Method and apparatus for generating training samples for earthquake data

Abstract

This disclosure presents a method and apparatus for generating training samples from seismic data. The specific implementation of the training sample generation method includes: performing anomalous energy interference attenuation processing on acquired seismic data using a custom threshold method to obtain attenuated seismic data; subtracting the seismic data before and after anomalous energy interference attenuation to obtain the difference, which is the anomalous energy interference data; binarizing the anomalous energy interference data obtained in the previous step to obtain sample labels; using the acquired seismic data as sample images and the sample labels obtained in the previous step to form training samples; and processing all seismic data to obtain a training sample set. This implementation diversifies the training samples for seismic data in applications involving the identification of anomalous energy interference areas and makes training samples easier to obtain.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of seismic data processing and interpretation, specifically to a method and apparatus for generating seismic data training samples. Background Technology

[0002] Seismic methods are crucial in exploration geophysics. In artificially generated seismic data, strong-energy interference is a typical type of interference wave, prevalent in raw seismic data. Currently, attenuation of strong-energy interference in seismic data typically involves first determining the interference distribution range, then using amplitude truncation or median filtering. Determining the range of strong-energy interference usually employs a custom threshold method; amplitude or energy exceeding a given threshold is identified as interference. In practice, repeated experiments are necessary to determine the distribution range of strong-energy signals and reasonable amplitude (energy) thresholds, and these parameters vary depending on the excitation and reception locations. The difficulty and complexity of this work increase further when near-surface conditions are complex. Summary of the Invention

[0003] The summary section of this disclosure is intended to provide a brief overview of concepts that will be described in detail in the subsequent detailed description section. This summary section is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions. Some embodiments of this disclosure propose training sample generation methods and apparatus to address the technical problems mentioned in the background section above.

[0004] In a first aspect, some embodiments of this disclosure provide a method for generating earthquake data training samples. The method includes: performing abnormal energy interference attenuation processing on the acquired earthquake data to obtain earthquake data after interference attenuation; subtracting the earthquake data before and after the abnormal energy interference attenuation to obtain the difference between the two, i.e., abnormal energy interference data; performing binarization processing on the abnormal energy interference data obtained in the previous step to obtain sample labels; using the acquired earthquake data as sample images and the sample labels obtained in the previous step to form training samples; and processing all earthquake data to obtain a training sample set.

[0005] Secondly, some embodiments of this disclosure provide an apparatus for generating earthquake data training samples. The apparatus includes: an earthquake dataset acquisition unit, an anomaly energy interference generation unit, a sample and label image generation unit, and a sample set generation unit. The earthquake dataset acquisition unit is configured to acquire an acquired earthquake dataset; the anomaly energy interference generation unit is configured to perform anomaly energy interference attenuation processing on the earthquake data in the aforementioned earthquake dataset to obtain anomaly energy interference data; the sample and label image generation unit is configured to combine the earthquake data and the binarized anomaly energy interference data to form a sample image; and the sample set generation unit is configured to compose training samples based on the sample images and anomaly energy interference data label images, and to process the earthquake data in all acquired earthquake datasets to obtain a training sample set.

[0006] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, which, when executed by one or more processors, cause the one or more processors to implement the method described in the first aspect.

[0007] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in the first aspect. Attached Figure Description

[0008] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0009] Figure 1 This is a schematic diagram illustrating an application scenario of the training sample generation method disclosed herein.

[0010] Figure 2 This is a flow diagram of some embodiments of the training sample generation method according to the present disclosure.

[0011] Figure 3 This is a schematic diagram of the statistical threshold value for abnormal energy interference attenuation based on the training sample generation method of this disclosure.

[0012] Figure 4 This is a schematic diagram of the anomalous energy interference attenuation results and differences based on the training sample generation method of this disclosure.

[0013] Figure 5 This is a schematic diagram of the binarization of anomalous energy interference data according to the training sample generation method of this disclosure.

[0014] Figure 6 It is a training sample image generated according to the training sample generation method of this disclosure.

[0015] Figure 7 This is a flow chart of an embodiment of the image classification generation method according to the present disclosure.

[0016] Figure 8 This is an image segmentation result image generated according to the image classification method of this disclosure.

[0017] Figure 9 These are schematic diagrams illustrating the structure of some embodiments of the training sample generation apparatus according to this disclosure.

[0018] Figure 10 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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.

[0020] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Figure 1 This is a schematic diagram illustrating an application scenario of the training sample generation method according to some embodiments of this disclosure.

[0022] exist Figure 1In application scenario 100, firstly, computing device 101 can receive seismic dataset 102 acquired in the field. Seismic data acquired in the field is typically arranged in the form of a common shot set. Computing device 101 can extract acquired seismic data 103 from the seismic dataset 102. Here, the acquired seismic data is usually data from one common shot set, and the acquired seismic data 103 is also the source of training samples. Then, computing device 101 performs anomalous energy interference attenuation processing on the acquired seismic data 103 to obtain processed seismic data 104. Next, computing device 101 subtracts the processed seismic data 104 from the acquired seismic data 103 to obtain the difference 105. Then, the difference 105 is binarized to obtain sample labels 106, where non-zero values ​​are changed to 1 during binarization. Then, the acquired seismic data 103 is graphically represented as a sample image 107. Finally, sample image 107 and sample labels 106 together form training samples 108. In the schematic diagram 100 of this application scenario, the label is: Image of the distribution area of ​​anomalous energy interference in seismic data. The label values ​​are set to 0 and 1. The value is set to 1 within the anomalous energy interference area and 0 outside the area.

