Seismic data denoising method and device based on mathematical morphology

Abstract

The invention provides a seismic data denoising method and device based on mathematical morphology. According to the method, mathematical morphology is applied to seismic data processing, the characteristics and advantages of noise can be accurately identified through mathematical morphology operation, and seismic wave field information is detected by using structural elements in a specified shape in combination with the characteristics of seismic data, so that the purpose of high-fidelity denoising of effective wave detail information is kept while interference is removed is achieved. The embodiment of the invention also has the beneficial effects of simple algorithm, strong noise adaptability, high automation degree, convenient design of implementation parameters and the like.

Description

Technical Field

[0001] The present invention relates to the field of seismic data processing in the field of earth sciences, and more particularly, to a seismic data denoising method and apparatus based on mathematical morphology. Background Art

[0002] Low signal-to-noise ratio of seismic data is a key issue affecting the current seismic exploration accuracy. Especially, numerous secondary seismic sources are formed by the outcropping old strata reef peaks, presenting strong energy pseudo-linear noise in the seismic records, which submerges the effective seismic reflection waves. A large amount of strong pseudo-linear noise covers the effective signals, resulting in a very low signal-to-noise ratio of the data, affecting the processing and interpretation of seismic data. The existence of pseudo-linear noise makes it difficult for us to identify the in-phase axis of the effective wave, affecting the processing processes and results such as dynamic and static correction, velocity analysis, and migration of seismic data, and reducing the credibility of seismic interpretation and inversion results. Usually existing problems include: reducing the resolution of seismic data, strong noise will cover the effective reflection wave, weakening the amplitude of the useful signal, resulting in a reduced clarity of the formation interface, generating unclear false images, and being unable to identify formation details; affecting seismic wave correction, noise mixing will change the actual amplitude and frequency characteristics of the waveform, increasing the correction difficulty, resulting in waveform mismatch before and after correction, and generating errors; endangering the seismic migration results, the migration algorithm is very sensitive to signals and noise, noise will cause migration, generating misaligned formation positions, and being unable to perform accurate imaging; reducing the accuracy of seismic interpretation; strong noise interference will make it impossible for interpreters to accurately interpret formation information, increasing the difficulty of interpretation, and reducing the result reliability; affecting the seismic inversion results, noise will cause deviations in parameter selection during the inversion process, generating inaccurate rock physical property information; misleading geological structure interpretation, strong noise is easily misjudged as a geological structure, increasing the difficulty of geological interpretation, and possibly obtaining wrong conclusions, and so on.

[0003] Currently, the commonly used signal processing methods for improving the signal-to-noise ratio of seismic data mainly include:

[0004] (1) Wavelet transform denoising method

[0005] This method uses wavelet analysis to extract signal features, designs a wavelet function suitable for the statistical characteristics of the signal, and realizes targeted denoising. The advantages are good denoising effect and high fidelity, and the disadvantage is that it is difficult to design the wavelet function;

[0006] (2) Wavelet packet transform denoising method

[0007] This method uses an adaptive wavelet packet basis to perform multi-resolution analysis on the signal, extracts useful information, and realizes adaptive filtering denoising. The advantages are high automation degree, and the disadvantage is large computational amount;

[0008] (3) Neural network denoising method

[0009] This method uses the self-learning ability of neural networks to model complex noise and achieve intelligent denoising. Its advantage is strong adaptability, but its disadvantage is that it requires a large number of samples for training;

[0010] (4) Statistical signal processing denoising method

[0011] This method uses high-order statistical features to analyze signal characteristics and designs statistical filters to achieve denoising. The advantage is that the algorithm is simple and effective, but the disadvantage is that it is difficult to balance the denoising effect with the damage to the useful signal;

[0012] (5) Adaptive filtering denoising method

[0013] This method uses an adaptive algorithm to adjust the filter parameters to adapt to changes in signal statistical characteristics. Its advantage is strong real-time performance, but its disadvantage is that the algorithm is complex and difficult to implement.

[0014] These methods have their own advantages and disadvantages, but overall there is still room for improvement in denoising effect and fidelity. Often, while removing noise, information details are lost and effective signals cannot be well maintained, and the denoising effect is difficult to meet the requirements. It is difficult for existing technologies to completely remove the influence of strong noise, and the signal-to-noise ratio of the data is still low, especially in complex noise environments. The effect is even worse, and it may cause more or less damage to the effective signal, resulting in errors in the interpretation results. Low fidelity is an obvious defect of current technology. In short, there is still a lot of room for improvement in denoising effect and fidelity.

