Signal enhancement methods, devices, equipment, media, and products based on high signal-to-noise ratio tag-and-spot (TAS) guns and deep networks.

Abstract

This application provides a signal enhancement method, apparatus, equipment, medium, and product based on high signal-to-noise ratio (SNR) tag-shot and deep network, including: optimizing the design of observation system parameters according to the target geological task; selecting seed shots in noisy areas and generating tag-shot data; performing mathematical transformations on both the tag-shot data and the original data; extracting noise features from the tag-shot data and the transformed multi-domain feature data, and constructing a deep network model using the denoised training data; comparing and analyzing the denoised training data with the original data; inputting the seismic data to be denoised into the trained deep network model for denoising; and monitoring the performance of the deep network model and the denoising effect of the seismic data in real time and adjusting and optimizing the parameters to enhance the effective signal in the seismic data. This application significantly improves the SNR in complex noisy scenarios while effectively preserving the characteristics and waveform integrity of the effective signal, and can be widely used in fields such as oil and gas exploration, geological disaster early warning, and underground structure detection.

Description

Technical Field

[0001] This application relates to the fields of geophysical exploration and seismic data processing technology, and in particular to a signal enhancement method, apparatus, equipment, medium and product based on high signal-to-noise ratio tag-and-shot and depth network. Background Technology

[0002] Seismic exploration, as a core tool of geophysical investigation, plays an irreplaceable role in oil and gas resource development, geological disaster early warning, and underground structure detection. Specifically, it constructs underground media models by artificially generating seismic waves and receiving reflected signals. However, in reality, due to the complex and variable field acquisition environment, seismic data is often affected by environmental noise, instrument noise, and secondary interference waves, causing the effective signals in the seismic data to be submerged by noise. Statistics show that in land exploration, areas with low signal-to-noise ratios can account for more than 30%, which inevitably severely restricts the accuracy of stratigraphic interface identification and the reliability of structural interpretation.

[0003] Currently, traditional seismic denoising techniques mainly rely on manually designed filters based on experience, such as FK filtering (Frequency-Wavenumber Filtering) and wavelet thresholding. While traditional seismic denoising techniques can reduce noise impact to some extent, they also have some significant limitations, as follows:

[0004] (1) Fixed parameter filters are difficult to adapt to the non-stationary characteristics of complex noise, which can easily lead to loss of effective signal;

[0005] (2) Methods based on statistical assumptions are highly dependent on noise distribution, such as PCA (Principal Component Analysis) denoising, whose performance drops sharply in strong non-Gaussian noise scenarios;

[0006] (3) Early machine learning methods (such as support vector machines) rely on manual feature engineering and have insufficient generalization ability. Although deep learning technology has made some progress in seismic denoising in recent years, it still faces many bottlenecks, namely: training depends on ideal labeled data, but noise-free labels are difficult to obtain in actual exploration; the vanishing gradient of the network leads to insufficient extraction of deep features; noise and signal are seriously mixed in the time and frequency domains, and it is difficult to achieve effective separation using only single-domain features.

[0007] In summary, given the technical bottlenecks of traditional seismic denoising methods, such as high degree of human intervention, insufficient environmental adaptability, and poor performance in suppressing complex noise, there is an urgent need to provide a signal enhancement scheme for seismic data to overcome the many limitations of existing technologies. Summary of the Invention

[0008] This application discloses a signal enhancement method, apparatus, device, medium, and product based on high signal-to-noise ratio tag-and-shot and deep network to solve the problem of poor data quality caused by noise interference in seismic data processing.

[0009] In a first aspect, this application discloses a signal enhancement method based on high signal-to-noise ratio tag-and-gun and deep networks, the method comprising:

[0010] Based on the target geological task, the parameters of the observation system are optimized and designed; the observation system parameters include at least the receiver line spacing, the number of receiver lines, the number of per shot channel, and the number of sampling points per channel.

