A seismic data adaptive denoising method and system based on quantum mechanics

Abstract

The application discloses a kind of seismic data adaptive denoising method and system based on quantum mechanics, belong to seismic data denoising field, this method includes the following steps: based on the predictability of seismic signal in frequency-space domain, seismic data is mapped into quantum potential energy field, and the eigenvalue problem of solving Schrodinger equation is used to construct adaptive quantum base function;Quantum base function is used to carry out sparse representation to seismic data, and the expansion coefficient of signal on quantum base is obtained;Using energy-based threshold processing mechanism, the expansion coefficient is suppressed to noise, and the coefficient after denoising is obtained;The seismic data after denoising is reconstructed by linear combination of quantum base function using the coefficient after denoising.The application can be seamlessly connected with existing seismic data processing procedure, facilitate the use of seismic data processing personnel, improve the analysis precision of seismic data.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of seismic data denoising, and particularly relates to a seismic data adaptive denoising method and system based on quantum mechanics. BACKGROUND

[0002] Seismic data random noise suppression is a key processing step for improving interpretation and inversion accuracy and realizing efficient seismic exploration. Random noise is irregularly distributed on a seismic profile, covering effective signals, and needs to be suppressed by using a denoising algorithm. Random noise suppression algorithms for post-stack seismic data include a structure-oriented algorithm, an fx domain prediction filtering algorithm, a sparse transform domain noise attenuation method based on compressive sensing, a deep learning algorithm, and a non-local similarity algorithm introduced from image processing.

[0003] The above algorithms all have their own application conditions. For example, the structure-oriented algorithm can effectively preserve large dip angle data such as faults, so the algorithm is preferentially used in data processing with rich faults. The fx domain prediction filtering removes high wave number information according to the predictability of the same phase axis, so the fx domain prediction filtering is generally suitable for data with relatively flat strata. The sparse transform domain denoising method has different application data ranges according to the different types of transform domains. The method is based on the characteristics that effective signals are sparsely or densely distributed in the transform domain, while random noise is randomly distributed, and noise suppression is realized through threshold processing. Common transform methods include fk transform, wavelet transform, curvelet transform, and Radon transform. It should be noted that this kind of global transform algorithm may introduce new noise components in the transform domain when applied. The random noise suppression method based on deep learning relies on high-quality training data. Although transfer learning improves the generalization ability of the algorithm, the processing effect is still limited in areas with dramatic changes in structure. The non-local similarity denoising algorithm uses the self-similarity characteristics of seismic data to suppress noise by matching and fusing similar data blocks. However, due to the influence of the fractal characteristics of strata, the method may lose part of the effective signal details in the denoising process, so it is more suitable for strata with strong similarity or stable sedimentary characteristics.

[0004] Existing seismic data denoising techniques generally suffer from key limitations: while structure-guided algorithms are highly effective in protecting steeply dipping faults, they struggle to accurately identify and preserve minute geological structures; fx-domain predictive filtering methods perform well in simple stratigraphic structures but are less adaptable to complex wavefields and high-frequency noise; sparse transform-based techniques (such as wavelet and curvelet transforms) are constrained by the global transform assumption, potentially introducing artifacts and requiring strictly sparsity signals; deep learning methods are highly dependent on the quality of training data and are prone to overfitting or underfitting in areas with complex geological conditions; nonlocal similarity algorithms are ideal in homogeneous stratigraphy but can obscure subtle geological features; and traditional denoising algorithms (such as BM3D) effectively suppress noise but often at the expense of stratigraphic detail.

[0005] In summary, these technical bottlenecks severely restrict the realization of high-precision seismic interpretation, and there is an urgent need to develop innovative methods that can effectively suppress noise while accurately preserving complex small-scale geological features. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention proposes an adaptive denoising method and system for seismic data based on quantum mechanics, thereby resolving the issues present in the prior art.

