Seismic Data Denoising Method Based on Quaternion Generalized S-Transform

Abstract

This invention relates to a seismic data denoising method based on the quaternion generalized S-transform, relating to the field of seismic exploration. The seismic data denoising method based on the quaternion generalized S-transform includes: acquiring multi-component seismic data; converting the multi-component seismic data into quaternion functions according to a preset quaternion relationship; performing a quaternion generalized S-transform on the quaternion functions based on the preset quaternion generalized S-transform to obtain a quaternion generalized S-transform spectrum; filtering the quaternion generalized S-transform spectrum to obtain a filtering result; and performing an inverse quaternion generalized S-transform on the filtering result to convert the filtering result into denoised multi-component seismic data. By converting the multi-component seismic data into quaternion functions, it is possible to perform a quaternion generalized S-transform on the multi-component seismic data, effectively utilizing the correlation information between the components of the multi-component seismic data while performing denoising, thus fully preserving the vector wavefield characteristics of the multi-component seismic data.

Description

Technical Field

[0001] This application relates to the field of seismic exploration, specifically to a seismic data filtering method based on quaternion generalized S-transform, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Multi-component seismic data is used to simultaneously record the motion of particles in both vertical and horizontal directions, providing a complete record of the seismic wave vector field. However, the signal-to-noise ratio of acquired multi-component seismic data is relatively low, and denoising / noise reduction processing is usually required before it can be fully utilized.

[0003] The generalized S-transform is a method for analyzing non-stationary signals and removing time-varying noise. Existing technologies utilize the generalized S-transform to process multi-component seismic data. However, current methods for denoising multi-component seismic data using the generalized S-transform typically only separate and process the different components individually. This approach fails to leverage the correlation between the components and results in the loss of vector wavefield characteristics in the multi-component seismic data. Summary of the Invention

[0004] In view of this, the present invention aims to provide a seismic data denoising method based on quaternion generalized S-transform, an electronic device, and a computer-readable storage medium to process each component of multi-component seismic data simultaneously, make full use of the inter-correlation information between the components of multi-component seismic data, better suppress noise, and preserve the vector wavefield characteristics of multi-component seismic data.

[0005] In a first aspect, this application provides a seismic data denoising method based on quaternion generalized S-transform, comprising: acquiring multi-component seismic data; converting the multi-component seismic data into a quaternion function according to a preset quaternion relation; performing a quaternion generalized S-transform on the quaternion function based on the preset quaternion generalized S-transform to obtain a quaternion generalized S-transform spectrum; filtering the quaternion generalized S-transform spectrum to obtain a filtering result; and performing an inverse quaternion generalized S-transform on the filtering result to convert the filtering result into denoised multi-component seismic data.

[0006] In this embodiment, the multi-component seismic data is converted into a quaternion function according to a preset quaternion relationship. This allows multiple components of the multi-component seismic data to be transformed into a single quaternion function. Consequently, a generalized quaternion S-transform can be directly performed on the quaternion function to obtain the generalized quaternion S-transform spectrum. Compared to existing technologies, this avoids splitting each component of the multi-component seismic data for separate processing, thus effectively utilizing the correlation information between the components. Furthermore, the generalized quaternion S-transform can process vectors, and performing it on the quaternion function completely preserves the vector wavefield characteristics of the multi-component seismic data. Filtering the generalized quaternion S-transform spectrum effectively removes noise from the multi-component seismic data. The filtered result is then subjected to an inverse generalized quaternion S-transform to obtain the denoised multi-component seismic data.

[0007] In one embodiment, converting the multi-component seismic data into a quaternion function according to a preset quaternion relation includes: the preset quaternion relation is:

[0008] m The ordinal number of the discrete-time sampling points. p(m) For pressure component seismic data, x(m) for x Component seismic data, y(m) for y Component seismic data, z(m) For z-component seismic data, i、j、k It is the imaginary unit of quaternions.