[0015] Existing technologies still have the following problems to a greater or lesser extent: poor adaptability. Existing technologies have different denoising effects on noises of different types and complexity, and are not universal, requiring manual parameter selection; difficult parameter selection. The parameter selection of the signal processing algorithm has a decisive influence on the denoising effect, but it is very difficult to select the best parameters; low computational efficiency. Some algorithms are highly complex, require high computer performance, and have slow computational speed; low automation. Most algorithms require manual selection and design of filter parameters, and low automation, etc.

[0016] All these have resulted in the limitation of denoising through signal processing in practical applications. SUMMARY OF THE INVENTION

[0017] In view of this, the present invention discloses a high-fidelity seismic data denoising scheme that removes interference while maintaining effective wave detail information.

[0018] According to one aspect of the present invention, a seismic data denoising method based on mathematical morphology is proposed, the method comprising:

[0019] Step 1, obtain seismic data;

[0020] Step 2, constructing a radial structural element, wherein the radial structural element is composed of two orthogonal conventional structural elements;

[0021] Step 3, perform opening and closing operations on the seismic data using the radial structuring element to obtain the opening-closing operation result;

[0022] Step 4, perform morphological filtering on the opening-closing operation result to obtain the denoised seismic signal.

[0023] In some embodiments, the shape of the conventional structuring element is linear, elliptical or triangular.

[0024] In some embodiments, the radial structuring element is constructed from the spatio-temporal domain according to the following formula:

[0025]

[0026] where, T denotes transpose, the column vector b i =[b i (τ)] and b j =[b j (τ)] are two conventional structuring elements that are orthogonal in the radial direction, follow the information of the reflection event axis in the horizontal direction, and correspond to the normal reflectivity in the vertical direction.

[0027] In some embodiments, the specific steps of Step 3 include:

[0028] Perform opening and closing operations on the seismic signal using the radial structuring element based on the following formula to obtain the opening-closing operation result

[0029]

[0030] where, d represents a trace of seismic data, denotes performing an opening operation first and then a closing operation on the seismic data d using the structuring element , denotes performing a closing operation first and then an opening operation on the seismic data d using the structuring element , and then taking the average of the above two combined operations as the opening-closing operation result

[0031] In some embodiments, the specific steps of Step 4 include performing morphological filtering on the opening-closing operation result based on the following formula to obtain the denoised seismic signal d':

[0032]

[0033] where, d represents a trace of seismic data, denotes using the radial structuring element The opening and closing operation result obtained by performing opening and closing operations on seismic data d represents the scale of the radial structuring element .

[0034] In some embodiments, the method further includes:

[0035] According to the obtained interference noise trajectory, adaptively adjusting the scale and direction of the radial structuring element so that when performing the morphological filtering process, the interference noise trajectory is always parallel to the direction of the large scale in the radial structuring element.

[0036] According to one aspect of the present invention, there is also provided a seismic data denoising device based on mathematical morphology, the device includes:

[0037] A seismic data acquisition unit for acquiring seismic data;

[0038] A radial structuring element construction unit for constructing a radial structuring element, the radial structuring element being composed of two orthogonal conventional structuring elements;

[0039] An opening and closing calculation unit for performing opening and closing operations on the seismic data using the radial structuring element to obtain an opening and closing operation result;

[0040] A filtering unit for performing morphological filtering on the opening and closing operation result to obtain a denoised seismic signal.

[0041] In some embodiments, the shape of the conventional structuring element is linear, elliptical or triangular.

[0042] In some embodiments, the radial structuring element construction unit is specifically used to construct the radial structuring element from the spatio-temporal domain according to the following formula:

[0043]

[0044] wherein, T represents transpose, and the column vector b i = [b i (τ)] and b j = [b j (τ)] are two conventional structuring elements orthogonal in the radial direction, following the information of the reflection event axis in the horizontal direction and corresponding to the normal reflectivity in the vertical direction.

[0045] In some embodiments, the opening and closing calculation unit is specifically used to perform opening and closing operations on the seismic signal using the radial structuring element based on the following formula to obtain an opening and closing operation result

[0046]

[0047] Among them, d represents a trace of seismic data, represents performing an opening operation first and then a closing operation on the seismic data d using the structuring element and then performing a closing operation first and then an opening operation on the seismic data d using the structuring element represents using the structuring element and taking the average value of the above two combined operations as the result of the opening and closing operation

[0048] In some embodiments, the filtering unit is specifically configured to perform morphological filtering on the result of the opening and closing operation based on the following formula to obtain the denoised seismic signal d':

[0049]

[0050] Among them, d represents a trace of seismic data, represents the result of the opening and closing operation obtained by performing the opening and closing operations on the seismic data d using the radial structuring element and represents the scale of the radial structuring element b.