[0007] To achieve the above objectives, this invention provides an adaptive denoising method for seismic data based on quantum mechanics, comprising:

[0008] Based on the predictability of seismic signals in the frequency-space domain, seismic data is mapped to a quantum potential field, and an adaptive quantum basis function is constructed by solving the eigenvalue problem of the Schrödinger equation.

[0009] The quantum basis functions are used to perform sparse representation of seismic data to obtain the expansion coefficients of the signal on the quantum basis.

[0010] An energy-based thresholding mechanism is used to suppress noise in the expansion coefficients to obtain denoised coefficients.

[0011] The denoised seismic data is reconstructed by linear combination of quantum basis functions using the denoised coefficients.

[0012] Optionally, the expression for the predictability of seismic signals in the frequency-spatial domain is:

[0013] ;

[0014] In the formula, For the first Earthquake data at circular frequency Spectral values ​​at that location For the first Earthquake data at circular frequency a spectral value at a location, denotes a circular frequency, denotes a contribution weight of a th neighboring trace to a current trace, is a linear combination of neighboring th traces, denotes a th trace, denotes a th trace.

[0015] Optionally, the expression of mapping the seismic data into a quantum potential field is:

[0016] ;

[0017] wherein m is a mass of a particle, is a reduced Planck constant, is a Laplacian operator, representing a second-order derivative of a space, E is an energy of the particle, is a wave function of the particle, is a potential field.

[0018] Optionally, the process of constructing an adaptive quantum basis function comprises:

[0019] constructing a quantum potential field matrix based on the seismic data;

[0020] processing the quantum potential field matrix by using a matrix eigenvalue decomposition method to obtain corresponding eigenvalues and eigenvectors;

[0021] obtaining the adaptive quantum basis function based on the eigenvectors.

[0022] Optionally, the expression of solving the adaptive quantum basis function is:

[0023] ;

[0024] wherein is an eigenvalue, is a vector form of a wave function , H QAB denotes a Hamiltonian operator of a quantum system.

[0025] Optionally, the coefficient after denoising is:

[0026] ;

[0027] wherein is a threshold function, is a coefficient after threshold processing, is an expansion coefficient on a quantum basis.

[0028] Optionally, the denoised seismic data is reconstructed as:

[0029]

[0030] wherein, is the denoised seismic data, is the wave function of the i-th particle, is the threshold processed coefficient.

[0031] The application further provides a quantum-mechanics-based seismic data adaptive denoising system for implementing the method, and the system comprises:

[0032] a quantum potential field mapping module, configured to map the seismic data into a quantum potential field based on a corresponding relationship between seismic signal amplitude and the quantum potential field;

[0033] an adaptive quantum basis function construction module, configured to process the quantum potential field by using a method for solving an eigenvalue problem of a Schrodinger equation to obtain an adaptive quantum basis function;

[0034] a sparse representation module, configured to perform sparse representation processing on the seismic data based on the adaptive quantum basis function to obtain expansion coefficients of the signal on the quantum basis;

[0035] a noise suppression module, configured to process the expansion coefficients by using an energy-based threshold processing mechanism to obtain denoised coefficients;

[0036] a denoised data reconstruction module, configured to reconstruct the denoised seismic data by using the denoised coefficients through a linear combination method of the quantum basis function to obtain final denoised seismic data.

[0037] The application further provides a computer comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method when executing the computer program.

[0038] The application further provides a storage medium having a computer program stored thereon, wherein the program is executable on a processor to implement the method.