[0009] In this embodiment, by transforming each component of multi-component seismic data into the same quaternion function according to a preset quaternion relationship, the quaternion function can effectively preserve the correlation between the components of multi-component seismic data. Performing a quaternion generalized S-transform on the quaternion function can also preserve the vector characteristics of each component of the multi-component seismic data.

[0010] In one embodiment, the step of performing a quaternion generalized S-transform on the quaternion function based on a preset quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum includes: performing a quaternion generalized S-transform on the quaternion function based on a time-domain quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum; wherein, the time-domain quaternion generalized S-transform formula is:

[0011]

[0012] m It is any pure identity quaternion. pm *=1 , f For frequency, t The time variable before the transformation. t For the transformed time variable, For frequency-dependent window functions, d(t) For the quaternion function, S (τ, t) The generalized S-transform spectrum of the quaternion is given.

[0013] In this embodiment, seismic data is a typical non-stationary signal. The generalized S-transform, as a tool for analyzing non-stationary signals and removing time-varying noise, is used to analyze multi-component seismic data by performing a quaternion generalized S-transform, which can effectively analyze the characteristics of multi-component seismic data. The preset time-domain quaternion generalized S-transform formula uses a frequency-related window function, and the time-frequency resolution of the frequency band containing the multi-component seismic data can be adjusted based on the window function, facilitating the analysis of multi-component seismic data.

[0014] In one embodiment, the step of performing a quaternion generalized S-transform on the quaternion function based on a preset quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum includes: performing a quaternion generalized S-transform on the quaternion function based on a frequency domain quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum; the frequency domain quaternion generalized S-transform formula is:

[0015]

[0016] in, m It is any pure identity quaternion. pm * =1 , α The frequency variable before transformation, f The transformed frequency variable, t It is a time variable. D (a+f) The quaternion Fourier transform of the quaternion function is... G f （α） Here is the frequency domain expression for the Gaussian window function. S (τ, t) The generalized S-transform spectrum of the quaternion is given.

[0017] In this embodiment of the application, since the generalized S-transform formula of frequency domain quaternions can be calculated using the fast Fourier transform of quaternions, the generalized S-transform of quaternions on quaternion functions using the generalized S-transform formula of frequency domain quaternions can effectively improve the computational efficiency.

[0018] In one embodiment, before filtering the quaternion generalized S-transform spectrum to obtain the filtering result, the method further includes: acquiring the time range and frequency range of effective seismic data; calculating the average amplitude of spectral coefficients based on the quaternion generalized S-transform spectrum outside the time range and outside the frequency range; and constructing a time-frequency domain filter function based on the average amplitude of the spectral coefficients; wherein the time-frequency domain filter function is:

[0019]

[0020] in, m The ordinal number of the discrete-time sampling points. n The discrete frequency sampling point number, N This represents the total number of samples. T The time sampling interval is... Where β is the frequency sampling interval, and β is the threshold adjustment factor, taking the form of a positive real number. abs Calculated for absolute value. These are the time-frequency domain discrete sampled values ​​of the quaternion generalized S-transform spectrum. The average amplitude of the spectral coefficients. The time range is defined as follows. [f1, f2] This refers to the frequency range.

[0021] In this embodiment of the application, by constructing a time-frequency domain filter to filter the quaternion generalized S-transform spectrum, noise in the quaternion generalized S-transform spectrum can be quickly filtered out, thereby improving the efficiency of filtering and noise reduction.

[0022] In one embodiment, filtering the quaternion generalized S-transform spectrum to obtain a filtering result includes: performing a dot product between the time-frequency domain filter function and the quaternion generalized S-transform spectrum to obtain the filtering result.

[0023] In this embodiment, by multiplying the time-frequency domain filter function with the quaternion generalized S-transform spectrum, noise signals in the quaternion generalized S-transform spectrum can be quickly filtered out, thereby obtaining the denoised quaternion generalized S-transform spectrum.

[0024] In one embodiment, before performing a dot product between the time-frequency domain filter function and the quaternion generalized S-transform spectrum, the method further includes: smoothing the time-frequency domain filter function.

