Imaging domain least squares migration method, device, electronic equipment and medium

Abstract

The application discloses an imaging domain least square migration method and device, electronic equipment and medium. The method can comprise: acquiring a global space-varying scattering function, interpolating and reconstructing the global space-varying scattering function to generate a plurality of scattering function data volumes; calculating the spatial deconvolution of the plurality of scattering function data volumes according to a depth domain imaging volume through an MPI parallel strategy to obtain corresponding least square migration images; and reducing the least square migration images corresponding to the plurality of scattering function data volumes to obtain a final imaging result. The application encrypts the point spread function through the interpolation and reconstruction strategy, and further deconvolves the reconstruction to suppress discontinuous noise through multiple superimpositions to obtain higher imaging accuracy.

Description

Technical Field

[0001] This invention relates to the field of seismic wave imaging technology, and more specifically, to an imaging domain least squares migration method, apparatus, electronic device, and medium. Background Technology

[0002] Least squares reverse time migration (LSTM) is currently the most accurate imaging method. However, data-domain LSTM theoretically requires multiple inverse migration / migration iterations to match the observed seismic data. Therefore, factors such as unknown seismic wavelets, amplitude mismatch, and insufficient accuracy of the initial model have a significant impact on the imaging results. Furthermore, the inverse migration / migration process requires extensive 3D seismic wavefield forward modeling calculations, and data convergence is slow, all of which lead to significant problems in its practical application with real-world data.

[0003] Image-domain least-squares migration can directly process conventional migration results by estimating the inverse Hessian operator, addressing issues such as insufficient image illumination and low resolution. The point spread function (PSF), a row element of the Hessian algorithm, physically describes the impact of a single scattering point on underground space and is a commonly used Hessian approximation operator. Because the PSF has spatial focusing properties, conventional least-squares migration algorithms typically divide the model into regions. This approach effectively increases the sampling density of scattering points and improves imaging accuracy. However, in poorly illuminated areas, it can cause abrupt boundary changes at the edges of these regions, resulting in image discontinuities.

[0004] Therefore, given the severe discontinuity noise generated by the current region-block calculation method for least squares reverse time migration in the imaging domain, it is necessary to develop an imaging domain least squares migration method, device, electronic device, and medium based on interpolation reconstruction. This method refines the point spread function based on the interpolation reconstruction idea, overcoming the spatial sampling constraints imposed by the point spread function. Simultaneously, it performs spatial deconvolution calculation based on a CPU / GPU heterogeneous parallel strategy, achieving accurate and efficient calculation of least squares migration, obtaining high-quality, high-resolution imaging results, and accelerating the practical application of imaging domain least squares reverse time migration technology.

[0005] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] This invention proposes an imaging domain least squares offset method, device, electronic device, and medium, which can encrypt the point spread function through an interpolation reconstruction strategy. Furthermore, the reconstructed deconvolution can suppress discontinuity noise through multiple superpositions to obtain higher imaging accuracy.

[0007] In a first aspect, embodiments of this disclosure provide an imaging domain least-squares offset method, including:

[0008] Obtain the global spatially variable scattering function, and reconstruct the global spatially variable scattering function by interpolation to generate multiple scattering function data volumes;

[0009] Based on the depth domain imaging volume, the spatial deconvolution of multiple scattering function data volumes is calculated using the MPI parallel strategy to obtain the corresponding least squares offset image;

[0010] The least-squares offset images corresponding to multiple scattering function data volumes are reduced to obtain the final imaging results.

[0011] Preferably, obtaining the global space-varying scattering function includes:

[0012] Establish a scattering point model;

[0013] Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data.

[0014] The global spatially variable scattering function is obtained by performing a reverse time migration based on the velocity model and the inverse migration data.

[0015] Preferably, the two-way wave equation and its inverse migration equation are numerically calculated using high-order finite difference of acoustic waves and optimal matching layer to obtain inverse migration data.

[0016] Preferably, calculating the spatial deconvolution of multiple scattering function data volumes to obtain the corresponding least-squares migration image includes:

[0017] The depth domain imaging volume and the scattering function data volume are respectively converted to the wavenumber domain;

[0018] Calculate the least squares shift image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume;

[0019] The wavenumber domain least squares offset image is transformed into the spatial domain to obtain the corresponding least squares offset image.

