A method and system for scattering wave imaging using deep learning
Through the deep learning neural network model, the scattered imaging data is directly extracted from the imaging data, which solves the problems of low accuracy and low efficiency of scattered wave imaging in the prior art, and efficient and automated scattered imaging is achieved, which improves the data support capabilities of seismic exploration.
Patent Information
- Application Number
- CN202011209057.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-11-03
- Publication Date
- 2025-07-04
- Estimated Expiration
- 2040-11-03
AI Technical Summary
The existing scattered wave imaging methods have problems such as low accuracy, low efficiency and insufficient automation in seismic exploration, and it is difficult to effectively identify scattering bodies such as faults and cracks.
The deep learning neural network model is adopted to generate training samples through forward simulation and offset imaging, and the training model is used to map input data to output data, extract scattered imaging data directly from the imaging data, and add it with the original scattered imaging data to improve accuracy and efficiency.
It realizes highly automated scattering imaging, significantly improves scattering imaging accuracy and efficiency, and provides strong data support for seismic exploration.
Smart Images
Figure CN114442163B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of exploration seismic data scattering imaging and ultrasonic data scattering imaging, and particularly to a method and system for scattering wave imaging using deep learning. Background Art
[0002] According to Huygens' principle, the seismic signals received on the surface in seismic exploration are the interference superposition of seismic waves generated by the vibrations of all points underground at the geophone. According to the size of the underground structural interface, the collected seismic waves can be divided into reflected waves and scattered waves. When the interface size is larger than the Fresnel zone, reflected waves are generated, and when the interface size is smaller than the Fresnel zone, scattered waves are generated.
[0003] Currently, conventional seismic imaging processing mainly uses reflected wave signals to reflect large-scale structural forms and velocity information. Scattered waves are markers of structures and lithological anomalies. Scattered waves are generated when there are discontinuities underground, such as faults, fractures, formation annihilation points, reef blocks, karst caves, weathered crusts, etc. In areas with extremely strong heterogeneity such as volcanic rocks and carbonate rocks, these scatterers are important exploration targets. Since the energy of the scattered waves generated by scattered anomaly bodies is usually one order of magnitude lower than the reflected energy, even if accurately imaged, the scattered information is submerged by the strong-energy reflected waves and the scatterers beneficial to oil and gas accumulation cannot be identified. There is an urgent need for an imaging method that highlights scattering targets.
[0004] Among them, conventional scattering imaging methods can be divided into three categories. One is the pre-imaging scattering imaging method. Based on the difference between the reflection wave travel-time curve and the scattered wave travel-time curve in the pre-stack gather, methods such as Radon transform, KL transform, plane wave decomposition, focusing and defocusing are used to separate the scattered waves, and then the scattered waves are imaged. In the actual pre-stack data of seismic data, due to the low signal-to-noise ratio, it is very difficult to identify and separate the scattered waves, which affects the application effect of this type of method. The second is the in-imaging scattering imaging method. By setting a weighting factor, the reflected energy is directly suppressed during the imaging process. However, since the underground medium conditions are not known in advance, it is relatively difficult to set the weighting factor. The third is to separate the scattered waves from the post-imaging scattering angle gather. In the scattering angle gather, the reflected energy is concentrated at a certain angle, while the scattered energy is dispersed at various angles. By means of median filtering and other means, the reflected energy in the scattering angle gather is filtered out, and the processed gather is stacked to obtain the scattering imaging profile. This method theoretically has high precision. The main problems are that the generation of the scattering angle gather requires a large amount of computing resources, and the quality of the angle gather is not high. The median filtering cannot completely remove the reflected energy and is prone to damaging the scattered energy.
[0005] In summary, the existing scattered wave imaging methods all have certain deficiencies, and there is an urgent need for a scattering imaging technical solution that can overcome the above deficiencies and improve the exploration efficiency, accuracy and automation degree. Summary of the Invention
[0006] To overcome the deficiencies of the existing technologies, the present invention proposes a method and system for scattering wave imaging using deep learning. Based on a deep learning neural network, a large amount of model data is generated through forward modeling using the Born approximation of the wave equation for training, so as to directly extract scattering imaging from the imaging data, improve the accuracy and efficiency of scattering imaging, and provide strong data support for seismic exploration.