[0039] Compared with the prior art, the application has the following advantages and technical effects:

[0040] ​​The application utilizes the self-adaptive generated quantum basis function to perform sparse representation on signals, can accurately capture the complex local features in seismic data, and especially the tiny geological structure. The threshold processing mechanism based on energy can intelligently distinguish effective signals and random noises, and realizes high-precision noise suppression. Compared with the traditional denoising algorithm, the method can better retain the detail information of small-scale geological structures such as faults, thin layers and river channels while effectively removing random noises, significantly improves the signal-to-noise ratio and structural similarity of the denoised data, and provides a more reliable data basis for high-precision seismic interpretation and inversion. BRIEF DESCRIPTION OF DRAWINGS

[0041] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application and are incorporated herein for a purpose of explanations and are not intended as improper limitations on the present application. In the drawings:

[0042] Figure 1 The flow chart of the quantum mechanics fx domain seismic data self-adaptive denoising of the embodiment of the present application;

[0043] Figure 2 The test signal of the embodiment of the present application;

[0044] Figure 3 The self-adaptive quantum basis function of the embodiment of the present application;

[0045] Figure 4 The self-adaptive quantum basis function expansion coefficient of the embodiment of the present application;

[0046] Figure 5 The clean data and noisy data participating in the test of the embodiment of the present application;

[0047] Figure 6 The non-sparse theoretical model test result of the embodiment of the present application;

[0048] Figure 7 The noise suppressed by the algorithm of the embodiment of the present application;

[0049] Figure 8 The local structural similarity of the embodiment of the present application. DETAILED DESCRIPTION

[0050] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0051] It should be noted that the steps shown in the flow chart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flow chart, in some cases, the steps shown or described herein can be executed in a different order.

[0052] The present application aims to solve the problem that the conventional seismic data random noise suppression algorithm faces when processing seismic data containing small structures. The existing method often has difficulty in effectively preserving small-scale structural details in seismic data while removing noise, resulting in signal distortion or reduced resolution. For example, structure-oriented algorithms can protect large dip data, but they do not adequately protect small-scale structural details; fx-domain predictive filtering is suitable for flat layers but has poor adaptability to complex structures; deep learning methods rely on training data and are limited in areas with large structural differences; and non-local similarity algorithms may blur signal details due to excessive smoothing. Therefore, an fx-domain seismic data random noise suppression algorithm based on quantum mechanics is proposed, which combines seismic signal potential energy distribution with quantum state representation by constructing an adaptive quantum basis function, achieving high-precision separation of noise and signal in the energy domain. This technology not only effectively suppresses random noise, but also fully preserves the details of complex small-scale structures, significantly improving the interpretation accuracy and reliability of seismic data, and is particularly suitable for exploration areas with complex geological structures.

[0053] The main work of the present technology is as follows: 1. Map the amplitude of the seismic signal to the quantum potential energy field, and construct an adaptive quantum basis function by solving the eigenvalue problem of the Schrödinger equation; 2. Develop a new fx-domain quantum mechanics denoising algorithm (fx-QABD), which uses quantum basis functions to sparsely represent seismic data, and then uses an energy-based threshold processing mechanism to suppress random noise in the expansion coefficients on the quantum basis; 3. Model tests verify the superiority of the algorithm in processing seismic data containing complex small structures, especially its ability to remove random noise while preserving small-scale structural details; 4. The algorithm's versatility and effectiveness are further tested using actual seismic data, demonstrating its applicability and superiority under different geological conditions. The present technology has good integration in terms of algorithm, can improve the analysis accuracy of seismic data, is convenient for seismic data processing personnel to use, and has high promotional value.

[0054] As shown in Figure 1 , the present embodiment provides a quantum mechanics-based adaptive seismic data denoising method, including the following steps: based on the predictability of seismic signals in the frequency-space domain, map the seismic data to a quantum potential energy field, and construct an adaptive quantum basis function by solving the eigenvalue problem of the Schrödinger equation; use the quantum basis function to sparsely represent the seismic data to obtain the expansion coefficients of the signal on the quantum basis; use an energy-based threshold processing mechanism to suppress noise in the expansion coefficients to obtain the denoised coefficients; and use the denoised coefficients to reconstruct the denoised seismic data through linear combination of the quantum basis function.