[0020] Preferably, the wavenumber domain least squares migration image is calculated using formula (1):

[0021]

[0022] Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

[0023] Preferably, the depth domain imaging volume and the scattering function data volume are converted to the wavenumber domain by a three-dimensional fast Fourier transform.

[0024] Preferably, the wavenumber domain least squares offset image is transformed to the spatial domain by a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares offset image.

[0025] As one specific implementation of this disclosure,

[0026] Secondly, embodiments of this disclosure also provide an imaging domain least squares offset device, comprising:

[0027] The global spatial variation scattering module acquires the global spatial variation scattering function, performs interpolation and reconstruction on the global spatial variation scattering function, and generates multiple scattering function data volumes;

[0028] The deconvolution module calculates the spatial deconvolution of multiple scattering function data volumes based on the depth domain imaging volume using the MPI parallel strategy, and obtains the corresponding least-squares offset image.

[0029] The reduction module performs reduction on the least-squares offset images corresponding to multiple scattering function data volumes to obtain the final imaging results.

[0030] Preferably, obtaining the global space-varying scattering function includes:

[0031] Establish a scattering point model;

[0032] Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data.

[0033] The global spatially variable scattering function is obtained by performing a reverse time migration based on the velocity model and the inverse migration data.

[0034] Preferably, the two-way wave equation and its inverse migration equation are numerically calculated using high-order finite difference of acoustic waves and optimal matching layer to obtain inverse migration data.

[0035] Preferably, calculating the spatial deconvolution of multiple scattering function data volumes to obtain the corresponding least-squares migration image includes:

[0036] The depth domain imaging volume and the scattering function data volume are respectively converted to the wavenumber domain;

[0037] Calculate the least-squares shifted image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume.

[0038] The wavenumber domain least squares offset image is transformed into the spatial domain to obtain the corresponding least squares offset image.

[0039] Preferably, the wavenumber domain least squares migration image is calculated using formula (1):

[0040]

[0041] Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

[0042] Preferably, the depth domain imaging volume and the scattering function data volume are converted to the wavenumber domain by a three-dimensional fast Fourier transform.

[0043] Preferably, the wavenumber domain least squares offset image is transformed to the spatial domain by a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares offset image.

[0044] Thirdly, embodiments of this disclosure also provide an electronic device, the electronic device comprising:

[0045] Memory, which stores executable instructions;

[0046] A processor that executes the executable instructions in the memory to implement the imaging domain least squares offset method.

[0047] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the imaging domain least squares offset method.

[0048] Its beneficial effects are as follows: This invention uses interpolation reconstruction technology to encrypt the point spread function, breaks through the sampling constraint of the point spread function by the wave field interference effect, and the encryption superposition can significantly suppress imaging noise, solve the discontinuity problem that occurs when performing region block deconvolution in the current imaging domain least squares migration algorithm, obtain high-quality least squares migration images, and thus promote the practical application of imaging domain least squares reverse time migration.

[0049] The methods and apparatus of the present invention have other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0050] The above and other objects, features and advantages of the present invention will become more apparent from the more detailed description of exemplary embodiments of the invention in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same parts.

[0051] Figure 1 A flowchart illustrating the steps of an imaging domain least-squares offset method according to an embodiment of the present invention is shown.

[0052] Figure 2 A two-dimensional schematic diagram of an Overthrust model according to an embodiment of the present invention is shown.

[0053] Figure 3 A two-dimensional partial schematic diagram of a scattering point model according to an embodiment of the present invention is shown.

[0054] Figure 4 It shows that according to Figure 2 and Figure 3 A schematic diagram of the globally space-varying scattering function generated from the data.

[0055] Figure 5 A schematic diagram of a depth domain imaging body according to an embodiment of the present invention is shown.

[0056] Figure 6 It shows the Figure 4 , Figure 5 A schematic diagram of the imaging results obtained by performing conventional least-squares reverse time migration on the data.

[0057] Figure 7 It shows the Figure 4 , Figure 5 A schematic diagram of the imaging results obtained by interpolation-based least-squares reverse time migration of the imaging domain.