[0051] In some embodiments, the device further includes an adaptive adjustment unit, configured to adaptively adjust the scale and direction of the radial structuring element according to the acquired interference noise trajectory, so that when performing the morphological filtering process, the interference noise trajectory is always parallel to the direction of the large scale in the radial structuring element.

[0052] According to another aspect of the present invention, an electronic device is further provided, and the electronic device includes:

[0053] a memory storing executable instructions;

[0054] a processor, and the processor runs the executable instructions in the memory to implement the above-mentioned method for denoising seismic data based on mathematical morphology.

[0055] According to another aspect of the present invention, a computer-readable storage medium is further provided, and the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned method for denoising seismic data based on mathematical morphology.

[0056] According to the technical solution of the present invention, mathematical morphology is applied to seismic data processing. Through mathematical morphological operations, the characteristics and advantages of noise can be accurately identified. Combining the characteristics of seismic data, the seismic wave field information is detected using structuring elements with specified shapes, so as to achieve the purpose of high-fidelity denoising while removing interference and maintaining the detailed information of effective waves. The technical solution has at least the following advantages:​

[0057] (1) Good denoising effect

[0058] The technical solution of the present invention makes full use of the morphological operations in mathematical morphology, can accurately identify the morphological characteristics of noise, uses a structural element with a specified shape to match the noise, and effectively removes the noise signal. Especially for linear noise with strong energy, the denoising effect is remarkable;

[0059] (2) High fidelity

[0060] Traditional denoising methods usually damage the detailed information of useful signals while filtering out noise. However, denoising by mathematical morphology realizes denoising by matching the noise morphology, can retain useful signals to the greatest extent, and has high fidelity;

[0061] (3) The adopted algorithm is simple

[0062] Compared with complex adaptive filtering algorithms, the mathematical morphology filtering algorithm is simple and direct, based on basic morphological operations, easy to implement, has a small amount of calculation, and high operation efficiency;

[0063] (4) High degree of automation

[0064] After obtaining the interference noise trajectory, it can adaptively adjust the scale and direction of the radial structural element to achieve filtering, without the need for manual design of the filter, and has a high degree of automation;

[0065] (5) Strong noise adaptability

[0066] By designing structural elements of different morphologies, according to the technical solution of the present invention, different types and scales of noise can be matched for denoising, with strong pertinence and good adaptability;

[0067] (6) Simple parameters

[0068] The main parameters involved in the present invention are limited, such as the shape and size of the structural element, etc. There is no need for a complex parameter selection process and it is simple to use;

[0069] (7) Improve the interpretation effect

[0070] It can effectively improve the signal-to-noise ratio, increase the characteristics of effective signals, and improve the interpretation effect and result accuracy of seismic data;

[0071] (8) Wide range of extended applications

[0072] This technology can be widely applied to the noise reduction of seismic data, and has broad application prospects.

[0073] The methods and apparatuses of the present invention have other characteristics and advantages, which will be apparent in the accompanying drawings and the following detailed description incorporated herein, or will be described in detail in the accompanying drawings and the following detailed description incorporated herein. The accompanying drawings and the detailed description together are used to explain the specific principles of the present invention. Description of the Drawings

[0074] The above and other objects, features, and advantages of the present invention will become more apparent by describing the exemplary embodiments of the present invention in more detail in conjunction with the accompanying drawings, wherein, in the exemplary embodiments of the present invention, the same reference numerals generally represent the same components.

[0075] Figure 1 The flowchart of a seismic data denoising method based on mathematical morphology according to an embodiment of the present invention is shown.

[0076] Figure 2 (a), (b), and (c) show schematic diagrams of the shapes of conventional structural elements according to embodiments of the present invention.

[0077] Figure 3 The schematic diagram of the results of mathematical morphology filtering of seismic data using different structural elements according to an embodiment of the present invention is shown.

[0078] Figure 4 (a), (b), and (c) show schematic diagrams of images of radial structural elements constructed according to embodiments of the present invention.

[0079] Figure 5 (a), (b), and (c) show schematic diagrams of noise suppression according to a certain prior art.

[0080] Figure 6 (a), (b), and (c) show schematic diagrams of noise suppression according to embodiments of the present invention. Detailed Description of the Embodiments

[0081] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0082] The conceptual background of the present invention is introduced below.

[0083] The theoretical basis of the morphological filtering adopted by the present invention is mathematical morphology.

