Sample data selection method and device, equipment and storage medium
By obtaining the grid feature values of a geographic region during seismic exploration, the seismic data of the target shot point is automatically selected as a sample, solving the problem of low efficiency of manual selection, realizing the reasonable distribution and efficient selection of sample data, and improving the accuracy of seismic exploration tasks.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2026-03-31
AI Technical Summary
In existing technologies, the sample data for artificial intelligence models in seismic exploration are mostly selected manually, which is inefficient and affected by the skill level of the operators, making it difficult to achieve efficient selection of sample data in different work areas.
By acquiring grid feature values of a geographical region, target shot points are automatically selected, and their seismic data is used as sample data. Deep neural networks and high-performance computing devices are used to achieve automated selection and reasonable distribution of sample data.
This improved the efficiency and accuracy of sample data selection, ensured the effective application of artificial intelligence models in different terrains, and enhanced the accuracy of seismic exploration tasks.
Smart Images
Figure CN121763369A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of oil and gas geophysical exploration technology, and in particular to a sample data selection method, apparatus, equipment and storage medium. Background Technology
[0002] Seismic exploration is a geophysical exploration method for oil and gas that infers the properties and morphology of underground rock strata by observing and analyzing the propagation patterns of artificially generated seismic waves underground.
[0003] In the field of seismic exploration, artificial intelligence (AI) models are often used to analyze seismic waves. However, the sample data used to train AI models is mostly selected manually, which is inefficient. Summary of the Invention
[0004] This application provides a sample data selection method, apparatus, device, and storage medium. The technical solution provided by this application is as follows:
[0005] According to one aspect of the embodiments of this application, a sample data selection method is provided, the method comprising:
[0006] Obtain the feature values of multiple grids, which correspond to different sub-regions in the first geographic region. The feature values of the grids are used to characterize the terrain complexity of the sub-regions corresponding to the grids.
[0007] Based on the feature values of each of the multiple grids, n target firing points are determined from the m firing points included in the first geographical region, where n is an integer greater than 1 and m is an integer greater than n;
[0008] The seismic data of each of the n target shot points are selected as sample data, and the seismic data of the shot points are used to reflect the seismic waves generated from the shot points.
[0009] According to one aspect of the embodiments of this application, a sample data selection apparatus is provided, the apparatus comprising:
[0010] The feature acquisition module is used to acquire the feature values of multiple grids, which correspond to different sub-regions in the first geographic region. The feature values of the grids are used to characterize the terrain complexity of the sub-regions corresponding to the grids.
[0011] The firing point determination module is used to determine n target firing points from m firing points included in the first geographical region based on the feature values of each of the multiple grids, where n is an integer greater than 1 and m is an integer greater than n;
[0012] The data acquisition module acquires the seismic data of each of the n target shot points, and the seismic data of the shot points is used to characterize the seismic waves generated from the shot points.
[0013] According to one aspect of the embodiments of this application, a computer device is provided, the computer device including a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the above-described sample data selection method.
[0014] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, the computer program being loaded and executed by a processor to implement the above-described sample data selection method.
[0015] According to one aspect of the embodiments of this application, a computer program product is provided, the computer program product including a computer program stored in a computer-readable storage medium, and a processor reading from the computer-readable storage medium and executing the computer program to implement the above-described sample data selection method.
[0016] The technical solutions provided in this application have at least the following beneficial effects:
[0017] By mapping different sub-regions within the first geographic region to multiple grids, and based on the characteristic values of each grid, n target shot points are determined from the first geographic region for collecting sample data. Seismic data reflecting the seismic waves generated by these n target shot points are then selected as sample data. On one hand, this achieves automated sample data selection; on the other hand, since the characteristic values characterize the topographic complexity of the sub-regions corresponding to the grids, the method also enables the reasonable selection of sample data based on the actual topography of the first geographic region. Attached Figure Description
[0018] Figure 1 This is a schematic diagram of a computer system provided in one embodiment of this application;
[0019] Figure 2 This is a flowchart of a sample data selection method provided in one embodiment of this application;
[0020] Figure 3 This is a schematic diagram of multiple grids provided in one embodiment of this application;
[0021] Figure 4 This is a schematic diagram of the first shot point and the (p-1)th target shot point in a plurality of grids provided in one embodiment of this application;
[0022] Figure 5This is a flowchart illustrating a specific example of a sample data selection method provided in one embodiment of this application;
[0023] Figure 6 This is a schematic diagram of a gridded representation of a first geographic region according to an embodiment of this application;
[0024] Figure 7 This is a schematic diagram of the target firing point selection result provided in one embodiment of this application;
[0025] Figure 8 This is a block diagram of a sample data selection device provided in one embodiment of this application;
[0026] Figure 9 This is a structural block diagram of a computer device provided in one embodiment of this application. Detailed Implementation
[0027] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.