[0058] Figure 8 A block diagram of an imaging domain least squares offset device according to an embodiment of the present invention is shown.

[0059] Explanation of reference numerals in the attached figures:

[0060] 201. Global spatially variable scattering module; 202. Deconvolution module; 203. Reduction module. Detailed Implementation

[0061] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention are described below, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0062] This invention provides an imaging domain least squares offset method, comprising:

[0063] Obtain the global spatially varying scattering function, interpolate and reconstruct the global spatially varying scattering function to generate multiple scattering function data volumes;

[0064] Based on the depth domain imaging volume, the spatial deconvolution of multiple scattering function data volumes is calculated using the MPI parallel strategy to obtain the corresponding least squares offset image;

[0065] The least-squares offset images corresponding to multiple scattering function data volumes are reduced to obtain the final imaging results.

[0066] In one example, obtaining the global space-varying scattering function includes:

[0067] Establish a scattering point model;

[0068] Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data.

[0069] The global spatially variable scattering function is obtained by performing inverse time migration based on the velocity model and inverse migration data.

[0070] In one example, the two-way wave equation and its inverse migration equation are numerically calculated using high-order finite difference of acoustic waves and optimal matching layer to obtain inverse migration data.

[0071] In one example, the spatial deconvolution of multiple scattering function data volumes is calculated to obtain the corresponding least-squares offset image, including:

[0072] Transform the depth domain imaging volume and the scattering function data volume to the wavenumber domain, respectively;

[0073] Calculate the least-squares shifted image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume.

[0074] The wavenumber domain least squares migration image is transformed into the spatial domain to obtain the corresponding least squares migration image.

[0075] In one example, the wavenumber domain least-squares migration image is calculated using formula (1):

[0076]

[0077] Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

[0078] In one example, the depth domain imaging volume and the scattering function data volume are transformed to the wavenumber domain using a three-dimensional fast Fourier transform.

[0079] In one example, the wavenumber domain least squares migration image is transformed to the spatial domain using a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares migration image.

[0080] Specifically, the reading speed model and main frequency are used; the sampling interval and initial position are determined by formula (2):

[0081]

[0082] Where r is the sampling interval, w is the weight (typically 1.5), and v max The maximum speed is given by f, which is the dominant frequency. The initial position is generally taken as... Establish a scattering point model;

[0083] Three-dimensional seismic wavefield simulations were performed using the two-way wave equations and their inverse migration equations to obtain inverse migration data. The two-way wave equations are as follows:

[0084]

[0085] Where v is the velocity, p represents the seismic wave field, and s represents the source function. For the Laplace operator, t represents time, and x, y, and z represent the three spatial directions; the inverse offset equation is:

[0086]

[0087] Where v is the velocity, p represents the seismic wave field, and s represents the source function. For the Laplace operator, δ represents the perturbation field.

[0088] The global spatially variable scattering function is obtained by performing inverse time migration based on the velocity model and inverse migration data.

[0089] Based on the encryption multiplier r in the x, y, and z directions x r y r z Interpolation reconstruction is performed on the global space-varying scattering function to generate a total N = r x ·r y ·r z A scattering function data volume. Taking the x-direction as an example, the current position is x. n Using x n -2,x n -1,x n x n +1, x n Calculate the global space-varying scattering function at position +2. Global spatially varying scattering function at location, i = 1, ..., r x This ultimately results in a total N = rx ·r y ·r z A scattering function data volume.

[0090] Based on the depth domain imaging volume, the spatial deconvolution of multiple scattering function data volumes is calculated using the MPI parallel strategy to obtain the corresponding least squares offset image. The depth domain imaging volume and the scattering function data volume are transformed to the wavenumber domain by the three-dimensional fast Fourier transform. Based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume, the wavenumber domain least squares offset image is calculated using formula (1). The wavenumber domain least squares offset image is transformed to the spatial domain by the three-dimensional fast Fourier inverse transform to obtain the corresponding least squares offset image.

[0091] When calculating the deconvolution of the spread function at any point, the model is divided into regions, and the spatial deconvolution within the region is accelerated by using a GPU parallel algorithm. All regions are superimposed to obtain the least squares offset image corresponding to the scatter function data volume of a single point.