[0084] Mathematical morphology is an image analysis method based on set theory and logical operations. It describes and analyzes the geometric structure of an image by extracting its morphological features. Mathematical morphology was first developed in the 1960s by the brothers Georges Matheron and Fernand Matheron, who were tutors at the French telecommunications company. After half a century of development, it has become an important theory and technology in the fields of image processing and computer vision.

[0085] The basic idea of mathematical morphology is to use a structuring element to probe the image. By shifting the structuring element and interacting with the image, morphological information of the image can be extracted. There are four common basic operations in mathematical morphology: erosion, dilation, opening, and closing. These operations extract the shape features of the image based on completely different algorithms and can be used alone or in combination.

[0086] The basic transformations of mathematical morphology for binary images are the above four: namely, erosion, dilation, opening, and closing. They can be combined into complex image processing techniques. Let A be the set of the target image and B be the set of the structuring element. Then

[0087] The erosion operation is defined as:

[0088]

[0089] The dilation operation is defined as:

[0090]

[0091] The opening operation is defined as:

[0092]

[0093] The closing operation is defined as:

[0094]

[0095] The opening operation first performs erosion and then dilation, which plays a role in separation and filtering. Isolated parts smaller than the structuring element will be filtered out to suppress signal peak (positive pulse) noise; the closing operation first performs dilation and then erosion, which plays a role in filling gaps and internal connection to suppress signal trough (negative pulse) noise. In actual operations, the closing operation is often used to eliminate small dark details compared with the structuring element while keeping the overall gray level of the image and large dark regions basically unaffected. Specifically, in the closing operation, the dilation in the first step removes small dark details and at the same time enhances the image brightness, and the erosion in the second step weakens (basically restores) the image brightness but does not reintroduce the details removed earlier.

[0096] Compared with traditional image processing methods based on gray-scale statistics, mathematical morphology has several prominent advantages:

[0097] 1. Preserve the basic shape features of the image and extract the geometric structure information of the image;

[0098] 2. The algorithm is simple and direct, with a small amount of calculation and is easy to be implemented in hardware;

[0099] 3. Structural elements of different shapes can be designed to match different shape features;

[0100] 4. Morphological operations can be flexibly combined to achieve complex image analysis tasks.

[0101] Mathematical morphology has been widely applied, including image enhancement, segmentation, compression, shape detection, edge detection, filtering and denoising, etc., and has gradually been extended to 3D and high-dimensional image processing. In recent years, with the development of deep learning in the field of computer vision, the theories and algorithms based on mathematical morphology have also obtained new developments, such as morphological layers and morphological pooling, etc., showing great application potential.

[0102] When applying mathematical morphology to filtering, the basic idea is to transform, measure and extract the corresponding shapes in the target image with structural elements of a certain morphology, simplify the image data, preserve the basic shape characteristics of the image, and exclude irrelevant graphic structures, so as to achieve the purpose of morphological filtering.

[0103] The present invention introduces mathematical morphology into seismic data processing, designs a radial structural element to match the noise signal, and realizes effective filtering and denoising. This reflects the innovative application of the mathematical morphology theory in the field of signal processing, explores and utilizes the shape matching advantage of mathematical morphology, achieves remarkable noise reduction effects, and shows the great application value of this technology in the field of seismic data processing.

[0104] Example 1

[0105] Figure 1 The flowchart of a seismic data denoising method based on mathematical morphology according to an embodiment of the present invention is shown. As shown in the figure, the method includes Step 1 to Step 4.

[0106] Step 1, obtain seismic data.

[0107] The seismic data can be obtained according to the technical means considered applicable by those skilled in the art, and the seismic data can also be preprocessed according to the technical means considered applicable by those skilled in the art before performing the subsequent steps.

[0108] Step 2, construct a radial structural element, and the radial structural element is composed of two orthogonal conventional structural elements.

[0109] Appropriate conventional structural elements can be selected to construct the radial structural element.

[0110] In some embodiments, the inventors select structural elements having linear, elliptical, and / or triangular shapes according to the characteristics of seismic signals and noise.

[0111] The shape of the structural element plays a crucial role in the filtering result.

[0112] Essentially, a structural element consists of discrete points, which often have certain specific forms or satisfy certain functions. Generally speaking, a structural element can be discrete points of any form, and different forms of structural elements can obtain filtering results with different meanings. During use, it is necessary to select a suitable structural element according to the specific target set, and the form of the structural element will cause the result of mathematical morphology filtering to shift in a certain direction to some extent.