[0007] In the first aspect of the embodiments of the present invention, a method for scattering wave imaging using deep learning is proposed. The method includes:
[0008] Obtain imaging data, and extract original scattering imaging data from the imaging data;
[0009] Establish a deep learning neural network model;
[0010] Through forward modeling and migration imaging, obtain training samples for training the deep learning neural network model;
[0011] Use the training samples to train the deep learning neural network model to perform mapping from input data to output data, where the input data includes imaging data of reflected and scattered waves, and the output data is scattering wave imaging data;
[0012] Use the trained deep learning neural network model to process actual imaging data to obtain scattering imaging data;
[0013] Multiply the scattering imaging data by a scale factor and add it to the original scattering imaging data to obtain a scattering wave imaging result.
[0014] Further, obtaining the training samples for training the deep learning neural network model through forward modeling and migration imaging includes:
[0015] Make a scattering wave reflection coefficient model and a reflected wave reflection coefficient model;
[0016] Add the scattering wave reflection coefficient model and the reflected wave reflection coefficient model to obtain a total reflection coefficient model;
[0017] Set a background velocity field;
[0018] According to the scattering wave reflection coefficient model and the background velocity field, use the wave equation Born approximation forward modeling method for processing to obtain a first pre-stack simulation record of scattering waves;
[0019] According to the total reflection coefficient model and the background velocity field, use the wave equation Born approximation forward modeling method for processing to obtain a second pre-stack simulation record of scattering waves;
[0020] Based on the first pre-stack simulated scattering wave records and the background velocity field, migration imaging is performed to obtain scattering wave imaging data;
[0021] Based on the second pre-stack simulated scattering wave records and the background velocity field, migration imaging is performed to obtain imaging data including reflected and scattering waves;
[0022] Based on the scattering wave imaging data and the imaging data including reflected and scattering waves, training samples for training a deep learning neural network model are generated.
[0023] Furthermore, the method further includes:
[0024] Model making, forward simulation, and migration imaging are repeated to generate multiple scattering wave imaging data and multiple imaging data including reflected and scattering waves as training samples for training a deep learning neural network model.
[0025] Furthermore, using the training samples to train a deep learning neural network model for mapping input data to output data includes:
[0026] Multiple scattering wave imaging data and multiple imaging data including reflected and scattering waves are input into the deep learning neural network model for mapping input data to output data, and multiple hidden layers are set, and the deep learning neural network model is trained through multiple error backpropagations and iterations.
[0027] In the second aspect of the embodiments of the present invention, a system for scattering wave imaging using deep learning is proposed. The system includes:
[0028] A data acquisition module for acquiring imaging data and extracting original scattering imaging data from the imaging data;
[0029] A model establishment module for establishing a deep learning neural network model;
[0030] A forward simulation and migration imaging module for obtaining training samples for training a deep learning neural network model through forward simulation and migration imaging;
[0031] A training module for training a deep learning neural network model using the training samples to perform mapping from input data to output data, where the input data includes imaging data of reflected and scattering waves, and the output data is scattering wave imaging data;
[0032] A data processing module for processing actual imaging data using the trained deep learning neural network model to obtain scattering imaging data;
[0033] A calculation module for multiplying the scattering imaging data by a proportionality coefficient and adding it to the original scattering imaging data to obtain a scattering wave imaging result.
[0034] Furthermore, the forward modeling and migration imaging module includes:
[0035] A model making unit, configured to make a scattered wave reflection coefficient model and a reflected wave reflection coefficient model, and add the scattered wave reflection coefficient model and the reflected wave reflection coefficient model to obtain a total reflection coefficient model;
[0036] A setting unit, configured to set a background velocity field;
[0037] A first forward modeling unit, configured to process according to the scattered wave reflection coefficient model and the background velocity field by using the wave equation Born approximation forward modeling method to obtain a first pre-stack simulation record of scattered waves;
[0038] A second forward modeling unit, configured to process according to the total reflection coefficient model and the background velocity field by using the wave equation Born approximation forward modeling method to obtain a second pre-stack simulation record of scattered waves;
[0039] A first migration imaging unit, configured to perform migration imaging according to the first pre-stack simulation record of scattered waves and the background velocity field to obtain scattered wave imaging data;
[0040] A second migration imaging unit, configured to perform migration imaging according to the second pre-stack simulation record of scattered waves and the background velocity field to obtain imaging data including reflections and scattered waves;
[0041] A training sample generation unit, configured to generate training samples for training a deep learning neural network model according to the scattered wave imaging data and the imaging data including reflections and scattered waves.