[0092] The least-squares offset images corresponding to multiple scattering function data volumes are reduced to obtain the final imaging results.

[0093] The present invention also provides an imaging domain least squares offset device, comprising:

[0094] The global spatial variation scattering module obtains the global spatial variation scattering function, performs interpolation and reconstruction on the global spatial variation scattering function, and generates multiple scattering function data volumes;

[0095] The deconvolution module calculates the spatial deconvolution of multiple scattering function data volumes based on the depth domain imaging volume using the MPI parallel strategy, and obtains the corresponding least-squares offset image.

[0096] The reduction module performs reduction on the least-squares offset images corresponding to multiple scattering function data volumes to obtain the final imaging results.

[0097] In one example, obtaining the global space-varying scattering function includes:

[0098] Establish a scattering point model;

[0099] Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data.

[0100] The global spatially variable scattering function is obtained by performing inverse time migration based on the velocity model and inverse migration data.

[0101] In one example, the two-way wave equation and its inverse migration equation are numerically calculated using high-order finite difference of acoustic waves and optimal matching layer to obtain inverse migration data.

[0102] In one example, the spatial deconvolution of multiple scattering function data volumes is calculated to obtain the corresponding least-squares offset image, including:

[0103] Transform the depth domain imaging volume and the scattering function data volume to the wavenumber domain, respectively;

[0104] Calculate the least-squares shifted image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume.

[0105] The wavenumber domain least squares migration image is transformed into the spatial domain to obtain the corresponding least squares migration image.

[0106] In one example, the wavenumber domain least-squares migration image is calculated using formula (1):

[0107]

[0108] Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

[0109] In one example, the depth domain imaging volume and the scattering function data volume are transformed to the wavenumber domain using a three-dimensional fast Fourier transform.

[0110] In one example, the wavenumber domain least squares migration image is transformed to the spatial domain using a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares migration image.

[0111] Specifically, the reading speed model and main frequency are used; the sampling interval and initial position are determined by formula (2), and the initial position is generally taken as... Establish a scattering point model;

[0112] Three-dimensional seismic wavefield simulation was performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data. The two-way wave equation is Equation (3), and the inverse migration equation is Equation (4).

[0113] The global spatially variable scattering function is obtained by performing inverse time migration based on the velocity model and inverse migration data.

[0114] Based on the encryption multiplier r in the x, y, and z directions x r y r z Interpolation reconstruction is performed on the global space-varying scattering function to generate a total N = r x ·r y ·r z A scattering function data volume. Taking the x-direction as an example, the current position is x. n Using x n -2,x n-1,x n x n +1, x n Calculate the global space-varying scattering function at position +2. Global spatially varying scattering function at location, i = 1, ..., r x This ultimately results in a total N = r x ·r y ·r z A scattering function data volume.

[0115] Based on the depth domain imaging volume, the spatial deconvolution of multiple scattering function data volumes is calculated using the MPI parallel strategy to obtain the corresponding least squares offset image. The depth domain imaging volume and the scattering function data volume are transformed to the wavenumber domain by the three-dimensional fast Fourier transform. Based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume, the wavenumber domain least squares offset image is calculated using formula (1). The wavenumber domain least squares offset image is transformed to the spatial domain by the three-dimensional fast Fourier inverse transform to obtain the corresponding least squares offset image.

[0116] When calculating the deconvolution of the spread function at any point, the model is divided into regions, and the spatial deconvolution within the region is accelerated by using a GPU parallel algorithm. All regions are superimposed to obtain the least squares offset image corresponding to the scatter function data volume of a single point.

[0117] The least-squares offset images corresponding to multiple scattering function data volumes are reduced to obtain the final imaging results.

[0118] The present invention also provides an electronic device, comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the above-described imaging domain least squares offset method.

[0119] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described imaging domain least squares offset method.

[0120] To facilitate understanding of the solutions and effects of the embodiments of the present invention, four specific application examples are given below. Those skilled in the art should understand that these examples are merely for the purpose of understanding the present invention, and any specific details therein are not intended to limit the present invention in any way.

[0121] Example 1

[0122] Figure 1 A flowchart illustrating the steps of an imaging domain least-squares offset method according to an embodiment of the present invention is shown.