[0011] The observation system is used to determine the noise area of ​​the target work area, a seed gun is selected in the noise area, and target gun data based on the location of the seed gun is generated as tag gun data;

[0012] Based on a preset time-frequency domain transformation rule, both the tag gun data and the original data are mathematically transformed to obtain the transformed multi-domain feature data.

[0013] A codec network structure is used to extract noise features from the tag-and-shot data and the transformed multi-domain feature data to obtain denoised data. Then, a deep network model based on deep residual attention is constructed using the denoised training data.

[0014] The denoising effect of earthquake data is obtained by comparing and analyzing the denoised training data with the original data from the dimensions of energy and waveform.

[0015] The seismic data to be denoised is acquired and input into the trained deep network model for denoising processing to obtain denoised seismic data; the performance of the deep network model and the denoising effect of the seismic data are monitored in real time, and parameters are adjusted and optimized to enhance the effective signals in the seismic data, thereby realizing model deployment and dynamic optimization.

[0016] Preferably, the step of determining the noise area of ​​the target work area using the observation system and selecting a seed cannon in the noise area includes:

[0017] In the target work area, select seed guns in noise areas where the noise level exceeds a set threshold, and select seed guns from the total number of guns in the noise areas according to a set percentage.

[0018] Preferably, the step of generating tag gun data from the target gun data based on the location of the seed gun includes:

[0019] Under the condition that the source excitation parameters are the same at the seed gun position, the excitation is repeated multiple times to obtain multiple gun data at the same position;

[0020] The obtained multi-shot data is vertically superimposed to synthesize single-shot data, and the single-shot data is used as tagged shot data.

[0021] Preferably, the step of generating tag gun data from the target gun data based on the location of the seed gun includes:

[0022] Under the condition of firing the seed cannon once, the noise at different times during the acquisition process is recorded, a noise library is constructed based on the recorded noise, and the noise data at the time of data acquisition for each shot is obtained by spatial interpolation.

[0023] In the target work area, a seed shot is selected in a region with a high signal-to-noise ratio. The noise from the noise library is superimposed on the high signal-to-noise ratio shot data of the seed shot to obtain the tag shot data.

[0024] Preferably, the step of comparing and analyzing the denoised training data with the original data from the dimensions of energy and waveform to obtain the denoising effect of the seismic data includes:

[0025] The denoised training data and the original data are compared and analyzed from the dimensions of energy and waveform. The denoising effect of the seismic data is evaluated based on the mean square error, peak signal-to-noise ratio and structural similarity index.

[0026] Preferably, the step of repeatedly exciting the source under the condition that the source excitation parameters at the seed shot location are the same to obtain multiple shot data at the same location includes:

[0027] The number of repeated excitations is determined based on the severity of the environmental noise, and the number of repeated excitations is inversely proportional to the severity of the environmental noise.

[0028] Preferably, the step of selecting a seed shot in a high signal-to-noise ratio area of ​​the target work area and superimposing the noise from the noise library with the high signal-to-noise ratio shot data of the seed shot to obtain the tagged shot data includes:

[0029] In the target work area, a seed shot is selected in the noise area where the noise exceeds a set threshold. Within a circular area with the seed shot location as the center and a set radius, the seismic data of the adjacent K shot are vertically superimposed to obtain the tag shot data.

[0030] Preferably, the step of vertically stacking seismic data from multiple nearby shots to obtain labeled shot data includes:

[0031] The values ​​of neighboring shots are determined based on the complexity of the surface features, and the values ​​of neighboring shots are inversely proportional to the complexity of the surface features.

[0032] Preferably, the step of performing mathematical transformations on both the tag-gun data and the original data based on preset time-frequency domain transformation rules to obtain transformed multi-domain feature data includes:

[0033] Both the tag gun data and the original data are mathematically transformed based on Fourier transform and / or wavelet transform to obtain multi-domain feature data after transformation and separation of noise and effective signal.