[0042] Furthermore, the forward modeling and migration imaging module is further configured to:
[0043] Repeat model making, forward modeling and migration imaging to generate multiple scattered wave imaging data and multiple imaging data including reflections and scattered waves as training samples for training a deep learning neural network model.
[0044] Furthermore, the training module is specifically configured to:
[0045] Input multiple scattered wave imaging data and multiple imaging data including reflections and scattered waves into the deep learning neural network model, perform mapping from input data to output data, set multiple hidden layers, and train the deep learning neural network model through multiple error backpropagations and iterations.
[0046] In a third aspect of the embodiments of the present invention, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, a method for scattering wave imaging using deep learning is implemented.
[0047] In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, a method for scattering wave imaging using deep learning is implemented.
[0048] The method and system for scattering wave imaging using deep learning proposed by the present invention obtain imaging data, extract original scattered imaging data from the imaging data; establish a deep learning neural network model; obtain training samples for training the deep learning neural network model through forward modeling and migration imaging, where the input data includes imaging data of reflected and scattered waves, and the output data is scattered wave imaging data; use the training samples to train the deep learning neural network model to perform mapping from input data to output data; use the trained deep learning neural network model to process actual imaging data to obtain scattered imaging data; multiply the scattered imaging data by a scale factor and add it to the original scattered imaging data to obtain a scattered wave imaging result, realizing the direct extraction of scattered imaging from imaging data, with high automation and effectively improving the accuracy and efficiency of scattered imaging, providing strong data support for seismic exploration. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0050] Figure 1 It is a schematic flowchart of a method for scattering wave imaging using deep learning according to an embodiment of the present invention.
[0051] Figure 2 It is a schematic flowchart of forward modeling and migration imaging according to a specific embodiment of the present invention.
[0052] Figure 3 It is a schematic diagram of the system architecture for scattering wave imaging using deep learning according to an embodiment of the present invention.
[0053] Figure 4 It is a schematic diagram of the architecture of a forward modeling and migration imaging module according to a specific embodiment of the present invention.
[0054] Figure 5Schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed implementation manners
[0055] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and then implement the present invention, rather than limiting the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to be able to fully convey the scope of the present disclosure to those skilled in the art.
[0056] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, a device, an equipment, a method, or a computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: completely hardware, completely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.
[0057] According to an embodiment of the present invention, a method and a system for scattering wave imaging using deep learning are proposed.
[0058] Below, with reference to several representative embodiments of the present invention, the principles and spirit of the present invention will be elaborated in detail.
[0059] Figure 1 Schematic diagram of the method flow for scattering wave imaging using deep learning according to an embodiment of the present invention. As Figure 1 shown, the method includes:
[0060] Step S101, obtaining imaging data and extracting original scattered imaging data from the imaging data;
[0061] Step S102, establishing a deep learning neural network model;
[0062] Step S103, obtaining training samples for training the deep learning neural network model through forward modeling and migration imaging;
[0063] Step S104, training the deep learning neural network model using the training samples to perform mapping from input data to output data, where the input data includes imaging data of reflected and scattered waves, and the output data is scattered wave imaging data;
[0064] Step S105, processing actual imaging data using the trained deep learning neural network model to obtain scattered imaging data;
[0065] Step S106, multiplying the scattered imaging data by a scale factor and adding it to the original scattered imaging data to obtain a scattered wave imaging result.
[0066] For a clearer explanation of the above method for scattering wave imaging using deep learning, the following will be described in conjunction with each step.
[0067] Step S101:
[0068] Directly extract scattering imaging from the imaging data based on a deep learning neural network.