[0113] In the field of seismic exploration, according to the characteristics of seismic data, the inventors select the above-mentioned structural elements having linear, elliptical, and / or triangular shapes and being non-concave functions with axial symmetry.

[0114] The expressions of these three structural elements are shown in the following formulas (1), (2), and (3) respectively, and the shapes of the three structural elements are as shown in Figure 2 (a), (b), and (c), where Figure 2 (a) is a linear structural element, Figure 2 (b) is an elliptical structural element, Figure 2 (c) is a triangular structural element.

[0115] SE straight = A (1)

[0116]

[0117]

[0118] In the above formula, A represents the amplitude of the structural element, and W represents the number of sampling points of the structural element.

[0119] A structural element is usually composed of shape, amplitude, and the number of sampling points. As shown above, for the amplitude of the structural element, corresponding adjustments can be made according to the amplitude value of each seismic data trace. And the number of sampling points of the structural element, usually also called the scale of the structural element, is generally selected as an odd number. Figure 2 (a), (b), and (c) show the maximum amplitudes of the three structural elements are all set to 1, and the scales of their structural elements are all 21.

[0120] Figure 3 Shows a schematic diagram of the result of mathematical morphology filtering of seismic data using different structural elements according to an embodiment of the present invention. In the figure, the blue curve is a certain seismic data trace, that is, the target set; the red curve, black curve, and green curve are respectively obtained by usingFigure 2 The results of mathematical morphological filtering for the three structural elements in (a), (b), and (c). It can be seen that different structural elements lead to relatively obvious differences in the results of mathematical morphological filtering.

[0121] In some embodiments, the radial structural element can be constructed from the spatio-temporal domain according to the following formula:

[0122]

[0123] where, T represents transpose, and the column vector b i = [b i (τ)] and b j = [b j (τ)] are two conventional structural elements that are orthogonal in the radial direction, following the information of the reflection event axis in the horizontal direction and corresponding to the normal reflectivity in the vertical direction.

[0124] To improve the noise suppression effect and signal amplitude preservation, we construct the radial structural element from the spatio-temporal domain according to the interference trajectory characteristics. The radial structural element is composed of two conventional structural elements in two orthogonally radial directions, which are 90 degrees to each other, following the information of the reflection event axis in the horizontal direction and corresponding to the normal reflectivity in the vertical direction, so as to extract the main reflection information. Therefore, the seismic trace morphology information can be detected in two directions to obtain the correlation of adjacent traces.

[0125] Figure 4 (a), (b), and (c) show schematic diagrams of the images of the radial structural elements constructed according to an exemplary embodiment of the present invention. Among them, Figure 4 (a) shows a schematic diagram of the image of a conventional elliptical structural element; Figure 4 (b) shows a schematic diagram of the image of a conventional triangular structural element; Figure 4 (c) shows a schematic diagram of the image of a radial structural element composed of an orthogonal elliptical structural element and a triangular structural element.

[0126] Return to Figure 1 , step 3, use the radial structural element to perform opening and closing operations on the seismic data to obtain the opening and closing operation results.

[0127] In some embodiments, based on the following formula, use the radial structural element to perform opening and closing operations on the seismic signal to obtain the opening and closing operation results

[0128]

[0129] where, d represents a seismic data trace, represents using the structural element Perform an opening-then-closing operation on the seismic data d, which means using a structuring element to perform a closing-then-opening operation on the seismic data d, and then taking the average of the above two combined operations as the result of the opening-closing operation

[0130] Step 4: Perform morphological filtering on the result of the opening-closing operation to obtain a denoised seismic signal.

[0131] Although traditional morphological filtering uses the morphological information along the time direction in each single trace to separate signals and noise and has achieved some effects, morphological filtering based on a radial structuring element separates signals and noise from the spatio-temporal domain perspective and has better effects.

[0132] In some embodiments, according to the obtained interference noise trajectory, adaptively adjust the scale and direction of the radial structuring element so that when performing the morphological filtering process, the interference noise trajectory is always parallel to the direction of the large scale in the radial structuring element.

[0133] In some embodiments, step 4 specifically includes performing morphological filtering on the result of the opening-closing operation based on the following formula to obtain a denoised seismic signal d':

[0134]

[0135] where d represents a trace of seismic data, represents using the radial structuring element the result of the opening-closing operation obtained by performing the opening and closing operations on the seismic data d, represents the scale of the radial structuring element b.

[0136] According to this embodiment, if a trace of seismic data d, the result of the opening-closing operation calculated in step 3 is greater than the scale of the radial structuring element then the signal output after filtering is If the result of the opening-closing operation calculated for the seismic data d in step 3 is less than the scale of the radial structuring element then the signal output after filtering is the scale of then the signal output after filtering is The two-dimensional morphological filtering of the radial structuring element allows large-scale components to pass through and rejects small-scale components from passing through, thereby dividing the input data into two subspaces.