[0123] like Figure 1As shown, the imaging domain least squares migration method includes: Step 101, obtaining the global spatially varying scattering function, interpolating and reconstructing the global spatially varying scattering function to generate multiple scattering function data volumes; Step 102, based on the depth domain imaging volume, calculating the spatial deconvolution of the multiple scattering function data volumes using an MPI parallel strategy to obtain the corresponding least squares migration image; Step 103, reducing the least squares migration image corresponding to the multiple scattering function data volumes to obtain the final imaging result.

[0124] Figure 2 A two-dimensional schematic diagram of an Overthrust model according to an embodiment of the present invention is shown.

[0125] Figure 3 A two-dimensional partial schematic diagram of a scattering point model according to an embodiment of the present invention is shown.

[0126] Figure 4 It shows that according to Figure 2 and Figure 3 A schematic diagram of the globally space-varying scattering function generated from the data.

[0127] Input as follows Figure 2 The velocity model shown has a velocity v and a main frequency f of 30Hz, and as shown in the figure. Figure 3 The scattering point model m is shown; using the velocity model and the scattering point model m, wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data d. dem Using the velocity model v and the inverse offset data d dem Perform a reverse time offset to obtain, as shown Figure 4 The global space-varying scattering function is shown.

[0128] Based on the encryption multiplier r in the x, y, z directions x r y r z Interpolation reconstruction is performed on the global space-varying scattering function to generate a total N = r x ·r y ·r z A scattering function data volume.

[0129] Figure 5 A schematic diagram of a depth domain imaging body according to an embodiment of the present invention is shown.

[0130] Figure 6 It shows the Figure 4 , Figure 5 A schematic diagram of the imaging results obtained by performing conventional least-squares reverse time migration on the data.

[0131] Figure 7 It shows the Figure 4 , Figure 5A schematic diagram of the imaging results obtained by interpolation-based least-squares reverse time migration of the imaging domain.

[0132] Input as follows Figure 5 The depth-domain imaging volume shown employs an MPI parallel strategy to simultaneously compute the spatial deconvolution of N scattering function data volumes. When computing the deconvolution of any scattering function data volume, the model is divided into regions, and a GPU parallel algorithm is used to accelerate the spatial deconvolution within each region. All regions are then superimposed to obtain a least-squares offset image under a single-point spread function, as shown below. Figure 6 As shown; the least-squares migration images corresponding to the N scattering function data volumes are reduced to obtain the final imaging result, as shown. Figure 7 As shown.

[0133] Conventional imaging domain least squares migration algorithms employ a region-blocking strategy, which effectively addresses issues such as amplitude imbalance, insufficient deep illumination, and low resolution in RTM results. Figure 6 As shown, however, various types of noise appear significantly at depth. By employing an interpolation reconstruction strategy to refine the point spread function, the migration results are significantly improved, discontinuities are noticeably reduced, and noise is decreased, as shown in the figure. Figure 7 As shown.

[0134] Example 2

[0135] Figure 8 A block diagram of an imaging domain least squares offset device according to an embodiment of the present invention is shown.

[0136] like Figure 8 As shown, the imaging domain least squares offset device includes:

[0137] The global spatial variation scattering module 201 acquires the global spatial variation scattering function, performs interpolation and reconstruction on the global spatial variation scattering function, and generates multiple scattering function data volumes;

[0138] The deconvolution module 202 calculates the spatial deconvolution of multiple scattering function data volumes based on the depth domain imaging volume using the MPI parallel strategy to obtain the corresponding least squares offset image.

[0139] The reduction module 203 reduces the least-squares offset image corresponding to multiple scattering function data volumes to obtain the final imaging result.

[0140] As an optional approach, obtaining the global space-varying scattering function includes:

[0141] Establish a scattering point model;

[0142] Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data.

[0143] The global spatially variable scattering function is obtained by performing inverse time migration based on the velocity model and inverse migration data.

[0144] As an alternative, the two-way wave equation and its inverse migration equation are numerically calculated using high-order finite difference of acoustic waves and optimal matching layer to obtain inverse migration data.