[0069] Step S102:
[0070] Establish a deep learning neural network model.
[0071] Step S103:
[0072] Refer to Figure 2 , which is a schematic flow diagram of the forward modeling and migration imaging of a specific embodiment of the present invention. As Figure 2 shown, the detailed processes of forward modeling and migration imaging are as follows:
[0073] Step S1031, fabricate a scattering wave reflection coefficient model m1 and a reflected wave reflection coefficient model m2;
[0074] Step S1032, add the scattering wave reflection coefficient model m1 and the reflected wave reflection coefficient model m2 to obtain a total reflection coefficient model m;
[0075] Step S1033, set the background velocity field V;
[0076] Step S1034, according to the scattering wave reflection coefficient model m1 and the background velocity field V, use the wave equation Born approximation forward modeling method to process and obtain a pre-stack simulation record P1 of the scattering wave;
[0077] Step S1035, according to the total reflection coefficient model m and the background velocity field V, use the wave equation Born approximation forward modeling method to process and obtain a pre-stack simulation record P of the scattering wave;
[0078] Step S1036, according to the pre-stack simulation record P1 of the scattering wave and the background velocity field V, perform migration imaging to obtain scattering wave imaging data I1;
[0079] Step S1037, according to the pre-stack simulation record P of the scattering wave and the background velocity field V, perform migration imaging to obtain imaging data I including reflections and scattering waves;
[0080] Step S1038, generate training samples for training the deep learning neural network model according to the scattering wave imaging data I1 and the imaging data I including reflections and scattering waves.
[0081] Furthermore, the process of step S1038 also includes:
[0082] Repeat the model making, forward simulation, and migration imaging of steps S1031 to S1037 to generate multiple scattered wave imaging data I1 and multiple imaging data I including reflection and scattered waves as training samples for training a deep learning neural network model.
[0083] Step S104:
[0084] Input the multiple scattered wave imaging data and the multiple imaging data including reflection and scattered waves into the deep learning neural network model for mapping from input data to output data, set multiple hidden layers, and train the deep learning neural network model through multiple error backpropagations and iterations.
[0085] Step S105:
[0086] Use the trained deep learning neural network model to process the actual imaging data to obtain scattered wave imaging data;
[0087] Step S106:
[0088] Multiply the scattered wave imaging data by a scale factor and add it to the original scattered wave imaging data to obtain a scattered wave imaging result.
[0089] It should be noted that although the operations of the method of the present invention are described in a specific order in the above embodiments and the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the shown operations must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.
[0090] After introducing the method of the exemplary embodiment of the present invention, next, refer to Figures 3 to 4 to introduce the system for scattered wave imaging using deep learning of the exemplary embodiment of the present invention.
[0091] The implementation of the system for scattered wave imaging using deep learning can refer to the implementation of the above method, and the repeated parts will not be elaborated. The terms "module" or "unit" used hereinafter may be a combination of software and / or hardware for implementing a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.
[0092] Based on the same inventive concept, the present invention also proposes a system for scattered wave imaging using deep learning, as Figure 3 shown, the system includes:
[0093] A data acquisition module 310, configured to acquire imaging data and extract original scattered wave imaging data from the imaging data;
[0094] The model establishment module 320 is used to establish a deep learning neural network model;
[0095] The forward modeling and migration imaging module 330 is used to obtain training samples for training the deep learning neural network model through forward modeling and migration imaging;
[0096] The training module 340 is used to train the deep learning neural network model with the training samples to perform the mapping from input data to output data, where the input data includes the imaging data of reflected and scattered waves, and the output data is the scattered wave imaging data;
[0097] The data processing module 350 is used to process the actual imaging data with the trained deep learning neural network model to obtain the scattered imaging data;
[0098] The calculation module 360 is used to multiply the scattered imaging data by a scale factor and add it to the original scattered imaging data to obtain the scattered wave imaging result.