[0025] In this embodiment, by smoothing the time-frequency domain filter function, the quaternion generalized S-variance spectrum of multi-component seismic data is made smoother after filtering and denoising by the time-frequency domain filter function, and no signal abrupt changes occur.

[0026] In one embodiment, smoothing the time-frequency domain filter function includes: applying smoothing to each... Win (m, n)=0 Using the center point as the center point, obtain a set of signal points with a time-frequency domain filter function value of 1 within a preset radius of the center point; obtain the shortest distance between each signal point in the signal point set and the center point; assign a value to the center point based on the shortest distance.

[0027] In this embodiment of the application, Win(m,n)=0 The point is determined as the center point. If there is a signal point with a value of 1 within a preset radius from the center point, then this center point is located at the boundary of the multi-component seismic data. Therefore, by smoothly assigning values ​​to the center point, the filtered result obtained after filtering and denoising the quaternion generalized S-variance spectrum of the multi-component seismic data through the time-frequency domain filter function is smoother at the boundary and is not affected by signal abrupt changes.

[0028] Secondly, embodiments of this application provide an electronic device, including a memory and a processor. The memory stores computer-readable instructions, which, when executed by the processor, cause the processor to perform the seismic data denoising method based on quaternion generalized S-transform as described in the first aspect.

[0029] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the seismic data denoising method based on quaternion generalized S-transform as described in the first aspect.

[0030] Other features and advantages of this disclosure will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0031] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0032] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart of a seismic data denoising method based on quaternion generalized S-transform provided in an embodiment of this application.

[0034] Figure 2 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0036] Please see Figure 1 , Figure 1 The flowchart of a seismic data denoising method based on quaternion generalized S-transform is provided for an embodiment of this application. The multi-component seismic data denoising method may include the following steps.

[0037] S110, acquire multi-component seismic data.

[0038] In this embodiment, multi-component seismic data is used to record the motion of particles in the vertical and horizontal directions. It is a record of the seismic wave vector field. Therefore, multi-component seismic data typically includes multiple components, such as pressure components, horizontal and vertical components in the horizontal plane, and vertical components. The seismic data can be physical quantities such as particle velocity, acceleration, or displacement.

[0039] In this embodiment, the acquired multi-component seismic data is multi-component seismic data with a fixed sampling interval. The time sampling interval can be set to T, and the number of sampling points is N.

[0040] In this embodiment, the method of acquiring multi-component seismic data can be either by directly acquiring it through a sampling device for acquiring multi-component seismic data, or by acquiring it from a database that pre-stores multi-component seismic data. Here, this application does not limit the method of acquiring multi-component seismic data.

[0041] S120 converts multi-component seismic data into quaternion functions according to a preset quaternion relationship.

[0042] In one embodiment, each component of the multi-component seismic data is converted into a preset quaternion relation.

[0043] In this embodiment, a quaternion contains one real part and three imaginary parts, and the quaternion can be represented as:

[0044]

[0045] Where i, j, and k are three imaginary units, and t is a time variable. , , , It is a real number function. Therefore, it can be used. , , , This represents different components of earthquake data, and multiple components can be fused into a single signal data for processing.

[0046] In this embodiment, the preset quaternion relation is:

[0047]

[0048] in, m The ordinal number of the discrete-time sampling points. p(m) For pressure component seismic data, x(m) for x Component seismic data, y(m) for y Component seismic data, z(m) for z Component seismic data, i、j、k The imaginary unit of a quaternion. By pre-setting quaternion relationships, the data of each component of multi-component seismic data can be uniformly represented.

[0049] In this embodiment, the multi-component seismic data is a discrete signal. Therefore, m is the discrete-time sampling point number, m=0,1, 2,…, N-1.

[0050] In one embodiment, when a component of multi-component seismic data is missing, the missing component can be replaced with zero. Therefore, denoising can be performed on seismic data with no more than four components using quaternion functions.

[0051] S130, based on the preset quaternion generalized S-transform, performs a quaternion generalized S-transform on the quaternion function to obtain the quaternion generalized S-transform spectrum.