[0028] With breakthroughs in deep neural network technology and the introduction of high-performance computing devices such as GPUs (Graphics Processing Units), artificial intelligence (AI) has achieved a qualitative leap in general fields such as image recognition and speech recognition, far surpassing conventional algorithms. In the field of seismic exploration, corresponding intelligent algorithms have been developed for first arrival picking, velocity spectrum interpretation, noise suppression, and stratigraphic and fault interpretation. Some of these algorithms outperform conventional algorithms in both effectiveness and efficiency. This, to a certain extent, validates the potential of data-driven, artificial intelligence algorithms in handling complex problems.
[0029] Due to limited computing power and stability requirements, most currently developed artificial intelligence (AI) algorithms are based on supervised learning, requiring users to provide sample data to train the AI model. The sample data largely determines the effectiveness of the AI algorithm; it needs to have high accuracy, extremely high representativeness, and similar proportions of data with different characteristics. Furthermore, due to the significant variations in seismic data characteristics across different work areas and the high computational demands of general AI models, current AI models are mostly only used for seismic exploration tasks in work areas specified by technicians. When handling seismic exploration tasks in new work areas, retraining or transfer training based on the existing model is required. In this case, it is necessary to select new sample data for the new work area. Sample data selection is a common and extremely important task; the quality of the selected sample data directly affects the performance of the AI model.
[0030] Currently, sample data selection is primarily manual, requiring operators to perform extensive data browsing and analysis, resulting in a heavy workload and effectiveness dependent on the operators' skill levels. The most thorough solution would be to automatically analyze the characteristics of seismic data and automatically generate sample data; however, considering that current seismic data is typically in the TB or even PB range, the computational load is enormous, making practical implementation impractical. For land data, the quality of sample data is mainly affected by near-surface undulations and complex surface structures. Based on near-surface elevation, velocity, and other attributes, representative sample data can be selected or the range of sample data can be significantly narrowed.
[0031] Therefore, it is essential to study automated sample data selection based on surface features.
[0032] Please refer to Figure 1 This illustration shows a schematic diagram of a computer system provided in one embodiment of this application. The computer system includes a terminal device 10 and a server 20.
[0033] Terminal device 10 can be an electronic device such as a mobile phone, tablet computer, PC (Personal Computer), or in-vehicle terminal device. A client application for a target application can be installed and run on terminal device 10. This target application has the function of acquiring and filtering seismic data reflecting seismic waves. In some embodiments, the target application also has mapping and data analysis functions. In some embodiments, terminal device 10 is located in a first geographical region, or close to the first geographical region, thereby enabling edge computing for seismic data. The first geographical region is the region used for sampling seismic data in this embodiment of the application.
[0034] Server 20 can be used to provide background services for the client of the target application in terminal device 10. For example, server 20 can be the background server of the target application. Server 20 can be a single server, a server cluster consisting of multiple servers, or a cloud computing service center.
[0035] Terminal device 10 and server 20 can communicate via a network, such as a wireless or wired network.
[0036] Please refer to Figure 2 The diagram illustrates a flowchart of a sample data selection method provided in one embodiment of this application. The execution entity for each step of this method is a computer device, such as the terminal device 10 in the above embodiment, or the server 20 in the above embodiment. The method includes at least one of the following steps 210 to 230.
[0037] Step 210: Obtain the feature values of each of the multiple grids. The multiple grids correspond to different sub-regions in the first geographic region. The feature values of the grids are used to characterize the terrain complexity of the sub-regions corresponding to the grids.
[0038] In some embodiments, the eigenvalues of a grid are used to characterize the topographic complexity of the surface of the sub-region corresponding to the grid. For example, the greater the slope of the surface undulation in the sub-region and the more frequent the surface undulation, the larger the eigenvalue of the corresponding grid.