[0023] It should be noted that the aforementioned computing device 101 can be either hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed within the hardware devices listed above. It can be implemented as, for example, multiple software programs or software modules used to provide distributed services, or as a single software program or software module. No specific limitations are made here. Figure 1 The number of computing devices shown is merely illustrative. Any number of computing devices can be used depending on implementation needs.

[0024] refer to Figure 2 The diagram illustrates a flow 200 of some embodiments of a training sample generation method according to the present disclosure. This training sample generation method includes the following steps:

[0025] Step 201: Perform abnormal energy interference attenuation processing on the acquired seismic data to obtain the seismic data.

[0026] Currently, most methods for attenuating high-energy noise employ a custom threshold method, where the user provides a planar window and a vertical time window, defining S. n (t)={r1(t),r2(t),r3(t),...,r n (t)}, where n is the total number of seismic traces within the planar window, r n(t) is the amplitude value in the time window of the k-th seismic trace, S n (t) can be viewed as the set of amplitudes of multiple seismic data within a specified time window. For S... n (t) is used to obtain the threshold through statistical analysis:

[0027] The threshold represents the maximum possible amplitude of a valid signal. Any amplitude exceeding the threshold can be considered strong energy noise, and the energy of samples identified as noise will be attenuated or interpolated. The statistical method can be the mean, median, root mean square, or any S value. n The percentage value requires a reasonable statistical method based on S. n The distribution characteristics of the midpoints determine the denoising method. Denoising methods based on user-defined thresholds can be performed in either the time or frequency domain. Considering that seismic data amplitudes can be positive or negative in the time domain, when performing statistics in the time domain, the absolute value of the amplitude is usually taken first. If performed in the frequency domain, then the amplitude spectrum of the signal within the time window needs to be calculated first, and then the amplitude spectrum values ​​are statistically analyzed to obtain a reasonable threshold for the amplitude spectrum. Figure 3 This is a diagram illustrating the statistical threshold values, where... Figure 3 (a) shows the seismic data within the analysis window. Figure 3 (b) is... Figure 3 The data in (a) are taken as absolute values. Figure 3 (c) is correct. Figure 3 The data in (b) is taken from the envelope. Figure 3 (d) is about Figure 3 (c) The amplitude value of each channel is extracted at the 450ms position, and the result is obtained. Figure 3 The average amplitude in (d) is used as a threshold. Amplitudes exceeding the threshold will be attenuated as strong energy interference.

[0028] Step 202: Subtract the seismic data before and after the anomalous energy interference attenuation to obtain the difference as the anomalous energy interference data.

[0029] After attenuating the seismic data for abnormal energy interference, most of the abnormal energy interference has been attenuated. The difference between the processing results before and after attenuation is the abnormal energy interference that has been attenuated. Figure 4 (a) is seismic data from a single shot gather collected in the field. Figure 4 (b) shows the seismic data after attenuation due to anomalous energy interference. Figure 4 (c) shows the difference in constant energy interference data between the two.

[0030] Step 203: The difference between the seismic data before and after the abnormal energy interference attenuation obtained in step 202 is binarized to obtain the sample label.

[0031] Not all areas of the seismic data have anomalous energy interference. After attenuating the seismic data for anomalous energy interference, the locations with anomalous energy interference will have values, while the locations without anomalous energy interference will have values ​​of 0. The locations with values ​​will be set to 1, and these will be used as labels for training samples. The anomalous energy interference area is the area with a value of 1, and the area outside the area will have a value of 0. Figure 5 (a) The difference between the seismic data before and after the anomalous energy interference attenuation is the anomalous energy interference data. Figure 5 (b) Binarization process to obtain sample labels.

[0032] Step 204: The acquired seismic data is used as sample images and the sample labels obtained in the previous step to form training samples. After processing all the seismic data, a training sample set is obtained.

[0033] Training samples are generated based on the sample images and their labels. The training sample set is obtained by processing all images in the image set. (As an example) Figure 6 This is a training sample.

[0034] The method provided in some embodiments of this disclosure combines a labeled image obtained by binarizing each seismic data image in the aforementioned field-acquired seismic dataset with the anomalous interference data obtained by subtracting the data before and after the anomalous energy interference attenuation. This makes the training samples more diverse and easier to obtain.

[0035] Further reference Figure 7 The diagram illustrates a flow 700 of some embodiments of an image classification generation method. The flow 700 of this image classification generation method includes the following steps:

[0036] Step 701: Input seismic data images of the areas with abnormal energy interference to be identified. For example... Figure 8 (a).