[0052] In one embodiment, performing a quaternion generalized S-transform on a quaternion function includes: performing a quaternion generalized S-transform on the quaternion function based on the time-domain quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum.

[0053] The formula for the time-domain quaternion generalized S-transform is as follows:

[0054]

[0055] m It is any pure identity quaternion. pm * =1 , f For frequency, t The time variable before the transformation. tFor the transformed time variable, For frequency-dependent window functions, d(t) For the quaternion function, S (τ, f) The generalized S-transform spectrum of the quaternion is given.

[0056] in, For example, a window function related to frequency, Gaussian window functions include:

[0057]

[0058] f For frequency, t Let γ be the time variable and γ be a real factor.

[0059] In this embodiment, the window function can be adjusted as needed to adjust the resolution in the frequency domain or time domain. Specifically, γ is used to adjust the width of the Gaussian window. By adjusting γ, the time-frequency resolution can be adjusted. The smaller γ is, the narrower the window and the higher the time domain resolution. Conversely, the larger γ is, the wider the window and the higher the frequency domain resolution.

[0060] In one embodiment, performing a quaternion generalized S-transform on a quaternion function includes: performing a quaternion generalized S-transform on the quaternion function based on the frequency domain quaternion generalized S-transform formula to obtain the quaternion generalized S-transform spectrum.

[0061] In this embodiment, since the time-domain convolution in the time-domain quaternion generalized S-transform formula is equal to the frequency-domain multiplication, the time-domain quaternion generalized S-transform formula can be rewritten as the frequency-domain quaternion generalized S-transform formula. The frequency-domain quaternion generalized S-transform formula is:

[0062]

[0063] in, m It is any pure identity quaternion. pm * =1 , α The frequency variable before transformation, f The transformed frequency variable, t For time variables, D (a+f) The quaternion Fourier transform of quaternion functions G f （α） Here is the frequency domain expression for the Gaussian window function. S (τ, f) It is the quaternion generalized S-transform spectrum.

[0064] In this embodiment, the window function is a Gaussian window function, which can be considered as a quaternion with only a scale part, i.e., a vector part of zero. Therefore, based on the definition of the quaternary Fourier transform, the quaternary Fourier transform of the Gaussian window function can be determined as a quaternion with a zero vector part and a Gaussian scale part. Therefore, the frequency domain expression of the Gaussian window function... for:

[0065]

[0066] Where α is the frequency variable and γ is the real factor.

[0067] In this embodiment, since the frequency domain quaternion generalized S-transform can be calculated using the quaternion fast Fourier transform, the complexity of performing the quaternion generalized S-transform on quaternion functions can be effectively reduced using the frequency domain quaternion generalized S-transform formula, thereby improving computational efficiency. It is understandable that the frequency domain expression of the Gaussian window function... G f （α） Adjusting γ can also adjust the time domain resolution or the frequency domain resolution.

[0068] In this embodiment, after performing a quaternion generalized S-transform on the quaternion function, the quaternion generalized S-transform spectrum can be obtained. ,in, , m is the ordinal number of discrete-time sampling points (m=0, 1,2,…, N-1), and n is the ordinal number of discrete-frequency sampling points (n=0, 1, 2,…, N-1).

[0069] S140 performs filtering on the quaternion generalized S-transform spectrum to obtain the filtering result.

[0070] In one embodiment, the quaternion generalized S-transform spectrum is filtered. Before obtaining the filtering result, a filter is constructed to filter the quaternion generalized S-transform spectrum.

[0071] In this embodiment, a filter is constructed to process the quaternion generalized S-transform spectrum, thereby removing noise signals from the quaternion generalized S-transform spectrum.

[0072] In this embodiment, constructing a time-frequency domain filter includes: first, acquiring the time range and frequency range of valid seismic data; second, calculating the average amplitude of spectral coefficients based on the quaternion generalized S-transform spectrum outside the time range and outside the frequency range; and finally, constructing a time-frequency domain filter function based on the average amplitude of the spectral coefficients.