[0039] In some embodiments, a first geographical region is divided into multiple grids on a first plane, the first plane being parallel to a base plane, for example, the first plane being the base plane. In the embodiments of this application, the base plane may be specified by a person skilled in the art, or an absolute base plane commonly used in the art may be used, and this application does not limit this.
[0040] In some embodiments, multiple grids are interconnected.
[0041] In some embodiments, multiple grids have the same size. In the following embodiments, the length value of the grid in the first direction is referred to as the first length value, and the length value of the grid in the second direction is referred to as the second length value. The first direction and the second direction are different.
[0042] In some embodiments, please refer to Figure 3 Multiple grids are interconnected rectangular grids of the same size, with a first length of X meters and a second length of Y meters, where X and Y are both positive numbers.
[0043] In some embodiments, based on the location of m gun points included in the first geographic region within the first geographic region, the m gun points are drawn in multiple grids, where m is an integer greater than 1.
[0044] In some embodiments, based on the coordinates of m gun points included in the first geographical region in the first plane, the m gun points are drawn on multiple grids. It should be noted that the origin of the coordinates in the first plane can be arbitrarily set by those skilled in the art, and this application does not limit it.
[0045] In some embodiments, step 210 includes at least one of the following sub-steps 212 to 214.
[0046] Sub-step 212: For the first grid among multiple grids, determine the average elevation of each shot point included in the sub-region corresponding to the first grid as the elevation of the first grid.
[0047] The first grid is any one of a plurality of grids. In some embodiments, the same steps as for the first grid are performed for all grids in the plurality of grids to determine the characteristic values of each of the plurality of grids.
[0048] The elevation of a shot point refers to the distance from the shot point along the vertical line to the base surface.
[0049] For example, please refer to Figure 3 If multiple grids are Figure 3 Given an M×N grid (where M and N are both integers greater than 1), the first grid is the i-th grid along the first direction and the j-th grid along the second direction. Then the elevation of the first grid is... Where b is the number of shot points included in the sub-region corresponding to the first grid, and s a Let be the elevation of the a-th shot point included in the sub-region corresponding to the first grid, where b is a positive integer, a is a positive integer less than or equal to b, i is a positive integer less than or equal to M, and j is a positive integer less than or equal to N.
[0050] In some embodiments, the derivatives of the elevation of the first grid in the first direction and the derivatives in the second direction are calculated, and the modulus of the derivative in the first direction and the modulus of the derivative in the second direction are added together to obtain the eigenvalue of the first grid.
[0051] Sub-step 214: Determine the characteristic value of the first grid based on the elevation of the first grid and the elevation of the adjacent grids of the first grid.
[0052] In some embodiments, sub-step 214 includes at least one of the following steps:
[0053] 1. Determine the first derivative based on the elevation of the first grid and the elevations of the adjacent grids in the first direction. The first derivative is (determined by the finite difference method) the derivative of the elevation of the first grid in the first direction.
[0054] In some embodiments, the adjacent grids of the first grid in the first direction include the second grid. The second grid and the first grid are connected sequentially in the first direction. The step includes: dividing the difference between the elevation of the first grid and the elevation of the second grid by a first length value to obtain a first derivative.
[0055] In some embodiments, the adjacent grids of the first grid in the first direction include a third grid. The first grid and the third grid are connected sequentially in the first direction. The step includes: dividing the difference between the elevation of the first grid and the elevation of the third grid by a first length value to obtain a first derivative.
[0056] In some embodiments, the adjacent grids of the first grid in the first direction include a second grid and a third grid. The second grid, the first grid, and the third grid are connected sequentially in the first direction. The step includes: (1) subtracting the product of 2 and the elevation of the first grid from the sum of the elevations of the second grid and the third grid to obtain a first intermediate value. (2) dividing the first intermediate value by the product of 2 and a first length value to obtain a first derivative.
[0057] For example, please refer to Figure 3 If the first grid is the i-th grid along the first direction and the j-th grid along the second direction out of M×N grids, then the second grid is the (i-1)-th grid along the first direction and the j-th grid along the second direction out of M×N grids, and the third grid is the (i+1)-th grid along the first direction and the j-th grid along the second direction out of M×N grids. First derivative Among them, h (i+1,j) h represents the elevation of the third grid. (i-1,j) This represents the elevation of the second grid.