[0099] In a specific embodiment, refer to Figure 4 , which is a schematic diagram of the architecture of the forward modeling and migration imaging module in a specific embodiment of the present invention. As Figure 4 shown, the forward modeling and migration imaging module 330 includes:
[0100] The model production unit 331 is used to produce the scattered wave reflection coefficient model and the reflected wave reflection coefficient model, and add the scattered wave reflection coefficient model and the reflected wave reflection coefficient model to obtain the total reflection coefficient model;
[0101] The setting unit 332 is used to set the background velocity field;
[0102] The first forward modeling unit 333 is used to process according to the scattered wave reflection coefficient model and the background velocity field by using the wave equation Born approximation forward modeling method to obtain the first scattered wave pre-stack simulation record;
[0103] The second forward modeling unit 334 is used to process according to the total reflection coefficient model and the background velocity field by using the wave equation Born approximation forward modeling method to obtain the second scattered wave pre-stack simulation record;
[0104] The first migration imaging unit 335 is used to perform migration imaging according to the first scattered wave pre-stack simulation record and the background velocity field to obtain the scattered wave imaging data;
[0105] The second migration imaging unit 336 is used to perform migration imaging according to the second scattered wave pre-stack simulation record and the background velocity field to obtain the imaging data including reflected and scattered waves;
[0106] A training sample generation unit 337 is configured to generate training samples for training a deep learning neural network model based on the scattered wave imaging data and the imaging data including reflected and scattered waves.
[0107] In another embodiment, the forward simulation and migration imaging module 330 is further configured to:
[0108] Repeatedly call these units to perform multiple model fabrications, forward simulations, and migration imaging, and generate multiple scattered wave imaging data and multiple imaging data including reflected and scattered waves as training samples for training the deep learning neural network model.
[0109] In one embodiment, the training module 340 is specifically configured to:
[0110] Input the multiple scattered wave imaging data and the multiple imaging data including reflected and scattered waves into the deep learning neural network model, perform mapping from input data to output data, set multiple hidden layers, and train the deep learning neural network model through multiple error backpropagations and iterations.
[0111] It should be noted that although several modules of the system for scattered wave imaging using deep learning are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present invention, the features and functions of two or more of the above-described modules can be embodied in one module. Conversely, the features and functions of one module described above can be further divided and embodied by multiple modules.
[0112] Based on the foregoing inventive concept, as Figure 5 shown, the present invention also provides a computer device 500, including a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 530, the method for scattered wave imaging using deep learning described above is implemented.
[0113] Based on the foregoing inventive concept, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method for scattered wave imaging using deep learning described above is implemented.
[0114] The method and system for scattering wave imaging using deep learning proposed by the present invention extract original scattering imaging data from the acquired imaging data; establish a deep learning neural network model; obtain training samples for training the deep learning neural network model through forward modeling and migration imaging, where the input data includes imaging data of reflected and scattered waves, and the output data is scattering wave imaging data; use the trained deep learning neural network model to process actual imaging data to obtain scattering imaging data; multiply the scattering imaging data by a proportionality coefficient and add it to the original scattering imaging data to obtain a scattering wave imaging result, realizing the direct extraction of scattering imaging from imaging data, with high automation and effectively improving the accuracy and efficiency of scattering imaging, providing strong data support for seismic exploration.
[0115] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0116] The present invention is described with reference to the flowcharts and / or block diagrams of methods and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.
[0117] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that realizes the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.
[0118] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for implementing the functions specified in one block or a plurality of blocks.
[0119] Finally, it should be noted that the above embodiments are only specific embodiments of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily conceive of changes, or equivalently replace some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.