[0073] In this embodiment, the signal to be processed can be filtered, that is, the signal segment to be processed can be selected as the effective seismic signal for processing. Therefore, the time interval of the effective seismic signal can be set as follows: The frequency range is [f1, f2] .

[0074] After determining the time interval of the valid seismic signal, the quaternion generalized S-transform spectrum in the time-frequency domain is averaged along the time axis to obtain the average amplitude of the spectral coefficients for each frequency. ,in, The frequency sampling interval, n The discrete frequency sampling point number, N This represents the total number of samples. T The sampling interval is denoted as .

[0075] A time-frequency domain filter can be constructed using the average amplitude of the obtained spectral coefficients.

[0076] For example, a time-frequency domain filter can be:

[0077]

[0078] in, m The ordinal number of the discrete-time sampling points. n The discrete frequency sampling point number, T The time sampling interval is... N This represents the total number of samples. Where β is the frequency sampling interval, and β is the threshold adjustment factor, taking the form of a positive real number. abs Calculated for absolute value. The average amplitude of the spectral coefficients. For time range, [f1, f2] This refers to the frequency range.

[0079] It should be noted that the positive and negative frequency coefficients of the quaternion generalized S-transform are not conjugate symmetric, for example, and While not conjugate symmetric, the signal of any frequency component is described by the corresponding positive and negative frequency coefficients. Therefore, when designing a time-frequency domain filter, both the positive and negative frequency components must be considered simultaneously, resulting in a time-frequency domain filter where the positive and negative frequencies are symmetrical.

[0080] In one embodiment, the time-frequency domain filter function is multiplied by the quaternion generalized S-transform spectrum to obtain the filtering result.

[0081] In this embodiment, the time-frequency domain filter function can be used to determine whether the signal at a certain point in the quaternion generalized S-transform spectrum is seismic data signal or noise. When = 1, the signal at that point in the quaternion generalized S-transform spectrum is the signal of the seismic data. When the quaternion generalized S-transform spectrum is zero, the signal at that point is a noise signal. By performing a dot product between the quaternion generalized S-transform spectrum and the time-frequency domain filter function to zero out the noise signal, the noise signal can be effectively removed, thus obtaining the filtering result.

[0082] S150 performs an inverse quaternion generalized S-transform on the filtered result to convert the filtered result into denoised multi-component seismic data.

[0083] In this embodiment, the inverse transformation formula of the quaternion generalized S-transformation can be:

[0084]

[0085] in, S (τ, f) Let μ be the quaternion generalized S-transform spectrum, where μ is any pure unit imaginary quaternion. pm * =1, f is the frequency, and τ is the time.

[0086] In this embodiment, the multi-component seismic data is converted into a quaternion function according to a preset quaternion relationship. This allows multiple components of the multi-component seismic data to be integrated into a single data set, i.e., a quaternion function. A generalized quaternion S-transform is then performed on the quaternion function to obtain the generalized quaternion S-transform spectrum. Compared to existing technologies, this avoids the need to process each component of the multi-component seismic data separately, thus effectively utilizing the correlation information between the components. Furthermore, the generalized quaternion S-transform can process vectors, thereby preserving the vector wavefield characteristics of the multi-component seismic data. Filtering the generalized quaternion S-transform spectrum effectively removes noise from the multi-component seismic data. An inverse transform of the filtering result yields denoised multi-component seismic data.

[0087] In one embodiment, the time-frequency domain filter function is smoothed before performing a dot product with the quaternion generalized S-transform spectrum.

[0088] In this embodiment, since the time-frequency domain filter takes the value of 0 or 1, when the quaternion generalized S-transform spectrum of multi-component seismic data is filtered using the time-frequency domain filter, the signal will have abrupt changes. That is, when the time-frequency domain filter takes the value of 1 at a certain point and the value of 0 at the next point, these two points will show abrupt changes when the data is processed. Therefore, the time-frequency domain filter function can be smoothed to make the data obtained by the time-frequency domain filter function smoother.