[0037] Step 702: Input the target image into a pre-trained image segmentation network to obtain the segmentation result of the abnormal energy interference region. The training samples for the classification network are obtained through... Figure 2 Or generated by the corresponding method. For example, as shown below... Figure 8 (b)

[0038] In some embodiments, the pre-trained classification network may be a U-net network or an FCN network.

[0039] Further reference Figure 9 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a seismic data training sample generation device, which are similar to... Figure 2Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0040] like Figure 9 As shown, the training sample generation apparatus 900 in some embodiments includes: an earthquake dataset acquisition unit 901, an anomaly energy interference generation unit 902, a sample and label image generation unit 903, and a sample set generation unit 904. The earthquake dataset acquisition unit 901 is configured to acquire the acquired earthquake dataset; the anomaly energy interference generation unit 902 is configured to perform anomaly energy interference attenuation processing on the earthquake data in the aforementioned earthquake dataset to obtain anomaly energy interference data; the sample and label image generation unit 903 is configured to combine the earthquake data and the binarized anomaly energy interference data to form a sample image; and the sample set generation unit 904 is configured to compose training samples based on the sample images and anomaly energy interference data label images, and to process the earthquake data in all acquired earthquake datasets to obtain a training sample set.

[0041] The following is for reference. Figure 10 It shows a schematic diagram of the structure of an electronic device 1000 suitable for implementing the embodiments of this application.

[0042] like Figure 10 As shown, the electronic device 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate tasks and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage section 1008 into a random access memory (RAM) 1003. The RAM 1003 also stores various programs and data required for the operation of the system 1000. The CPU 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0043] The following components are connected to I / O interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to I / O interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 1010 as needed so that computer programs read from it can be installed in storage section 1008 as needed.

[0044] In particular, according to embodiments of the present invention, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, embodiments of the present invention include a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned dynamic data tracking method based on a front-end framework.

[0045] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1009, and / or installed from the removable medium 1011.

[0046] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0047] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0048] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0049] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0050] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0051] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for generating training samples, characterized in that... Includes the following steps: (1) The acquired seismic data were processed by a custom threshold method to attenuate abnormal energy interference, and the seismic data after interference attenuation was obtained. (2) Subtract the seismic data before and after the above-mentioned abnormal energy interference attenuation to obtain the difference between the two, which is the abnormal energy interference data; (3) Binarize the abnormal energy interference data obtained in the previous step to obtain sample labels; (4) The collected seismic data and the sample labels obtained in the previous step are used as sample images to form training samples. After processing all the seismic data, the training sample set is obtained.

2. The method according to claim 1, wherein, The acquired seismic data is processed by a custom threshold method to attenuate abnormal energy interference, and the attenuated seismic data is obtained through the following steps: The acquired seismic data is processed by anomalous energy interference attenuation in sequence according to the arrangement of common shot points or common receiver points to obtain the attenuated seismic data.

3. The method according to claim 1, wherein, The difference between the seismic data before and after the abnormal energy interference attenuation is obtained by subtracting the seismic data before and after the abnormal energy interference attenuation. This is achieved through the following steps: Subtract the seismic data before and after the abnormal energy interference attenuation to obtain the abnormal energy interference data. If the minuend is the seismic data after the abnormal energy interference attenuation, then the subtrahend is the seismic data before the abnormal energy interference attenuation.

4. The method according to claim 1, wherein, The process of binarizing the abnormal energy interference data obtained in the previous step to obtain sample labels is achieved through the following steps: In the abnormal energy interference data, the positions where abnormal energy interference exists will have values, and the positions where there is no abnormal energy interference will have values ​​of 0. The positions with values ​​will be set to 1, and these will be used as the labels for training samples. The abnormal energy interference region is the region with a value of 1, and the values ​​outside the region will be set to 0.

5. The method according to claim 1, wherein, The acquired seismic data is used as sample images, and the sample labels obtained in the previous step are combined to form training samples. The training sample set is obtained by processing all the seismic data. The process involves the following steps: the acquired seismic data is converted into images and used as sample images; the acquired sample images and their labels are combined to form training samples; and the training sample set is obtained by processing all the seismic data in the acquired seismic dataset.

6. A seismic data training sample generation device, comprising: The system comprises an earthquake dataset acquisition unit, an anomalous energy interference generation unit, a sample and label image generation unit, and a sample set generation unit. The earthquake dataset acquisition unit is configured to acquire the acquired earthquake dataset. The anomalous energy interference generation unit is configured to perform anomalous energy interference attenuation processing on the earthquake data in the aforementioned earthquake dataset to obtain anomalous energy interference data. The sample and label image generation unit is configured to combine the earthquake data with the binarized anomalous energy interference data to form a sample image. The sample set generation unit is configured to compose training samples based on the sample images and anomalous energy interference data label images, and to obtain a training sample set by processing all the acquired earthquake datasets.

7. An electronic device, comprising: One or more processors; A storage device having stored one or more programs thereon, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 5.

8. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 5.

Images