[0137] This filter is similar to a "large-scale pass" filter with morphological scales. By using the proposed morphological filtering based on a radial structuring element to suppress both coherent and random noise simultaneously, the difference in morphological scales between the signal and other unwanted energy from the horizontal and vertical directions can be utilized. During the suppression process of interference, we use the obtained interference noise trajectory to adaptively select the scale and direction of the radial structuring element at different spatio-temporal positions, such that the interference noise trajectory is always parallel to the large-scale direction in the radial structuring element, in order to obtain the best filtering effect. Since most of the information of the interference noise propagates along the large-scale direction in the radial structuring element and will be stored in the larger morphological scale band, while other signals, such as primary reflection waves, will be stored in the smaller morphological scale frequency band, the morphological filtering based on the radial structuring element can make full use of the difference in waveform shapes between the signal and the noise.

[0138] The method according to this embodiment applies mathematical morphology to seismic data processing. Through mathematical morphological operations, the characteristics and advantages of noise can be accurately identified. Combining the characteristics of seismic data, the seismic wavefield information is detected using structuring elements with a specified shape, so as to achieve the purpose of high-fidelity denoising while removing interference and maintaining the detailed information of the effective wave. This embodiment also has beneficial effects such as simple algorithm, strong noise adaptability, high degree of automation, and convenient design of implementation parameters, etc.

[0139] Example 2

[0140] According to an embodiment of the present invention, a seismic data denoising device based on mathematical morphology is provided. The device includes: a seismic data acquisition unit for acquiring seismic data; a radial structuring element construction unit for constructing a radial structuring element, which is composed of two orthogonal conventional structuring elements; an opening and closing calculation unit for performing opening and closing operations on the seismic data using the radial structuring element to obtain an opening and closing operation result; and a filtering unit for performing morphological filtering on the opening and closing operation result to obtain a denoised seismic signal.

[0141] In some embodiments, the shape of the conventional structuring element is linear, elliptical or triangular.

[0142] In some embodiments, the radial structuring element construction unit is specifically used to construct the radial structuring element from the spatio-temporal domain according to the following formula:

[0143]

[0144] where, T represents the transpose, and the column vector b i =[b i (τ)] and b j =[b j(τ) are two conventional structural elements that are orthogonally radial, following the information of the reflection event axis in the horizontal direction and corresponding to the normal reflectivity in the vertical direction.

[0145] In some embodiments, the opening and closing calculation unit is specifically configured to perform opening and closing operations on the seismic signal using the radial structural element based on the following formula to obtain an opening and closing operation result

[0146]

[0147] where d represents a trace of seismic data, represents performing an opening operation first and then a closing operation on the seismic data d using the structural element and represents performing a closing operation first and then an opening operation on the seismic data d using the structural element and then taking the average of the above two combined operations as the opening and closing operation result

[0148] In some embodiments, the filtering unit is specifically configured to perform morphological filtering on the opening and closing operation result based on the following formula to obtain a denoised seismic signal d':

[0149]

[0150] where d represents a trace of seismic data, represents the opening and closing operation result obtained by performing opening and closing operations on the seismic data d using the radial structural element and represents the scale of the radial structural element

[0151] In some embodiments, the device further includes an adaptive adjustment unit configured to adaptively adjust the scale and direction of the radial structural element according to the acquired interference noise trajectory so that when performing the morphological filtering, the interference noise trajectory is always parallel to the direction of the large scale in the radial structural element.

[0152] The device according to this embodiment applies mathematical morphology to seismic data processing. Through mathematical morphological operations, the characteristics and advantages of noise can be accurately identified. Combining the characteristics of seismic data, the seismic wavefield information is detected using structural elements with specified shapes, thereby achieving the purpose of high-fidelity denoising while removing interference and maintaining the detailed information of effective waves. This embodiment also has beneficial effects such as simple algorithm, strong noise adaptability, high automation degree, and convenient design of implementation parameters.

[0153] ​For other detailed descriptions and advantages of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, which will not be elaborated herein.

[0154] Example 3

[0155] According to another aspect of the present invention, an electronic device is further provided. The electronic device includes:

[0156] A memory storing executable instructions:

[0157] A processor that runs the executable instructions in the memory to implement the seismic data denoising method based on mathematical morphology according to the present invention.