[0145] As an optional approach, the spatial deconvolution of multiple scattering function data volumes can be calculated to obtain the corresponding least-squares migration image, including:

[0146] Transform the depth domain imaging volume and the scattering function data volume to the wavenumber domain, respectively;

[0147] Calculate the least squares shift image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume;

[0148] The wavenumber domain least squares migration image is transformed into the spatial domain to obtain the corresponding least squares migration image.

[0149] As an alternative, the wavenumber domain least squares migration image is calculated using formula (1):

[0150]

[0151] Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

[0152] As an alternative, the depth domain imaging volume and the scattering function data volume can be converted to the wavenumber domain by using a three-dimensional fast Fourier transform.

[0153] As an alternative, the wavenumber domain least squares migration image can be transformed to the spatial domain using a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares migration image.

[0154] Example 3

[0155] This disclosure provides an electronic device comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the aforementioned imaging domain least squares offset method.

[0156] An electronic device according to an embodiment of the present disclosure includes a memory and a processor.

[0157] This memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0158] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory.

[0159] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0160] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0161] Example 4

[0162] This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the imaging domain least squares offset method.

[0163] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods described in the foregoing embodiments of the present disclosure are performed.

[0164] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0165] Those skilled in the art should understand that the above description of the embodiments of the present invention is only intended to illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any of the examples given.

[0166] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A least-squares migration method for the imaging domain, characterized in that, include: Obtain the global spatially variable scattering function, and reconstruct the global spatially variable scattering function by interpolation to generate multiple scattering function data volumes; Based on the depth domain imaging volume, the spatial deconvolution of multiple scattering function data volumes is calculated using the MPI parallel strategy to obtain the corresponding least squares offset image; The least-squares migration image corresponding to multiple scattering function data volumes is reduced to obtain the final imaging result; Obtaining the global space-varying scattering function includes: Establish a scattering point model; Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data. The global spatially variable scattering function is obtained by performing a reverse time migration based on the velocity model and the inverse migration data.

2. The imaging domain least squares offset method according to claim 1, wherein, Numerical calculations of the two-way wave equation and its inverse migration equation are performed using high-order finite difference acoustic wave finite difference and optimal matching layer to obtain inverse migration data.

3. The imaging domain least squares offset method according to claim 1, wherein, Calculate the spatial deconvolution of multiple scattering function data volumes to obtain the corresponding least-squares migration images, including: The depth domain imaging volume and the scattering function data volume are respectively converted to the wavenumber domain; Calculate the least-squares shifted image in the wavenumber domain based on the wavenumber domain imaging volume and the wavenumber domain scattering function data volume. The wavenumber domain least squares offset image is transformed into the spatial domain to obtain the corresponding least squares offset image.

4. The imaging domain least squares offset method according to claim 3, wherein, The formula for calculating the wavenumber domain least squares migration image is: Where α is the regularization factor. This is a least-squares migration image in the wavenumber domain. For wavenumber domain imaging volume, This is the data volume of the wavenumber domain scattering function.

5. The imaging domain least squares offset method according to claim 3, wherein, The depth domain imaging volume and the scattering function data volume are respectively converted to the wavenumber domain using a three-dimensional fast Fourier transform.

6. The imaging domain least squares offset method according to claim 3, wherein, The wavenumber domain least squares offset image is transformed to the spatial domain by using a three-dimensional fast Fourier inverse transform to obtain the corresponding least squares offset image.

7. An imaging domain least-squares offset device, characterized in that, include: The global spatial variation scattering module acquires the global spatial variation scattering function, performs interpolation and reconstruction on the global spatial variation scattering function, and generates multiple scattering function data volumes; The deconvolution module calculates the spatial deconvolution of multiple scattering function data volumes based on the depth domain imaging volume using the MPI parallel strategy, and obtains the corresponding least-squares offset image. The reduction module reduces the least-squares offset image corresponding to multiple scattering function data volumes to obtain the final imaging result. Obtaining the global space-varying scattering function includes: Establish a scattering point model; Based on the velocity model and the real-time scattering point model, a three-dimensional seismic wavefield simulation is performed on the two-way wave equation and its inverse migration equation to obtain inverse migration data. The global spatially variable scattering function is obtained by performing a reverse time migration based on the velocity model and the inverse migration data.

8. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement the imaging domain least-squares offset method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the imaging domain least squares offset method as described in any one of claims 1-6.

Images