[0058] 2. Determine the second derivative based on the elevation of the first grid and the elevation of its adjacent grids in the second direction. The second derivative is the derivative of the elevation of the second grid in the second direction (determined by the finite difference method).
[0059] In some embodiments, the adjacent grids of the first grid in the second direction include a fourth grid. The fourth grid and the first grid are sequentially connected in the first direction. The step includes: dividing the difference between the elevation of the first grid and the elevation of the fourth grid by a second length value to obtain a second derivative.
[0060] In some embodiments, the adjacent grids of the first grid in the second direction include a fifth grid. The first grid and the fifth grid are connected sequentially in the first direction. The step includes: dividing the difference between the elevation of the first grid and the elevation of the fifth grid by a second length value to obtain a second derivative.
[0061] In some embodiments, the adjacent grids of the first grid in the first direction include the fourth grid and the fifth grid. The fourth grid, the first grid, and the fifth grid are connected sequentially in the first direction. The step includes: (1) subtracting the product of 2 and the elevation of the first grid from the sum of the elevations of the fourth grid and the fifth grid to obtain a second intermediate value. (2) dividing the second intermediate value by the product of 2 and the second length value to obtain a first derivative.
[0062] For example, please refer to Figure 3 If the first grid is the i-th grid along the first direction and the j-th grid along the second direction out of M×N grids, then the fourth grid is the i-th grid along the first direction and the (j-1)-th grid along the second direction out of M×N grids, and the fifth grid is the i-th grid along the first direction and the (j+1)-th grid along the second direction out of M×N grids. Second derivative Among them, h (i,j+1) h represents the elevation of the fifth grid. (i,j-1) This is the elevation of the fourth grid.
[0063] 3. Add the modulus of the first derivative and the modulus of the second derivative to obtain the eigenvalues of the first grid.
[0064] For example, please refer to Figure 3 If the first grid is the i-th grid along the first direction and the j-th grid along the second direction in an M×N grid, then the eigenvalues of the first grid are...
[0065] Step 220: Based on the characteristic values of each of the multiple grids, determine n target firing points from the m firing points included in the first geographical region.
[0066] n is an integer greater than 1, and m is an integer greater than n. n is set by the technicians as needed, and this application does not impose any restrictions on it.
[0067] In some embodiments, step 220 includes at least one of the following sub-steps 222 to 228.
[0068] Sub-step 222: Determine any one of the m gun points as the first target gun point among the n target gun points.
[0069] In this sub-step, one of the m gun points is randomly selected as the first target gun point among the n target gun points.
[0070] Sub-step 224: Based on the characteristic values of each of the multiple grids, determine the weighted distance of each of the other gun points among the m gun points, excluding the p-1 target gun points that have been determined. The weighted distance of the other gun points is used to reflect the spatial distance between the other gun points and the p-1th target gun point among the n target gun points, where p is an integer greater than 1 and less than or equal to n.
[0071] In some embodiments, sub-step 224 includes: for the first shot point among the other shot points, summing the planar distances of the first shot point according to the feature values of each associated grid of the first shot point to obtain the weighted distance of the first shot point.
[0072] The associated grid of the first shot point refers to the grid through which the line segment connecting the first shot point and the (p-1)th target shot point passes on the plane (i.e., the first plane) where multiple grids are located. The planar distance of the associated grid of the first shot point refers to the length of the line segment connecting the first shot point and the (p-1)th target shot point in the associated grid.
[0073] The first shot point is any of the other shot points. In some embodiments, the same steps as for the first shot point are performed for each of the other shot points to obtain the weighted distance of each of the other shot points.
[0074] In some embodiments, the weighted distance of the first shot point Where q is the number of grid cells traversed by the line segment connecting the first shot point and the (p-1)th target shot point (i.e., q is the number of associated grid cells of the first shot point), q is a positive integer, l k It is the planar distance (i.e., l) of the k-th associated grid among the q associated grids of the first shot point. k W is the length of the line segment connecting the first shot point and the (p-1)th target shot point in the kth associated grid. k It is the eigenvalue of the k-th associated grid, where k is a positive integer less than or equal to q.
[0075] For example, please refer to Figure 4 For the first shot point 41 and the (p-1)th target shot point 42, the line segment connecting the first shot point 41 and the (p-1)th target shot point 42 passes through 5 grids (i.e., the first shot point 41 has 5 associated grids). According to the characteristic values of each of these 5 associated grids, the planar distances (i.e., L1, L2, L3, L4 and L5) of each of these 5 associated grids are weighted and summed to obtain the weighted distance of the first shot point 41.