[0034] Secondly, this application discloses a signal enhancement device based on a high signal-to-noise ratio tag-and-gun system and a deep network, the device comprising:

[0035] The observation system optimization module is used to optimize the parameters of the observation system based on the target geological task; the observation system parameters include at least the receiver line spacing, the number of receiver lines, the number of per shot channel, and the number of sampling points per channel;

[0036] The tag gun data construction module is used to determine the noise area of ​​the target work area using the observation system, select seed guns in the noise area, and generate tag gun data based on the target gun data at the location of the seed guns.

[0037] The multi-domain data transformation module is used to perform mathematical transformations on both the tag gun data and the original data based on preset time-frequency domain transformation rules to obtain transformed multi-domain feature data.

[0038] The network model building module is used to extract noise features from the label-based bullet data and the transformed multi-domain feature data using an encoder-decoder network structure, obtain denoised data, and build a deep network model based on deep residual attention using the denoised training data.

[0039] The denoising analysis module is used to compare and analyze the denoised training data with the original data from the dimensions of energy and waveform to obtain the denoising effect of the seismic data.

[0040] The signal enhancement module is used to acquire the seismic data to be denoised and input it into the trained deep network model for denoising processing to obtain denoised seismic data; it monitors the performance of the deep network model and the denoising effect of the seismic data in real time and adjusts and optimizes the parameters to enhance the effective signals in the seismic data, thereby realizing model deployment and dynamic optimization.

[0041] Thirdly, this application discloses an electronic device comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to perform the method as described in any of the preceding aspects.

[0042] Fourthly, this application discloses a non-transitory computer-readable storage medium in which, when the instructions in the storage medium are executed by a processor of an electronic device, enable the electronic device to perform the methods described in any of the preceding aspects.

[0043] Fifthly, this application discloses a computer program product in which, when the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to perform the method described in any of the preceding aspects.

[0044] The technical solution provided in this application may include the following beneficial effects:

[0045] This application's solution significantly improves the signal-to-noise ratio (SNR) in complex noise scenarios through high SNR tag-shot construction and deep residual attention network optimization, while effectively preserving the characteristics and waveform integrity of the effective signal, thus enhancing the denoising effect and signal fidelity of seismic data. Especially in complex noise environments, the tag-shot generation mechanism can dynamically simulate real noise characteristics, and combined with multi-domain feature extraction technology, effectively separates noise features from signal features. The deep residual attention network achieves high-precision denoising through layer-by-layer noise stripping and signal reconstruction, while effectively preserving the dynamic characteristics and waveform integrity of seismic waves. Moreover, through quantitative evaluation of mean square error, peak SNR, and structural similarity, the denoised data outperforms traditional solutions in both global quality and structural fidelity. The dynamic optimization system further enhances the model's robustness and adaptability in complex exploration environments. The modular design of this application supports seamless integration with existing seismic processing systems, with flexible parameter configuration, providing high-quality seismic data support for fields such as oil and gas exploration, geological disaster early warning, and underground structure detection. It possesses significant engineering practical value and promotion potential, and can be widely applied in fields such as oil and gas exploration, geological disaster early warning, and underground structure detection. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating the signal enhancement method based on high signal-to-noise ratio tag-and-spot (TAS) guns and deep networks provided in this application;

[0049] Figure 2 To generate single-shot data;

[0050] Figure 3 The data from ten superimposed shots were generated at the same location;

[0051] Figure 4 This refers to data collected from a single shot during a single firing.