[0158] The method includes the following steps:

[0159] Step 1, obtaining seismic data;

[0160] Step 2, constructing a radial structuring element, which is composed of two orthogonal conventional structuring elements;

[0161] Step 3, performing opening and closing operations on the seismic data using the radial structuring element to obtain an opening-closing operation result;

[0162] Step 4, performing morphological filtering on the opening-closing operation result to obtain a denoised seismic signal.

[0163] In some embodiments, the shape of the conventional structuring element is linear, elliptical or triangular.

[0164] In some embodiments, the radial structuring element is constructed from the spatio-temporal domain according to the following formula:

[0165]

[0166] where, T represents transpose, and the column vectors b i = [b i (τ)] and b j = [b j (τ)] are two conventional structuring elements that are orthogonal in the radial direction, follow the information of the reflection event axis in the horizontal direction, and correspond to the normal reflectivity in the vertical direction.

[0167] In some embodiments, step 3 specifically includes:

[0168] Performing opening and closing operations on the seismic signal using the radial structuring element based on the following formula to obtain an opening-closing operation result

[0169]

[0170] Among them, d represents a trace of seismic data, represents performing an opening operation first and then a closing operation on the seismic data d with the structuring element and represents performing a closing operation first and then an opening operation on the seismic data d with the structuring element and then taking the average value of the above two combined operations as the result of the opening-closing operation.

[0171] In some embodiments, step 4 specifically includes performing morphological filtering processing on the result of the opening-closing operation based on the following formula to obtain the denoised seismic signal d':

[0172]

[0173] Among them, d represents a trace of seismic data, represents the result of the opening-closing operation obtained by performing opening and closing operations on the seismic data d using the radial structuring element and represents the scale of the radial structuring element .

[0174] In some embodiments, the method further includes:

[0175] Adapting to adjust the scale and direction of the radial structuring element according to the acquired interference noise trajectory, so that when performing the morphological filtering processing, the interference noise trajectory is always parallel to the direction of the large scale in the radial structuring element.

[0176] Specifically, the memory may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0177] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In an embodiment of the present invention, the processor is used to run the computer-readable instructions stored in the memory.

[0178] Applying mathematical morphology to seismic data processing according to this embodiment, the characteristics and advantages of noise can be accurately identified through mathematical morphological operations. Combining the characteristics of seismic data, the seismic wavefield information is detected using a structural element with a specified shape, so as to achieve the purpose of high-fidelity denoising that can remove interference while maintaining the detailed information of effective waves. This embodiment also has beneficial effects such as simple algorithm, strong noise adaptability, high degree of automation, and convenient design of implementation parameters, etc.

[0179] For the detailed description of this embodiment, reference can be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0180] Example 4

[0181] According to another aspect of the present invention, there is also provided a computer-readable storage medium storing a computer program, which when executed by a processor implements the method for denoising seismic data based on mathematical morphology according to the present invention.

[0182] The method includes the following steps:

[0183] Step 1, obtaining seismic data;

[0184] Step 2, constructing a radial structural element, which is composed of two orthogonal conventional structural elements;

[0185] Step 3, performing opening and closing operations on the seismic data using the radial structural element to obtain the result of the opening and closing operations;

[0186] Step 4, performing morphological filtering on the result of the opening and closing operations to obtain a denoised seismic signal.

[0187] In some embodiments, the shape of the conventional structural element is linear, elliptical or triangular.

[0188] In some embodiments, the radial structural element is constructed from the spatio-temporal domain according to the following formula:

[0189]

[0190] where, T denotes transpose, and the column vectors b i =[b i (τ)] and b j =[b j (τ)] are two conventional structural elements orthogonal in the radial direction, following the information of the reflection event axis in the horizontal direction and corresponding to the normal reflectivity in the vertical direction.

[0191] In some embodiments, step 3 specifically includes:

[0192] Use the radial structural element to perform opening and closing operations on the seismic signal according to the following formula to obtain the opening and closing operation result

[0193]

[0194] where d represents a trace of seismic data, represents performing an opening operation first and then a closing operation on the seismic data d using the structural element and represents performing a closing operation first and then an opening operation on the seismic data d using the structural element After that, take the average of the above two combined operations as the opening and closing operation result

[0195] In some embodiments, step 4 specifically includes performing morphological filtering on the opening and closing operation result according to the following formula to obtain the denoised seismic signal d':

[0196]

[0197] where d represents a trace of seismic data, represents the opening and closing operation result obtained by performing opening and closing operations on the seismic data d using the radial structural element and represents the scale of the radial structural element

[0198] In some embodiments, the method further includes:

[0199] According to the obtained interference noise trajectory, adaptively adjust the scale and direction of the radial structural element so that when performing the morphological filtering, the interference noise trajectory is always parallel to the direction of the large scale in the radial structural element.