[0076] It should be noted that in the process of calculating the weighted distance of the first shot point in the above embodiment, the eigenvalues of the associated grid serve as weight parameters for the weighted summation of the planar distances. Therefore, the weighted distance of the first shot point is affected by the terrain of the associated grid of the first shot point, thus reflecting the spatial distance between the first shot point and the (p-1)th target shot point.
[0077] Sub-step 226: Determine the distance discrimination value of each other shot point based on the weighted distance of each other shot point. The distance discrimination value of the other shot points is used to indicate the priority of the other shot points being selected.
[0078] In some embodiments, when p equals 2, sub-step 226 includes: taking the weighted distance of each of the other shot points as the distance discrimination value of each of the other shot points and recording it in the distance array.
[0079] The distance array is used to record the distance discrimination value of each shot point, and the distance discrimination value of the shot point is used to indicate the priority of the shot point being selected.
[0080] In some embodiments, when p is greater than 2, sub-step 226 includes at least one of the following steps:
[0081] 1. For the second shot point among all other shot points, if the weighted distance of the second shot point is less than the distance discrimination value recorded for the second shot point in the distance array, the weighted distance of the second shot point is determined as the distance discrimination value of the second shot point, and the distance discrimination value recorded for the second shot point in the distance array is replaced with the weighted distance of the second shot point.
[0082] 2. If the weighted distance of the second shot point is greater than the distance discrimination value recorded for the second shot point in the distance array, the distance discrimination value recorded for the second shot point shall be determined as the distance discrimination value of the second shot point.
[0083] The second firing point is any of the other firing points. In some embodiments, the same steps as for the second firing point are performed for each of the other firing points to obtain the distance discrimination value for each of the other firing points.
[0084] Sub-step 228: Among all other firing points, the firing point with the largest distance determination value is determined as the p-th target firing point among the n target firing points.
[0085] For example, if p is 2, then the second target gun point among the n target gun points can be determined through the above sub-steps 222 to 228. Similarly, if p is n, then the nth target gun point among the n target gun points can be determined through the above sub-steps 222 to 228, thereby determining all n target gun points.
[0086] Step 230: Select the seismic data of each of the n target shot points as sample data.
[0087] Seismic data from the shot point are used to reflect the seismic waves generated from the shot point.
[0088] In some embodiments, the first geographical area further includes r geophones, which are used to detect seismic waves generated by the shot point, where r is a positive integer. After a seismic wave is generated at a shot point, the data received by the r geophones within a set time period are used as the seismic data of that shot point. The set time period is determined by a technician as needed, and this application does not limit it.
[0089] The technical solution provided in this application, by mapping different sub-regions within a first geographic region to multiple grids, determines n target shot points for collecting sample data from the first geographic region based on the characteristic values of each grid, and selects seismic data reflecting the seismic waves generated by these n shot points as sample data. On one hand, this achieves automated sample data selection; on the other hand, since the characteristic values characterize the topographic complexity of the sub-regions corresponding to the grids, the method also achieves reasonable selection of sample data based on the actual topography of the first geographic region.
[0090] Please refer to Figure 5 This document illustrates a flowchart of a specific example of a sample data selection method provided in an embodiment of this application. In this example, 60 target firing points need to be determined from 62,289 firing points included in a first geographical region to perform sample data selection. This example includes the following steps:
[0091] 1. Divide the first geographic region into multiple grids, each grid being 100m × 100m, such as... Figure 6 As shown, each non-blank grid corresponds to a sub-region within the first geographic region 60. Figure 6 Different shades of gray represent different elevations.
[0092] 2. Obtain the information of the shot points (such as their location and elevation) in the first geographic region and record the obtained information in the shot point array. The shot point array is Source[S1,S2,…,S62289]. S1, S2,…,S62289 correspond to the information of different shot points. Then, draw the 62289 shot points on multiple grids.
[0093] 3. Using the finite difference method, determine the derivatives of each grid in the first direction and the derivatives in the second direction, and use the sum of the magnitudes of the two derivatives as the feature value used to characterize the terrain complexity of the sub-region corresponding to the grid.