[0052] Figure 5 To apply the signal enhancement method based on high signal-to-noise ratio tag-and-gun and deep network provided in this application, Figure 4 Denoising-reduced single-gun data;

[0053] Figure 6 A structural diagram of a signal enhancement device based on a high signal-to-noise ratio tag-and-gun and a deep network provided in this application;

[0054] Figure 7 A block diagram of an electronic device provided in this application;

[0055] Figure 8 A block diagram of another electronic device provided in this application. Detailed Implementation

[0056] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] In geological exploration, high-quality seismic data is crucial for analyzing geological structures. However, raw data is often susceptible to interference from environmental noise (such as mechanical vibration and wind noise) and instrument noise. Therefore, addressing the technical bottlenecks of traditional seismic denoising methods, such as high degree of human intervention, insufficient environmental adaptability, and poor suppression of complex noise, this application proposes a signal enhancement scheme based on high signal-to-noise ratio tag-shot and deep network. This scheme provides a denoising framework that integrates dynamic tag-shot generation and adaptive optimization of the deep network to solve the problem of poor data quality caused by noise interference in seismic data processing.

[0058] The following is combined Figures 1 to 5 The signal enhancement method based on high signal-to-noise ratio tag-and-shot array and deep network provided in this application is described.

[0059] Example 1

[0060] Reference Figure 1 This is a flowchart of a signal enhancement method based on a high signal-to-noise ratio tag-and-gun and a deep network provided in this application. This method can be applied to electronic devices, and specifically includes the following steps:

[0061] Step S101: Based on the target geological task, optimize the design of the observation system parameters; specifically, the observation system parameters include at least the receiver line spacing L meters, the number of receiver lines S, the number of samples per shot M, and the number of sampling points per shot N. It should be noted that the receiver line spacing, the number of receiver lines, the number of samples per shot, and the number of sampling points per shot are preferred combinations of the observation system parameters selected in this application. This application may also include other methods for setting observation system parameters.

[0062] Step S102: Use the observation system to determine the noise area of ​​the target work area, select a seed gun in the noise area, and generate tag gun data based on the target gun data at the location of the seed gun.

[0063] Specifically, before data collection, the field collection environment is investigated to identify areas with high noise levels. Seed cannon locations are then selected in these areas, and the high-noise areas can be determined based on a set noise threshold.

[0064] In step S102, seed guns can be selected in the noise area as follows: Seed guns are selected in the noise area of ​​the target work area where the noise level exceeds a set threshold, and a seed gun is selected from the total number of guns in the noise area according to a set percentage. Preferably, the default value of the set percentage δ is set to 100%. In practical applications, this value can be adjusted according to the actual situation to select a suitable seed gun.

[0065] It should be noted that seed shot points are selected in high-noise areas, and high signal-to-noise ratio labeled shots are generated by repeated firing and vertical superposition, or candidate labeled data are constructed by spatial interpolation of the noise library and superposition with neighboring shots, as follows:

[0066] In one implementation, under the condition that the source excitation parameters at the seed gun location are the same, the excitation is repeated multiple times to obtain multi-shot data at the same location. The obtained multi-shot data is then vertically superimposed to synthesize single-shot data, which is used as the tag gun data. In this method, the number of repeated excitations is determined based on the severity of environmental noise, and the number of repeated excitations is inversely proportional to the severity of environmental noise. Specifically, excitation at the same location can obtain S-shot data. If the environmental noise is severe, the value of S can be increased; conversely, if the environmental noise is not severe, the value of S can be decreased. Preferably, the number of repeated excitations S is defaulted to 10 times.

[0067] In another implementation, under the condition of one excitation at the seed shot location, noise at different times during the acquisition process is recorded. A noise library is constructed based on the recorded noise, and spatial interpolation is used to obtain the noise data at the time of each shot's data acquisition. Seed shots are selected in areas of high signal-to-noise ratio in the target work area, and the noise from the noise library is superimposed on the high signal-to-noise ratio shot data of the seed shots to obtain labeled shot data. Specifically, seed shots are selected in noise areas of the target work area where the noise exceeds a set threshold. Within a circular area centered on the seed shot location and with a set radius, seismic data from K neighboring shots are vertically superimposed to obtain labeled shot data. In this method, the values ​​of neighboring shots are determined based on the complexity of the surface features, and the value of neighboring shots is inversely proportional to the complexity of the surface features. Preferably, the default number of neighboring shots K is 3.