[0055] The quantum mechanics-based fx-domain adaptive seismic data denoising flowchart is as follows:Figure 1 This patent contains four modules: the first is to map the seismic signal amplitude to the quantum potential field, and to construct the adaptive quantum basis function by solving the eigenvalue problem of the Schrödinger equation; the second is to develop a new quantum mechanics denoising algorithm in the fx domain (fx-QABD), which uses quantum basis functions to sparsely represent seismic data, and then uses an energy-based threshold processing mechanism to suppress random noise in the expansion coefficients on the quantum basis; the third is to verify the superiority of the algorithm in processing seismic data containing complex small structures through model testing, especially in retaining small-scale structural details while removing random noise; the fourth is to further test the universality and effectiveness of the algorithm using real seismic data, which proves its applicability and superiority under different geological conditions.

[0056] The predictability of seismic data in the fx domain indicates that the fx domain predictive filtering random noise suppression method is based on the predictability of linear events in the frequency domain, and it is to predict all dip angle events at the same time. Assuming that the seismic data s is a plane wave function about time t and spatial position x, the seismic data s(t,x) can be represented as a linear superposition of N inclined events with a slope p i , that is:

[0057] (1)

[0058] where, is the Dirac function, indicating the ideal pulse reflection (no width, no distortion) produced by the i th reflection interface at time , represents a two-dimensional linear convolution, is the seismic wavelet related to the i th reflection event, which is convolved with the Dirac function located at time , is the intercept time at the position of a certain event , is the slope of the i th event, and the Fourier transform of equation (1) is:

[0059] (2)

[0060] ω is the circular frequency, is the imaginary unit, is the Fourier transform of the wavelet , and according to the Auto-Regressive Model, the i may be represented as the superposition of seismic traces at different spatial locations within a spatial window:

[0061] (3)

[0062] wherein, is the spectral value of the i-th seismic data at the circular frequency is the spectral value of the i-th seismic data at the circular frequency is the spectral value of the i-th seismic data at the circular frequency is the spectral value of the i-th seismic data at the circular frequency is the spectral value of the i-th seismic data at the circular frequency denotes the circular frequency, denotes the contribution weight of the i-th adjacent trace to the current trace, is a linear combination of the adjacent seismic traces, denotes the i-th seismic trace, denotes the i-th seismic trace. wherein, , reflects the frequency proportional relationship of the wavelet amplitude in the two seismic data, is the trace interval, the effective signal wave number is low, and the noise wave number is high. Within the spatial window, when

[0063] is small, the data is correlated; when is large, any one data cannot be represented by other data, that is, it is considered as high wave number noise or random noise. Therefore, in the fx domain, noise attenuation can be achieved by reducing high wave number energy. Similarly, adaptive signal decomposition in the fx domain can also achieve noise attenuation. Quantum mechanics and construction of quantum basis functions: the algorithm uses the regional predictability feature to realize the separation of signal and noise by constructing quantum basis functions. The core idea is to adaptively decompose the signal, and in the coefficient domain of the quantum basis function expansion, the effective signal presents a sparse feature, while the noise presents a non-sparse distribution, and the random noise suppression is realized based on the difference.

[0064] The process of constructing adaptive quantum basis functions includes: constructing a quantum potential field matrix based on seismic data; using matrix eigenvalue decomposition method to process the quantum potential field matrix to obtain corresponding eigenvalues and eigenvectors; and obtaining adaptive quantum basis functions based on the eigenvectors.

[0065] Quantum mechanics and construction of quantum basis functions: the algorithm uses the regional predictability feature to realize the separation of signal and noise by constructing quantum basis functions. The core idea is to adaptively decompose the signal, and in the coefficient domain of the quantum basis function expansion, the effective signal presents a sparse feature, while the noise presents a non-sparse distribution, and the random noise suppression is realized based on the difference.

[0066] The process of constructing adaptive quantum basis functions includes: constructing a quantum potential field matrix based on seismic data; using matrix eigenvalue decomposition method to process the quantum potential field matrix to obtain corresponding eigenvalues and eigenvectors; and obtaining adaptive quantum basis functions based on the eigenvectors.