[0094] 4. From the 62289 firing points, randomly select one firing point as the first target firing point among the 60 target firing points. Calculate the weighted distance of all other firing points from this firing point, and store the calculation result as the initial distance discrimination value in the distance array Distance[D1,D2……D62289], where D1, D2, …, D62289 are the distance discrimination values corresponding to different firing points.
[0095] 5. Select the firing point corresponding to the largest distance discrimination value in the distance array (excluding the already selected target firing points) as the next target firing point to be selected.
[0096] 6. Calculate the weighted distance from all other gun points (excluding the selected target gun points) to the newly selected target gun point. For a given gun point, if the weighted distance is less than the distance discrimination value stored in the distance array for that gun point, then update the distance discrimination value stored in the distance array for that gun point to the weighted distance.
[0097] 7. Repeat steps 5 and 6 until all 60 target shot points are selected. The seismic data of each of the 60 target shot points constitute the final selected sample data.
[0098] Experiments showed that each of the above steps takes about 3 seconds to execute. Considering the terrain, manually selecting target firing points would take more than ten minutes. Therefore, this solution greatly improves the efficiency of sample data selection.
[0099] in addition, Figure 7 The above example shows the selection results of the target firing point. Figure 7 The firing points marked with a black cross are the final selected target firing points. Figure 7The three regions 71 correspond to areas with high terrain complexity (significant surface undulations) in the first geographical region 60, such as basins, hills, and valleys. It is evident that most target firing points fall within region 71. Due to the significant vertical drop of the firing points in region 71, this scheme achieves a uniform spatial distribution (not planar distribution) of the selected target firing points. Compared to manual selection, this improves efficiency and ensures the rationality of the target firing point location distribution.
[0100] In summary, this application addresses the challenges of low efficiency and poor performance in manual sample data selection within the field of seismic exploration by introducing a uniform sampling method based on terrain complexity. This method requires only the input of the required number of target shot points by technicians, and automatically selects corresponding sample data based on the surface characteristics of the first geographic region. Furthermore, the selection results can uniformly cover various terrain features within the first geographic region, ensuring the smooth execution of seismic exploration tasks by subsequent artificial intelligence models. This improves the accuracy of tasks such as first arrival picking, velocity spectrum interpretation, noise suppression, and stratigraphic and fault interpretation performed by artificial intelligence models.
[0101] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.
[0102] Please refer to Figure 8 This diagram illustrates a block diagram of a sample data selection device according to an embodiment of this application. The device has the function of implementing the above-described sample data selection method; this function can be implemented in hardware or by hardware executing corresponding software. The device can be a computer device or can be installed within a computer device. The device 800 may include: a feature acquisition module 810, a shot point determination module 820, and a data acquisition module 830.
[0103] The feature acquisition module 810 is used to acquire the feature values of multiple grids, which correspond to different sub-regions in the first geographic region. The feature values of the grids are used to characterize the terrain complexity of the sub-regions corresponding to the grids.
[0104] The firing point determination module 820 is used to determine n target firing points from m firing points included in the first geographical region based on the feature values of each of the plurality of grids, where n is an integer greater than 1 and m is an integer greater than n.
[0105] The data acquisition module 830 is used to acquire the seismic data of each of the n target shot points, and the seismic data of the shot points is used to characterize the seismic waves generated from the shot points.
[0106] In some embodiments, the shot point determination module 820 includes: a target determination submodule, a distance determination submodule, and a discrimination value determination submodule.
[0107] The target determination submodule is used to determine any one of the m gun points as the first target gun point among the n target gun points.
[0108] The distance determination submodule is used to determine the weighted distance of each of the m shot points, excluding the p-1 target shot points that have been determined, based on the feature values of the multiple grids. The weighted distance of the other shot points is used to reflect the spatial distance between the other shot points and the p-1th target shot point among the n target shot points, where p is an integer greater than 1 and less than or equal to n.
[0109] The discrimination value determination submodule is used to determine the distance discrimination value of each other shot point based on the weighted distance of each other shot point. The distance discrimination value of the other shot points is used to indicate the priority of the other shot points being selected.
[0110] The target determination submodule is also used to determine the gun point with the largest distance determination value among the other gun points as the p-th target gun point among the n target gun points.