[0068] Step S103: Based on the preset time-frequency domain transformation rules, both the tag gun data and the original data are mathematically transformed to obtain the transformed multi-domain feature data.

[0069] In one scenario, both the tag-and-shoot data and the original data can be mathematically transformed based on Fourier transform and / or wavelet transform to separate the time-frequency characteristics of noise and effective signals, resulting in multi-domain feature data after transformation and separation of noise and effective signals. It should be noted that this application does not limit the specific mathematical transformation method for the multi-domain feature data.

[0070] Step S104: Use an encoder-decoder network structure to extract noise features from the labeled data and the transformed multi-domain feature data to obtain denoised data, and use the denoised training data to build a deep network model based on deep residual attention.

[0071] This deep network model accepts multiple feature inputs. The network model in this application can be based on autoencoders (AEs), denoising autoencoders (DAEs), and U-Net for computer vision semantic segmentation. By introducing an attention module, the network's focus on important regions is enhanced. Furthermore, a ResNet module is used to construct the deep network model, incorporating ResNet's residual structure and attention mechanism at each layer. Through layer-by-layer fine-tuning of features, the impact of noise is gradually reduced, avoiding the vanishing gradient problem in deep networks. It should be noted that a deep residual attention network based on an encoder-decoder architecture is constructed, incorporating ResNet residual modules to alleviate the vanishing gradient problem, and a channel-space dual attention mechanism is introduced to dynamically enhance effective signal features, achieving layer-by-layer noise stripping and high-fidelity signal reconstruction.

[0072] Step S105: Compare and analyze the denoised training data with the original data from the dimensions of energy and waveform to obtain the denoising effect of the earthquake data;

[0073] Step S106: Obtain the seismic data to be denoised and input it into the trained deep network model for denoising processing to obtain denoised seismic data; monitor the performance of the deep network model and the denoising effect of the seismic data in real time, and adjust and optimize the parameters to enhance the effective signals in the seismic data and realize model deployment and dynamic optimization.

[0074] In step S105, the denoised training data and the original data are compared and analyzed from the dimensions of energy and waveform. The denoising effect of the seismic data is evaluated based on mean square error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM). Specifically, the evaluation criteria may include mean square error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM). MSE is mainly used to quantify the difference between the denoised result and the original noise-free data; PSNR is mainly used to evaluate the quality of the denoised image; and SSIM mainly measures the structural similarity between the denoised image and the original image.

[0075] It is important to emphasize that this application constructs label data that approximates the real scene by using a dynamic label generation mechanism and employing repeated excitation vertical superposition or noise library interpolation techniques in high-noise areas, thus solving the problem of missing training data. A multi-domain feature fusion strategy is designed, combining mathematical tools such as Fourier transform and wavelet transform to achieve decoupling of the time-frequency features of noise and signal. A deep residual attention network is constructed, integrating ResNet residual modules and a channel-space dual attention mechanism. Through gradient stabilization and dynamic feature weighting, noise is stripped layer by layer and the signal is reconstructed with high fidelity.

[0076] Please see Figures 2 to 5 Experiments show that this application significantly improves the signal-to-noise ratio in several typical seismic exploration areas, demonstrating significant improvements in both signal fidelity and waveform structure integrity compared to traditional filtering methods. In practical applications in areas with strong interference, fault identification accuracy is greatly enhanced, validating its engineering applicability. This technology provides an innovative solution for high-precision seismic exploration under complex geological conditions, and is of great value in promoting the efficient development of oil and gas resources and the prevention and control of geological disasters.