[0067] ​​​Quantum theory gives the probability of finding a quantum particle at a certain point. If a quantum particle with energy E probes this surface, then the probability of this quantum particle existing at a certain position on the surface will be given by the wave function. The dynamic behavior of particles in a potential field is determined. This can be described by the Schrödinger equation:

[0068] (4)

[0069] in, It is the mass of the particle. It is the reduced Planck constant. It is the Laplace operator, representing the second derivative in space. It is the energy of the particles. It is the wave function of the particle, the square of its absolute value. Indicates the position of the particle The probability density distribution. Equation (4) can be transformed into solving the eigenvalue problem of the matrix:

[0070] (5)

[0071] The Hamiltonian operator of a quantum system can be represented as ,in, For eigenvalues, It is a wave function The vector form of this operator has two important properties regarding its eigenfunctions: first, they form an orthogonal set, meaning the wavefunctions corresponding to different energy levels are mutually orthogonal; second, they form a complete set, such that any quantum state can be represented as a linear superposition of these eigenstates. Its solution set gives a set of eigenvalues ​​(energies). Related stationary solutions .

[0072] (6)

[0073] in, Operator The One portion, and Represents the row and column indices of the matrix. Indicates the first Potential energy values ​​at each grid point.

[0074] By solving the Schrödinger equation (4) or solving for the eigenvalues ​​of the matrix (5), a set of orthogonal wave functions can be obtained. These wavefunctions constitute an adaptive quantum basis, which can be denoted as the Quantum Adaptive Basis (QAB) in this algorithm. Each wavefunction... associated with a particular energy and have different oscillation frequencies, which depend on the local potential energy value of the signal or image, i.e.:

[0075] (7)

[0076] where, is the energy of the th eigenstate, then, based on the quantum basis functions, the signal can be represented as a linear combination of these basis functions:

[0077] (8)

[0078] where, is the coefficient, is the wave function of the th particle, which can be calculated by the inner product:

[0079] (9)

[0080] where, denotes the th quantum basis function. In the spatial direction, noise mainly affects the high wave number components of the signal, and can be removed by thresholding the coefficients . The purpose of thresholding is to retain the basis function expansion coefficients associated with low energy , which correspond to the low wave number components of the signal, and remove the noise components associated with high energy .

[0081] The coefficients after thresholding can be represented as:

[0082] (10)

[0083] where is the threshold function, which determines whether to retain a certain coefficient according to the energy and the hyperparameters and , where, denotes the smoothing parameter, which is used to control the slope of the threshold function, is the threshold parameter, which is used to determine how many largest coefficients to retain. The denoised signal or image is reconstructed using the processed coefficients :

[0084] (11) ​

[0085] The spatial direction of Fourier transform of seismic data is similar to irregular harmonic, while random noise is still randomly distributed in the whole signal. In order to simulate this phenomenon, the embodiment constructs a similar signal model as shown in Figure 2 Based on equation (5), the quantum basis functions are obtained as shown in Figure 3 These basis functions can adaptively detect low potential energy region with higher frequency and high potential energy region with lower frequency. By applying these quantum basis functions to decompose the signal, the embodiment obtains the projection coefficients of the signal on the quantum adaptive basis as shown in Figure 4 In the projection, the effective signal presents the characteristics of sparse distribution, while the noise exists in the form of non-sparse. Based on soft threshold, the random noise can be effectively suppressed to obtain the denoised signal. The soft threshold processing realizes the effective separation of signal and noise by selectively reducing the non-sparse noise component and retaining the sparse signal component.

[0086] As mentioned above, the eigenvector (called wave vector in quantum physics) is an oscillation function, and the oscillation frequency of the function is usually proportional to the local value of QAB has adaptivity, and the difference between it and Fourier basis and wavelet basis is that the basis vector detects low potential energy region with higher frequency and high potential energy region with lower frequency adaptively, and the dependence of local frequency on data value can be adjusted by super parameter This means that compared with fixed basis function, the quantum basis function has higher degree of freedom, that is, higher adaptivity.