Claims
1. A method for scattering wave imaging using deep learning, characterized in that, The method includes: Obtaining imaging data and extracting original scattered imaging data from the imaging data; Establishing a deep learning neural network model; Obtaining training samples for training the deep learning neural network model through forward modeling and migration imaging; Training the deep learning neural network model using the training samples to perform mapping from input data to output data, where the input data includes imaging data of reflected and scattered waves, and the output data is scattered wave imaging data; Processing actual imaging data using the trained deep learning neural network model to obtain scattered imaging data; Multiplying the scattered imaging data by a scale factor and adding it to the original scattered imaging data to obtain a scattered wave imaging result; Among them, obtaining training samples for training the deep learning neural network model through forward modeling and migration imaging includes: Making a scattered wave reflection coefficient model and a reflected wave reflection coefficient model; Adding the scattered wave reflection coefficient model and the reflected wave reflection coefficient model to obtain a total reflection coefficient model; Setting a background velocity field; Processing according to the scattered wave reflection coefficient model and the background velocity field using the wave equation Born approximation forward modeling method to obtain a first pre-stack simulation record of scattered waves; Processing according to the total reflection coefficient model and the background velocity field using the wave equation Born approximation forward modeling method to obtain a second pre-stack simulation record of scattered waves; Performing migration imaging according to the first pre-stack simulation record of scattered waves and the background velocity field to obtain scattered wave imaging data; Performing migration imaging according to the second pre-stack simulation record of scattered waves and the background velocity field to obtain imaging data including reflected and scattered waves; Generating training samples for training the deep learning neural network model according to the scattered wave imaging data and the imaging data including reflected and scattered waves.
2. The method for scattering wave imaging using deep learning according to claim 1, wherein The method further includes: Repeatedly performing model making, forward modeling and migration imaging to generate multiple scattered wave imaging data and multiple imaging data including reflected and scattered waves as training samples for training the deep learning neural network model.
3. The method for scattering wave imaging using deep learning according to claim 2, wherein Training the deep learning neural network model using the training samples to perform mapping from input data to output data includes: Inputting multiple scattered wave imaging data and multiple imaging data including reflected and scattered waves into the deep learning neural network model to perform mapping from input data to output data, setting multiple hidden layers, and training the deep learning neural network model through multiple error backpropagations and iterations.
4. A system for scattering wave imaging using deep learning, characterized in that, The system includes: A data acquisition module for obtaining imaging data and extracting original scattered imaging data from the imaging data; A model establishment module for establishing a deep learning neural network model; A forward modeling and migration imaging module for obtaining training samples for training the deep learning neural network model through forward modeling and migration imaging; A training module for training the deep learning neural network model using the training samples to perform mapping from input data to output data, where the input data includes imaging data of reflected and scattered waves, and the output data is scattered wave imaging data; A data processing module for processing actual imaging data by using a trained deep learning neural network model to obtain scattered imaging data; A calculation module for multiplying the scattered imaging data by a scale factor and adding it to the original scattered imaging data to obtain a scattered wave imaging result; Wherein, the forward simulation and migration imaging module includes: A model making unit for making a scattered wave reflection coefficient model and a reflected wave reflection coefficient model, and adding the scattered wave reflection coefficient model and the reflected wave reflection coefficient model to obtain a total reflection coefficient model; A setting unit for setting a background velocity field; A first forward simulation unit for processing according to the scattered wave reflection coefficient model and the background velocity field by using the wave equation Born approximation forward simulation method to obtain a first pre-stack simulation record of scattered waves; A second forward simulation unit for processing according to the total reflection coefficient model and the background velocity field by using the wave equation Born approximation forward simulation method to obtain a second pre-stack simulation record of scattered waves; A first migration imaging unit for performing migration imaging according to the first pre-stack simulation record of scattered waves and the background velocity field to obtain scattered wave imaging data; A second migration imaging unit for performing migration imaging according to the second pre-stack simulation record of scattered waves and the background velocity field to obtain imaging data including reflected and scattered waves; A training sample generation unit for generating training samples for training a deep learning neural network model according to the scattered wave imaging data and the imaging data including reflected and scattered waves.
5. The system for scattered wave imaging using deep learning according to claim 4, wherein The forward simulation and migration imaging module is further configured to: Repeatedly perform model making, forward simulation and migration imaging to generate a plurality of scattered wave imaging data and a plurality of imaging data including reflected and scattered waves as training samples for training a deep learning neural network model.
6. The system for scattering wave imaging using deep learning according to claim 5, wherein The training module is specifically configured to: Input a plurality of scattered wave imaging data and a plurality of imaging data including reflected and scattered waves into a deep learning neural network model, perform mapping from input data to output data, set a plurality of hidden layers, and train the deep learning neural network model through multiple error backpropagations and iterations.
7. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 3.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the method according to any one of claims 1 to 3.