[0111] In some embodiments, the distance determination submodule is used to, for the first shot point among the other shot points, sum the planar distances of the associated grids of the first shot point according to the feature values of each associated grid of the first shot point, to obtain the weighted distance of the first shot point; wherein, the associated grid of the first shot point refers to the grid through which the line segment connecting the first shot point and the (p-1)th target shot point passes on the plane where the plurality of grids are located, and the planar distance of the associated grid of the first shot point refers to the length of the line segment connecting the first shot point and the (p-1)th target shot point in the associated grid.
[0112] In some embodiments, when p equals 2, the discrimination value determination submodule is used to record the weighted distance of each of the other shot points as the distance discrimination value of each of the other shot points in the distance array. When p is greater than 2, the discrimination value determination submodule is used to, for the second shot point among the other shot points, if the weighted distance of the second shot point is less than the distance discrimination value recorded for the second shot point in the distance array, determine the weighted distance of the second shot point as the distance discrimination value of the second shot point, and replace the distance discrimination value recorded for the second shot point in the distance array with the weighted distance of the second shot point; if the weighted distance of the second shot point is greater than the distance discrimination value recorded for the second shot point in the distance array, determine the distance discrimination value recorded for the second shot point as the distance discrimination value of the second shot point.
[0113] In some embodiments, the feature acquisition module 810 is configured to, for a first grid among the plurality of grids, determine the average elevation of each shot point included in the sub-region corresponding to the first grid as the elevation of the first grid; and determine the feature value of the first grid based on the elevation of the first grid and the elevation of the adjacent grids of the first grid.
[0114] In some embodiments, the adjacent grids of the first grid include a second grid, a third grid, a fourth grid, and a fifth grid.
[0115] The feature acquisition module 810 is used to: subtract the product of 2 and the elevation of the first grid from the sum of the elevations of the second grid and the third grid to obtain a first intermediate value; divide the first intermediate value by the product of 2 and a first length value to obtain a first derivative; subtract the product of 2 and the elevation of the first grid from the sum of the elevations of the fourth grid and the fifth grid to obtain a second intermediate value; divide the second intermediate value by the product of 2 and a second length value to obtain a second derivative; and add the modulus of the first derivative and the modulus of the second derivative to obtain the feature value of the first grid; wherein the second grid, the first grid, and the third grid are connected sequentially in a first direction, and the fourth grid, the first grid, and the fifth grid are connected sequentially in a second direction, the first length value is the length value of the grid in the first direction, and the second length value is the length value of the grid in the second direction.
[0116] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.
[0117] Please refer to Figure 9 The diagram illustrates a structural block diagram of a computer device provided in one embodiment of this application.
[0118] Typically, computer device 900 includes a processor 901 and a memory 902.
[0119] Processor 901 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 901 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field Programmable Gate Array), and PLA (Programmable Logic Array). Processor 901 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 901 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 901 may also include an AI processor, which is used to handle computational operations related to machine learning.
[0120] The memory 902 may include one or more computer-readable storage media, which may be tangible and non-transitory. The memory 902 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 902 stores a computer program that is loaded and executed by the processor 901 to implement the sample data selection method described above.
[0121] Those skilled in the art will understand that Figure 9The structure shown does not constitute a limitation on the computer device 900, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.
[0122] In some embodiments, a computer-readable storage medium is also provided, wherein a computer program is stored therein, the computer program being loaded and executed by a processor to implement the above-described sample data selection method.
[0123] Optionally, the computer-readable storage medium may include: ROM (Read-Only Memory), RAM (Random-Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).
[0124] In some embodiments, a computer program product is also provided, the computer program product including a computer program stored in a computer-readable storage medium, and a processor reading from the computer-readable storage medium and executing the computer program to implement the above-described sample data selection method.
[0125] It should be understood that "multiple" as used herein refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the step numbers described herein are merely illustrative of one possible execution order. In some other embodiments, the steps may not be executed in numerical order, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.
[0126] The above are merely exemplary embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application shall be included within the protection scope of this application.
Claims
1. A method of sample data selection, the method comprising: The method comprises: obtaining characteristic values of a plurality of grids, the plurality of grids corresponding to different sub-regions in a first geographical region, the characteristic value of the grid being used to represent the terrain complexity of the sub-region corresponding to the grid; determining n target shots from m shots included in the first geographical region according to the characteristic values of the plurality of grids, n being an integer greater than 1, and m being an integer greater than n; selecting the seismic data of the n target shots as sample data, the seismic data of the shot being used to reflect the seismic wave excited from the shot.