[0087] At the same time, the adaptive quantum basis function is constructed according to the characteristics of the signal itself, which means that the basis function can adapt to the local characteristics of the signal, which is similar to dictionary learning, so that the wave function is dynamically generated according to the specific content of the signal or image. This adaptivity enables the basis function to more accurately capture the characteristics of the signal while ignoring irrelevant noise components. The wave function in quantum mechanics has the characteristics of localization, and their oscillation frequency depends on the local potential energy of the signal. Therefore, these basis functions can use different frequency components in different regions of the signal, so as to better represent the local characteristics of the signal. Moreover, the high wave number components of the signal in fxdomain (corresponding to high energy quantum state) are often associated with noise, while the low wave number components are related to the main information of the signal. By threshold processing the coefficients in the energy domain, the high frequency components corresponding to the noise can be removed, while the low frequency components of the signal are retained.

[0088] In addition to the above discussion, the quantum localization phenomenon refers to the localization of wave function in disordered potential energy, which will lead to the excessive concentration of quantum basis function in noise signal or image. In order to solve this problem, the adaptive basis function can be obtained by applying the signal after pre-noise reduction to reduce the localization of the basis function.

[0089] Model test: In order to test the denoising effect of the algorithm, a model is synthesized in this embodiment, which is a profile of the Stanfold geological model. The model data shows delta facies and river facies deposition from shallow to deep, and the profile contains many small-scale structures such as branch channels, as shown in Figure 5 . The original data presents the characteristics of anticline structure. According to the provided horizon information, the data is flattened in this embodiment to facilitate the testing of the denoising ability of the algorithm. On this basis, random noise is added to simulate the noise in real geological data. The signal-to-noise ratio of the data after adding noise is shown in Table 1. The denoising results of different algorithms are intuitively compared, as shown in Figure 6 and Figure 7 . The test results are shown. Considering that the denoising effect of the deep learning method depends largely on the quality of the training data set, in order to ensure the fairness and objectivity of the comparison, the deep learning-based method is not included in the comparison range. Among the traditional denoising algorithms compared, the frequency-space domain smoothing filter (fx-smooth) has residual noise, the frequency-space domain prediction filter (fx-decon) loses part of the signal, and the three-dimensional block matching denoising algorithm (BM3D algorithm) shows advantages in maintaining the continuity of the same phase axis and spatial consistency, but its smoothing mechanism can have a significant impact on subtle geological structures. While suppressing noise, this method may cause the clarity of small-scale geological features such as micro-faults and thin layers to decrease, resulting in a loss of horizontal detail resolution in the processing results. This characteristic makes this algorithm more suitable for geological scenes dominated by large structures and with low requirements for subtle structures. In contrast, the proposed frequency-space domain quantum adaptive basis denoising algorithm (fx-QABD algorithm) basically retains the response of small-scale structures such as branch channels during the denoising process, while effectively suppressing noise, and the numerical value of local structure similarity is the largest, indicating that the similarity to the real clean data is better, as shown in Figure 8 . This shows that the fx-QABD algorithm can provide good denoising effect while preserving geological details.

[0090] Table 1

[0091]

[0092] Note: SNR: signal-to-noise ratio, the larger the value, the less the noise; PSNR: peak signal-to-noise ratio, the larger the value, the less the noise; SSIM: structural similarity, the larger the value, the closer to the clean signal.

[0093] The application first applies quantum mechanics principles to noise suppression of seismic data, and uses quantum basis functions to perform sparse representation on signals, which is a brand-new method; the application constructs an adaptive quantum basis function which can be automatically adjusted according to local characteristics of seismic signals, so that the algorithm can more accurately capture signal characteristics and suppress noise; the application introduces an adaptive threshold mechanism for suppressing non-sparse noise components, which can be dynamically adjusted according to local potential energy values of signals, thereby improving the accuracy of noise removal; the algorithm has good integrality, can be seamlessly connected with an existing seismic data processing flow, is convenient for use by seismic data processing personnel, and improves the analysis accuracy of seismic data.

