2ms, with a total of 2000 sampling points, which is the noisy data used for noise suppression.

[0119] See also Figure 3 , Figure 4 as well as Figure 5 , which are the noise suppression results of the three methods. Cadzow filtering and the method of the present invention are both examples of retaining two eigenvalues ​​during rank reduction.

[0120] See also Figure 3 , which is the denoising result and difference profile of the FX prediction filtering method, is processed by FX prediction filtering Figure 2 The result obtained after the random noise is basically removed, and there is basically no signal loss.

[0121] See also Figure 4 , which is the denoising result and difference profile of the Cadzow filter method, is processed by Cadzow filter Figure 2 The result obtained after the random noise is basically removed, and there is basically no signal loss.

[0122] See also Figure 5 , which is the denoising result and difference profile of the method of the present invention, is processed by the method of the present invention Figure 2 The result obtained after . It can effectively remove random noise without signal loss. Figure 3 , Figure 4 as well as Figure 5 The denoising result of the method of the present invention has a higher signal-to-noise ratio and a cleaner profile, which is sufficient to illustrate the effectiveness of the method of the present invention.

[0123] See also Figure 6, which is a quantitative comparison result diagram of the three methods. Since the synthetic noisy data contains two events with different inclination angles, two eigenvalues ​​are retained during Cadzow filtering and rank reduction by the method of the present invention. Figure 6 The horizontal axis is the filter length (ie, the prediction step length), and the vertical axis is the signal-to-noise ratio of the denoised signal; when the optimal solution is taken, the method of the present invention has at least a 2dB improvement in signal-to-noise ratio compared with the other two methods.

[0124] In summary, the seismic random noise suppression method and system based on the reduced-rank FX prediction filter of the present invention introduces the idea of ​​reduced-rank on the basis of FX prediction filtering, thereby improving the denoising effect. The FX prediction filter model is simple and has a small amount of computation, which is retained through the processing of reduced-rank.

[0125] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0126] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0128] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0131] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0132] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0135] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A seismic random noise suppression method based on reduced-rank FX prediction filtering, characterized in that: The following steps are involved: S1, using Fourier transform to transform the signal from space-time domain seismic data to space-frequency domain; S2, obtaining a frequency slice according to the spatial-frequency domain signal obtained in step S1; S3, using the prediction step size p and the frequency slice obtained in step S2 to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients; S4, using the data matrix M obtained in step S3 f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix S5. Using the reduced rank matrix obtained in step S4 Solve for the forward filtered signal vector S6, using the prediction step size p and the frequency slice obtained in step S2 to perform backward prediction filtering to obtain a backward filtering signal vector S7, the forward filtering signal vector obtained in step S5 The back-filtered signal vector obtained in step S6 After synthesis, the denoised signal is expressed in the space-frequency domain S8, repeating steps S2 to S7, after processing all frequency slices, the denoised signal is expressed in the frequency-space domain Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

2. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S1, the result of performing discrete Fourier transform on the m-th signal s[m,n in the two-dimensional data is: Among them, N is the number of time sampling points of the signal, k is the frequency domain index, n is the time domain index, and S(m,k) is the expression of the signal in the frequency-space domain.

3. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S2, the frequency domain index k is fixed, and the corresponding frequency slice is: S(1,k), S(2,k), S(3,k),…, S(M,k) Where M is the number of channels of 2D seismic data, and k is the frequency domain index.

4. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S3, the forward prediction filter signal vector S is constructed. f for: S f =(S(p+1,k),S(p+2,k),…,S(M,k)) T The data matrix M of the forward prediction filter f for: The linear equations are: M f A f =S f Among them, A f is the forward filter coefficient vector to be solved, k is the frequency domain index, and M is the number of channels of two-dimensional seismic data.

5. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S4, the matrix is ​​reconstructed as follows: Among them, q is the number of eigenvalues ​​retained, σ i For B f The i-th eigenvalue of i is the i-th eigenvalue σ i The corresponding feature vector.

6. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S5, the forward filtering signal vector for: Among them, M f is the original data matrix, is the forward filter coefficient vector obtained using the least squares method.

7. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S6, the back-filtered signal vector for: Where M is the number of channels of 2D seismic data, and k is the frequency domain index.

8. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S7, the denoised signal is expressed in the space-frequency domain as for: in, is the backward filtered signal vector, is the forward filtering signal vector, m is the number of signals, and M is the number of channels of 2D seismic data.

9. The seismic random noise suppression method based on reduced-rank FX prediction filtering according to claim 1 is characterized in that: In step S8, the denoised signal for: Among them, N is the number of time sampling points of the signal, k is the frequency domain index, j is the unit of the imaginary part in complex number operations, and n is the time domain index.

10. A seismic random noise suppression system based on reduced-rank FX predictive filtering, characterized in that: include: A transform module, which transforms the signal from the space-time domain seismic data to the space-frequency domain using Fourier transform; A processing module obtains a frequency slice according to the spatial-frequency domain signal obtained by the transformation module; The construction module uses the prediction step size p and the frequency slices obtained by the processing module to construct the signal vector S of the forward prediction filter f , data matrix M f and solving a system of linear equations for prediction filter coefficients; The rank reduction module uses the data matrix M obtained by the construction module f Constructing the covariance matrix Then the rank reduction process is performed to obtain the reconstructed matrix The first prediction module uses the reduced rank matrix obtained by the reduced rank module Solve for the forward filtered signal vector The second prediction module uses the prediction step size p and the frequency slice obtained by the processing module to perform backward prediction filtering to obtain the backward filtering signal vector The synthesis module converts the forward filtering signal vector obtained by the first prediction module into The backward filtering signal vector obtained by the second prediction module After synthesis, the denoised signal is expressed in the space-frequency domain The suppression module repeats all frequency slices and expresses the denoised signal in the frequency-space domain. Perform inverse Fourier transform back to the space-time domain to obtain the denoised signal Realize random noise suppression.

Images