2. The method of claim 1, wherein, The method comprises: determining any shot in the m shots as the first target shot in the n target shots; determining the weighted distance of each other shot in the m shots except the p-1 target shots according to the characteristic values of the plurality of grids, the weighted distance of the other shot being used to reflect the spatial distance between the other shot and the p-1 target shot in the n target shots, p being an integer greater than 1 and less than or equal to n; determining the distance discriminant value of each other shot according to the weighted distance of each other shot, the distance discriminant value of the other shot being used to indicate the priority of the other shot being selected; determining the shot with the largest distance discriminant value in the other shots as the p target shot in the n target shots.
3. The method of claim 2, wherein, The method comprises: for the first shot in the other shots, performing weighted summation on the plane distance of each adjacent grid of the first shot according to the characteristic value of each adjacent grid of the first shot, to obtain the weighted distance of the first shot; wherein the adjacent grid of the first shot refers to the grid passing through the line segment connecting the first shot and the p-1 target shot in the plane of the plurality of grids, and the plane distance of the adjacent grid of the first shot refers to the length of the line segment connecting the first shot and the p-1 target shot in the adjacent grid.
4. The method of claim 2, wherein, In the case where p is equal to 2, the method comprises: recording the weighted distance of each other shot as the distance discriminant value of each other shot into a distance array; In the case where p is greater than 2, the method comprises: For a second shot point of the respective other shot points, if the weighted distance of the second shot point is less than a distance discriminant recorded in the distance array for the second shot point, the distance discriminant recorded in the distance array for the second shot point is replaced by the weighted distance of the second shot point, and the distance discriminant recorded in the distance array for the second shot point is determined as the distance discriminant of the second shot point; If the weighted distance of the second shot point is greater than the distance discriminant recorded in the distance array for the second shot point, the distance discriminant recorded for the second shot point is determined as the distance discriminant of the second shot point.
5. The method of claim 1, wherein, The acquiring of the respective feature values of the plurality of grids comprises: For a first grid of the plurality of grids, an average elevation of respective shot points included in a sub-region corresponding to the first grid is determined as an elevation of the first grid; The feature value of the first grid is determined according to the elevation of the first grid and elevations of adjacent grids of the first grid.
6. The method of claim 5, wherein, The adjacent grids of the first grid comprise a second grid, a third grid, a fourth grid and a fifth grid; and the determining of the feature value of the first grid according to the elevation of the first grid and the elevations of the adjacent grids of the first grid comprises: A first intermediate value is obtained by subtracting a product of 2 and the elevation of the first grid from a sum of the elevation of the second grid and the elevation of the third grid; A first derivative is obtained by dividing the first intermediate value by a product of 2 and a first length value; A second intermediate value is obtained by subtracting a product of 2 and the elevation of the first grid from a sum of the elevation of the fourth grid and the elevation of the fifth grid; A second derivative is obtained by dividing the second intermediate value by a product of 2 and a second length value; The feature value of the first grid is obtained by adding a modulus of the first derivative and a modulus of the second derivative; The second grid, the first grid and the third grid are sequentially connected in a first direction, the fourth grid, the first grid and the fifth grid are sequentially connected in a second direction, the first length value is a length value of the grid in the first direction, and the second length value is a length value of the grid in the second direction.
7. A sample data selection device, characterized by comprising: The device comprises: a feature acquiring module configured to acquire respective feature values of a plurality of grids, the plurality of grids corresponding to different sub-regions in a first geographic region, and the feature value of the grid being used to represent a terrain complexity of the sub-region corresponding to the grid; a shot point determining module configured to determine n target shot points from m shot points included in the first geographic region according to the respective feature values of the plurality of grids, n being an integer greater than 1, and m being an integer greater than n; a data acquiring module configured to acquire respective seismic data of the n target shot points, the seismic data of the shot point being used to represent a seismic wave excited from the shot point.
8. A computer device, comprising: The computer device comprises a processor and a memory, and the memory stores a computer program, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, which is loaded and executed by the processor to implement the method of any one of claims 1 to 6.
10. A computer program product, characterised in that, The computer program product comprises a computer program stored in a computer readable storage medium, which is read and executed by the processor to implement the method of any one of claims 1 to 6.