[0094] The application further provides an adaptive seismic data denoising system based on quantum mechanics, which is used for implementing the method, and the system comprises a quantum potential energy field mapping module, an adaptive quantum basis function construction module, a sparse representation module, a noise suppression module and a denoised data reconstruction module.

[0095] The application further provides a computer comprising a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor implements the method when executing the computer program.

[0096] The application further provides a storage medium having a computer program stored thereon, and the program is executed by a processor to implement the method.

[0097] The above is only a preferred specific embodiment of the application, but the protection scope of the application is not limited to this, any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the application, which should be covered in the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1. A seismic data adaptive de-noising method based on quantum mechanics, characterized in that, The method comprises the following steps: mapping the seismic data into a quantum potential field based on the predictability of the seismic signal in the frequency-space domain, and constructing an adaptive quantum basis function by solving an eigenvalue problem of the Schrödinger equation; wherein the expression based on the predictability of the seismic signal in the frequency-space domain is: ； In the formula, For the first Earthquake data at circular frequency Spectral values ​​at that location For the first Earthquake data at circular frequency Spectral values ​​at that location Represents angular frequency. Indicates the first The contribution weight of each neighboring path to the current path. For the neighboring Linear combination of seismic traces. Indicates the first One earthquake channel, Indicates the first One earthquake track; the expression for mapping the seismic data into a quantum potential field is: ； where m is the mass of the particle, is the reduced Planck constant, is the Laplacian, representing the second derivative in space, E is the energy of the particle, is the wave function of the particle, is the potential field; The process of constructing the adaptive quantum basis function comprises: constructing a quantum potential field matrix based on the seismic data; processing the quantum potential field matrix by using a matrix eigenvalue decomposition method to obtain corresponding eigenvalues and eigenvectors; and obtaining the adaptive quantum basis function based on the eigenvectors; the expression for solving the adaptive quantum basis function is: ； wherein is an eigenvalue, is a wave function in vector form, HQAB represents a Hamiltonian operator of the quantum system; sparse representation of the seismic data is performed by using the quantum basis function to obtain the expansion coefficients of the signal on the quantum basis; noise suppression is performed on the expansion coefficients by using an energy-based threshold processing mechanism to obtain denoised coefficients; denoised seismic data is reconstructed by linear combination of the quantum basis function using the denoised coefficients.

2. The quantum-mechanics-based seismic data adaptive de-noising method according to claim 1, characterized in that, The denoised coefficients are: ； wherein is a threshold function, is a thresholded coefficient, is an expansion coefficient on a quantum basis.

3. The quantum-mechanics-based seismic data adaptive de-noising method according to claim 2, characterized in that, The reconstructed denoised seismic data is: ； In the formula, to reconstruct the denoised seismic data, is the first wave function of the i-th particle, is the threshold-processed coefficient.

4. A system for seismic data adaptive de-noising based on quantum mechanics, characterized in that, The system is used to implement the method according to any one of claims 1-3, and the system comprises: a quantum potential field mapping module configured to map the seismic data into a quantum potential field based on the corresponding relationship between the seismic signal amplitude and the quantum potential field; an adaptive quantum basis function construction module configured to process the quantum potential field by using a method of solving an eigenvalue problem of the Schrödinger equation to obtain an adaptive quantum basis function; a sparse representation module configured to perform sparse representation processing of the seismic data based on the adaptive quantum basis function to obtain expansion coefficients of the signal on the quantum basis; a noise suppression module configured to process the expansion coefficients by using an energy-based threshold processing mechanism to obtain denoised coefficients; a denoised data reconstruction module configured to reconstruct denoised seismic data by linear combination of the quantum basis function using the denoised coefficients to obtain final denoised seismic data.

5. A computer comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the method according to any one of claims 1-3 when executing the computer program.

6. A storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the method according to any one of claims 1-3.

Images