[0089] In one embodiment, smoothing the time-frequency domain filter can be performed to achieve two-dimensional mean filtering.

[0090] In this embodiment, the time-frequency domain filter function can be processed by two-dimensional mean filtering to make the time-frequency domain filter function smoother. It is understood that two-dimensional mean filtering can refer to existing techniques, and will not be elaborated upon here.

[0091] In one embodiment, smoothing the time-frequency domain filter function may further include: applying smoothing to each... The center point is selected as the center point; a set of signal points with a value of 1 is obtained from the signal points of the time-frequency domain filter function within a preset radius of the center point; the shortest distance between each signal point in the set of signal points and the center point is obtained; and the center point is assigned a value based on the shortest distance.

[0092] In this embodiment, for any point of the time-frequency domain filter function (m, n) In determining At that time, no action is taken on that point.

[0093] In this embodiment, for any point of the time-frequency domain filter function (m, n) In determining If the value is 1, then take that point as the center point and search within a preset radius of b to find if there is a point with a value of 1.

[0094] In this embodiment, if there is no point with a value of 1 within the preset radius range of radius b, then the value of the center point is determined to be 0.

[0095] In this embodiment, when it is determined that there are points with a value of 1 within a preset radius range of radius b, a set of signal points with a value of 1 is obtained, and the distance from each point in the signal point set to the center point is calculated. After obtaining the distances from each point to the center point, the distances are compared with each other, and the shortest distance is selected and set as d. It can be understood that if there are no points with a value of 1 within the preset range, it means that the center point is located in a non-seismic data signal segment. If there are points with a value of 1 within the preset radius range, it means that the center point is close to a seismic data signal segment, and the center point is at the boundary between seismic data and non-seismic data.

[0096] In this embodiment, after obtaining the shortest distance d, the value of the time-frequency domain filter function of the center point is assigned as... The original center point is about to be reached. Modified to Where b is the preset boundary width for smoothing, and b is a positive integer. Thus, since the shortest distance from the center point to the point with a value of 1 is d, assigning a value to the center point can insert a transition band where the value gradually decreases from 1 to 0 between the region with a value of 1 and the region with a value of 0, so that the quaternion generalized S-transform spectrum processed by the time-frequency domain filter function will not have signal abrupt changes at the boundary.

[0097] The method provided in the above embodiments can be implemented as a computer-readable instruction, which can be implemented in, for example... Figure 2 It runs on the electronic device shown.

[0098] This application also provides an electronic device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the program to implement the method provided in the above embodiments.

[0099] Figure 2 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application. The electronic device may be a server. Please refer to... Figure 2 The electronic device includes a processor, a non-volatile storage medium, internal memory, an input device, a display screen, and a network interface connected via a system bus. The non-volatile storage medium can store an operating system and computer-readable instructions. When these computer-readable instructions are executed, they cause the processor to perform the method provided in the above embodiments of this application. The specific implementation process of this method can be found in [reference needed]. Figure 1 The specific details will not be elaborated here. The processor of this electronic device provides computing and control capabilities to support the operation of the entire electronic device. The internal memory can store computer-readable instructions, which, when executed by the processor, cause the processor to perform the methods provided in the above embodiments. The input device of the electronic device is used for inputting various parameters, the display screen of the electronic device is used for display, and the network interface of the electronic device is used for network communication. Those skilled in the art will understand that... Figure 2 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0100] Based on the same inventive concept, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which executes the methods provided in the above embodiments when the computer program is run.

[0101] The storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0102] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0103] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0104] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0105] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for seismic data denoising based on quaternion generalized S transform, characterized in that, The method comprises: acquiring multi-component seismic data; converting the multi-component seismic data into a quaternion function according to a preset quaternion relationship; performing quaternion generalized S transform on the quaternion function based on a preset quaternion generalized S transform formula to obtain a quaternion generalized S transform spectrum; filtering the quaternion generalized S transform spectrum to obtain a filtering result; performing inverse quaternion generalized S transform on the filtering result to convert the filtering result into noise-reduced multi-component seismic data.