[0077] This application's solution significantly improves the signal-to-noise ratio (SNR) in complex noise scenarios through high SNR tag-shot construction and deep residual attention network optimization, while effectively preserving the characteristics and waveform integrity of the effective signal, thus enhancing the denoising effect and signal fidelity of seismic data. Especially in complex noise environments, the tag-shot generation mechanism can dynamically simulate real noise characteristics, and combined with multi-domain feature extraction technology, effectively separates noise features from signal features. The deep residual attention network achieves high-precision denoising through layer-by-layer noise stripping and signal reconstruction, while effectively preserving the dynamic characteristics and waveform integrity of seismic waves. Moreover, through quantitative evaluation of mean square error, peak SNR, and structural similarity, the denoised data outperforms traditional solutions in both global quality and structural fidelity. The dynamic optimization system further enhances the model's robustness and adaptability in complex exploration environments. The modular design of this application supports seamless integration with existing seismic processing systems, with flexible parameter configuration, providing high-quality seismic data support for fields such as oil and gas exploration, geological disaster early warning, and underground structure detection. It possesses significant engineering practical value and promotion potential, and can be widely applied in fields such as oil and gas exploration, geological disaster early warning, and underground structure detection.

[0078] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions involved are not necessarily required by this application.

[0079] This invention focuses on the construction and effective signal enhancement of high signal-to-noise ratio (SNR) tag-and-spot (TAS) systems, integrating dynamic tag construction, multi-domain feature fusion, and deep network optimization techniques to achieve high-precision noise suppression. A specific example is provided below to illustrate the signal enhancement based on high SNR tag-and-spot systems and deep networks provided in this application.

[0080] Step (1) Based on the geological task, the observation system was designed with a receiver line spacing of 200m, a channel spacing of 40m, and 42 receiver lines, with 480 channels per line. Single-shot seismic data was excited and collected and recorded, with 4000 sampling points per channel and a sampling interval of 1ms.

[0081] Step (2) Before data collection, investigate the field collection environment and determine the area with high noise in the field environment. Select the seed gun position in the area with high noise. Repeat the excitation 10 times at the selected seed gun position under the same source excitation parameters. 10 shots can be obtained at the same position. Vertically superimpose the 10 shots to synthesize one shot data. This shot data can be used as a tag shot. Figure 2 To generate single-shot data, Figure 3 By generating ten superimposed single-shot data at the same location, it can be seen that the signal-to-noise ratio of the first-arrival data is significantly improved after superposition.

[0082] Step (3) Perform Fourier transform and wavelet transform on both the tag gun data and the original data obtained in step (2) to obtain the transformed data.

[0083] The network model designed in step (4) has multiple feature inputs and uses an encoder-decoder network structure to extract noise features from the labeled data and the transformed data, and the output is the denoised data. The residual structure and attention mechanism of ResNet are incorporated into each layer of the network, and the influence of noise is gradually weakened by finely adjusting the features layer by layer.

[0084] Step (5) evaluates and compares the denoised data obtained in step (4) with the original data, mainly analyzing energy and waveform, and the evaluation criteria include mean square error, peak signal-to-noise ratio and structural similarity.

[0085] Step (6) Based on steps (4) and (5), a trained deep network model can be obtained. The data to be denoised is processed according to the trained deep network model to obtain the denoised seismic data.

[0086] Furthermore, as can be seen from the attached diagram, Figure 4 The data is from a single shot acquired during a single excitation, and the signal-to-noise ratio of the first arrival data is low. Figure 5 To apply the signal enhancement method based on high signal-to-noise ratio tag-and-gun and deep network provided in this application, Figure 4 After denoising the single-shot data, it is clear that the solution provided in this application effectively suppresses random noise and significantly improves the quality of the initial arrival data.

[0087] Example 2

[0088] Reference Figure 6 This is a structural diagram of a signal enhancement device based on a high signal-to-noise ratio tag-and-gun and a deep network provided in this application. The device includes:

[0089] The observation system optimization module 210 is used to optimize the parameters of the observation system according to the target geological task; the observation system parameters include at least the receiver line spacing, the number of receiver lines, the number of per shot channel, and the number of sampling points per channel;

[0090] Tag-fired gun data construction module 220 is used to determine the noise area of ​​the target work area using the observation system, select seed guns in the noise area, and generate tag-fired gun data based on the target gun data at the location of the seed guns.