2. The method of claim 1, wherein, The converting the multi-component seismic data into a quaternion function according to a preset quaternion relationship comprises that the preset quaternion relationship is: m is a discrete time sample point ordinal, p(m) is pressure component seismic data, x(m) is x component seismic data, y(m) is y component seismic data, z(m) is z component seismic data, i, j, k is the imaginary unit of quaternions.

3. The method of claim 1, wherein, The performing quaternion generalized S transform on the quaternion function based on a preset quaternion generalized S transform formula to obtain a quaternion generalized S transform spectrum comprises performing quaternion generalized S transform on the quaternion function based on a time-domain quaternion generalized S transform formula to obtain the quaternion generalized S transform spectrum. The time-domain quaternion generalized S transform formula is: μ is an arbitrary pure unit quaternion, μμ * =1 , f is a frequency, t is a time variable before the transform, τ is a time variable after the transform, is a window function associated with the frequency, d t is the quaternion function, S τ , f is the quaternion generalized S-transform spectrum.​​ 4. The method of claim 1, wherein, The performing quaternion generalized S transform on the quaternion function based on a preset quaternion generalized S transform formula to obtain a quaternion generalized S transform spectrum comprises performing quaternion generalized S transform on the quaternion function based on a frequency-domain quaternion generalized S transform formula to obtain the quaternion generalized S transform spectrum. The frequency-domain quaternion generalized S transform formula is: where μ is an arbitrary pure unit quaternion, μμ * =1 , α is the frequency variable before the transform, f is the frequency variable after the transform, τ is the time variable, D ( α+f ) is the quaternion Fourier transform of the quaternion function, G f ( α ) is the frequency domain representation of the Gaussian window function, S ( τ , f ) is the quaternion generalized S-transform spectrum.

5. The method of claim 1, wherein, Before the filtering the quaternion generalized S transform spectrum to obtain a filtering result, the method further comprises: acquiring a time range and a frequency range of effective seismic data; calculating an average value of spectral coefficient amplitudes according to the quaternion generalized S transform spectrum outside the time range and outside the frequency range; constructing a time-frequency domain filter function based on the average value of spectral coefficient amplitudes; The time-frequency domain filter function is: wherein, m is a discrete time sample point sequence number, n is a discrete frequency sample point sequence number, T is a time sample interval, is a frequency sample interval, N is a total number of samples, β is a threshold adjustment factor, and is a positive real number, abs is an absolute value calculation, is a time-frequency domain discrete sample value of the quaternion generalized S-transform spectrum, is an average value of the spectrum coefficient amplitude, is the time range, [f1, f2] is the frequency range.

6. The method of claim 5, wherein, The filtering the quaternion generalized S transform spectrum to obtain a filtering result comprises point multiplying the time-frequency domain filter function and the quaternion generalized S transform spectrum to obtain the filtering result.

7. The method of claim 6, wherein, Before the point multiplying the time-frequency domain filter function and the quaternion generalized S transform spectrum, the method further comprises smoothing the time-frequency domain filter function.

8. The method of claim 7, wherein, The smoothing of the time-frequency domain filter function comprises: respectively taking each point with a value of 0 as a center point, obtaining a signal point set with a value of 1 in a time-frequency domain filter function within a preset radius range of the center point; obtaining the shortest distance between each signal point in the signal point set and the center point; and assigning a value to the center point based on the shortest distance. Win m n = 0 as a center point, obtaining a signal point set with a value of 1 in a time-frequency domain filter function within a preset radius range of the center point; obtaining the shortest distance between each signal point in the signal point set and the center point; and assigning a value to the center point based on the shortest distance.​​​ 9. An electronic device, comprising: A device comprises a memory and a processor, the memory stores computer readable instructions, and the computer readable instructions are executed by the processor to make the processor execute the method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, A readable storage medium stores a computer program, and when the computer program runs on a computer, the computer executes the method according to any one of claims 1-8.

Images