[0091] The multi-domain data transformation module 230 is used to perform mathematical transformations on both the tag gun data and the original data based on a preset time-frequency domain transformation rule to obtain transformed multi-domain feature data.

[0092] The network model building module 240 is used to extract noise features from the label-based bullet data and the transformed multi-domain feature data using an encoding-decoding network structure, obtain denoised data, and build a deep network model based on deep residual attention using the denoised training data.

[0093] The denoising analysis module 250 is used to compare and analyze the denoised training data with the original data from the dimensions of energy and waveform to obtain the denoising effect of the seismic data.

[0094] The signal enhancement module 260 is used to acquire the seismic data to be denoised and input it into the trained deep network model for denoising processing to obtain denoised seismic data; it monitors the performance of the deep network model and the denoising effect of the seismic data in real time and adjusts and optimizes the parameters to enhance the effective signals in the seismic data and realize model deployment and dynamic optimization.

[0095] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0096] Example 3

[0097] Optionally, this application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0098] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0099] Figure 7 This application provides a block diagram of an electronic device 800. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0100] Reference Figure 7 The electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0101] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0102] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, images, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0103] Power supply component 806 provides power to various components of electronic device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.

[0104] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0105] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0106] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0107] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 may detect the on / off state of device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0108] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast operation information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0109] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0110] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of an electronic device 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0111] Example 4

[0112] Figure 8 A block diagram of another electronic device 1900 provided for this application. For example, electronic device 1900 may be provided as a server.

[0113] Reference Figure 8 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0114] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output (I / O) interface 1958. Electronic device 1900 can operate on an operating system stored in memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0115] Example 5

[0116] Fifthly, this application discloses a computer program product in which, when the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to perform the method described in any of the preceding aspects.

[0117] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0119] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0120] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0121] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0122] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0123] 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.

[0124] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0125] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0126] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A signal enhancement method based on high signal-to-noise ratio tag gun and deep network, characterized in that, The method comprises: According to the target geological task, the observation system parameters are optimized and designed; the observation system parameters at least include receiving line distance, receiving line number, number of traces per shot, and number of sampling points per trace; Determine the noise area of the target work area using the observation system, select a seed shot in the noise area, and generate tag shot data based on the target shot data at the location of the seed shot; Based on the preset time-frequency domain transformation rule, the tag shot data and the original data are both mathematically transformed to obtain transformed multi-domain feature data; Using an encoding-decoding network structure, noise features are extracted from the tag shot data and the transformed multi-domain feature data to obtain denoised data, and a deep network model based on deep residual attention is constructed using the denoised training data; Compare and analyze the denoised training data and the original data from the energy and waveform dimensions to obtain the denoising effect of the seismic data; Obtain the seismic data to be denoised and input the deep network model to perform denoising processing to obtain the denoised seismic data; monitor the performance of the deep network model and the denoising effect of the seismic data in real time and perform parameter adjustment and optimization to enhance the effective signal in the seismic data, realize model deployment and dynamic optimization.

2. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 1, characterized in that, The step of determining the noise area of the target work area using the observation system and selecting a seed shot in the noise area comprises: Select a seed shot in the noise area where the noise is greater than a set threshold, and select a seed shot from the total number of shots in the noise area according to a set percentage.

3. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 2, characterized in that, The step of generating tag shot data based on the target shot data at the location of the seed shot comprises: Under the condition that the source excitation parameters at the location of the seed shot are the same, repeatedly excite multiple times to obtain multi-shot data at the same location; Vertically stack the obtained multi-shot data to synthesize single-shot data, and use the single-shot data as the tag shot data.

4. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 2, characterized in that, The step of generating tag shot data based on the target shot data at the location of the seed shot comprises: Under the condition of one excitation at the location of the seed shot, record the noise at different times during the acquisition process, construct a noise library based on the recorded noise, and obtain noise data for each shot data acquisition using spatial interpolation; Select a seed shot in the area with high signal-to-noise ratio in the target work area, superimpose the noise in the noise library with the high signal-to-noise ratio shot data of the seed shot to obtain the tag shot data.

5. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 2, characterized in that, The step of comparing and analyzing the denoised training data and the original data from the energy and waveform dimensions to obtain the denoising effect of the seismic data comprises: Compare and analyze the denoised training data and the original data from the energy and waveform dimensions, and evaluate the denoising effect of the seismic data based on mean square error, peak signal-to-noise ratio, and structural similarity index.

6. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 3, characterized in that, The step of repeatedly exciting multiple times under the condition that the source excitation parameters at the location of the seed shot are the same to obtain multi-shot data at the same location comprises: Determine the number of repeated excitations based on the severity of environmental noise, and the number of repeated excitations is inversely proportional to the severity of environmental noise.

7. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 4, characterized in that, The step of selecting seed shots in the area of the target work area with high signal-to-noise ratio and superimposing the noise of the noise library on the high signal-to-noise ratio shot data of the seed shots to obtain the labeled shot data comprises: In the target work area, selecting seed shots in the noise area where the noise is greater than a set threshold, and selecting seismic data of adjacent K shots in a circular area with the seed shot position as the center and a set radius to obtain labeled shot data through vertical stacking.

8. The signal enhancement method based on high signal-to-noise ratio tag shot and deep network according to claim 7, characterized in that, The step of selecting seismic data of multiple adjacent shots to obtain labeled shot data through vertical stacking comprises: The number of adjacent shots is determined based on the complexity of the surface feature, and the number of adjacent shots is inversely proportional to the complexity of the surface feature.

9. The signal enhancement method based on high signal-to-noise ratio taggun and deep network according to any one of claims 1 to 8, characterized in that, The step of performing mathematical transformation on the labeled shot data and the original data based on a preset time-frequency domain transformation rule to obtain transformed multi-domain feature data comprises: The labeled shot data and the original data are both mathematically transformed based on Fourier transform and / or wavelet transform to obtain multi-domain feature data after transformation and separation of noise and effective signal.

10. A signal enhancement device based on high signal-to-noise ratio tag shot and deep network, characterized in that, The device comprises: An observation system optimization module for optimizing and designing observation system parameters according to a target geological task, wherein the observation system parameters at least include receiving line distance, receiving line number, number of shots per channel, and number of sampling points per channel; A labeled shot data construction module for determining a noise area of a target work area using the observation system, selecting seed shots in the noise area, and generating target shot data based on the position of the seed shots as labeled shot data; A multi-domain data transformation module for performing mathematical transformation on the labeled shot data and the original data based on a preset time-frequency domain transformation rule to obtain transformed multi-domain feature data; A network model construction module for extracting noise features from labeled shot data and transformed multi-domain feature data using an encoding-decoding network structure to obtain denoised data, and constructing a deep network model based on deep residual attention using the denoised training data; A denoising analysis module for comparing and analyzing denoised training data and original data from energy and waveform dimensions to obtain the denoising effect of seismic data; A signal enhancement module for obtaining seismic data to be denoised and inputting the deep network model obtained through training to perform denoising processing to obtain denoised seismic data, and for real-time monitoring of the performance of the deep network model and the denoising effect of the seismic data and parameter adjustment and optimization to enhance effective signals in the seismic data, thereby realizing model deployment and dynamic optimization.

11. An electronic device, comprising: It comprises: A processor, a memory, and a computer program stored on the memory and executable on the processor, wherein the computer program is executed by the processor to implement the method of any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium and is executed by the processor to implement the method of any one of claims 1 to 9.

13. A computer program product, characterised in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device implements the method of any one of claims 1 to 9.

Images