[0200] According to an embodiment of the present invention, a computer-readable storage medium stores non-temporary computer-readable instructions. When the non-temporary computer-readable instructions are run by a processor, all or part of the steps of the methods of the various embodiments of the present invention described above are executed.

[0201] The above computer-readable storage medium includes but is not limited to: optical storage media (such as CD-ROMs and DVDs), magneto-optical storage media (such as MOs), magnetic storage media (such as tapes or external hard drives), media with built-in rewritable non-volatile memories (such as memory cards), and media with built-in ROMs (such as ROM cartridges).

[0202] ​Those skilled in the art should be able to understand that, in order to solve the technical problem of how to obtain good user experience effects, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included in the protection scope of the present invention.

[0203] According to this embodiment, mathematical morphology is applied to seismic data processing. Through mathematical morphological operations, the characteristics and advantages of noise can be accurately identified. Combining the characteristics of seismic data, the seismic wavefield information is detected with a structural element of a specified shape, so as to achieve the purpose of high-fidelity denoising while removing interference and maintaining the detailed information of effective waves. This embodiment also has beneficial effects such as simple algorithm, strong noise adaptability, high degree of automation, and convenient design of implementation parameters, etc.

[0204] For the detailed description of this embodiment, reference can be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0205] Example 5

[0206] In order to verify the effect of the seismic data denoising scheme based on mathematical morphology according to the present invention, this embodiment selects a large ship interference environment for verification.

[0207] Figure 5 (a), (b) and (c) show the schematic diagrams of noise suppression according to a certain prior art. Figure 5 (a) is the original trace gather data. It can be seen that the seismic signals are interfered by large ships with strong energy, and they overlap with each other and have a complex morphology. It can be seen that this data has suffered serious damage from large ship interference, and the waveform characteristics of large ship interference are quite complex. Figure 5 (b) shows the result after denoising with a certain existing commercial software. Figure 5 (c) shows the differential trace gather. It can be seen that the noise suppression is not very thorough.

[0208] Figure 6 (a), (b) and (c) show the schematic diagrams of noise suppression based on mathematical morphology according to the present invention. Figure 6 (a) is the same as Figure 5 (a) the original trace gather data. Figure 6 (b) gives the display diagram of the seismic data after denoising according to the technical solution of the present invention. Figure 6 (c) shows all the large ship interferences extracted according to the present invention. It can be seen that almost all the large ship interferences are extracted according to the technical solution of the present invention, the main noise energy is removed, and the background signal is not damaged.

[0209] This example fully shows that the present invention can effectively remove noise signals, especially strong-energy linear noise, with remarkable denoising effect, and can retain the detailed information of effective waves to the greatest extent with high fidelity.

[0210] For other detailed descriptions of this exemplary embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, which will not be repeated herein.

[0211] The embodiments of the present invention have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, the practical application, or the technical improvement of the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A seismic data denoising method based on mathematical morphology, characterized in that: The method comprises: Step 1, obtaining seismic data; Step 2, constructing a radial structural element, wherein the radial structural element is composed of two orthogonal conventional structural elements; Step 3, using the radial structural element to perform opening and closing operations on the seismic data to obtain opening and closing operation results; Step 4: Perform morphological filtering on the opening and closing operation results to obtain a denoised seismic signal.

2. The method according to claim 1, characterized in that The shape of the conventional structural element is linear, elliptical or triangular.

3. The method according to claim 1, characterized in that The radial structure element is constructed from the time-space domain according to the following formula: in,[] T represents transpose, column vector b i =[b i (τ)] and b j =[b j (τ)] are two radially orthogonal regular structure elements, following the information of the reflection event in the horizontal direction and corresponding to the normal reflectivity in the vertical direction.

4. The method according to claim 1, characterized in that: The step 3 specifically includes: Based on the following formula, the radial structural element is used to perform opening and closing operations on the seismic signal to obtain the opening and closing operation results: Among them, d represents a seismic data, Structural elements Perform an open operation and then a closed operation on the seismic data d. Structural elements The seismic data d is first closed and then opened, and then the average value of the above two combined operations is used as the opening and closing operation result.

5. The method according to claim 1, characterized in that The step 4 specifically includes performing morphological filtering on the opening and closing operation results based on the following formula to obtain the denoised seismic signal d ′ : Among them, d represents a seismic data, Indicates the use of radial structural elements The opening and closing operation results obtained after performing opening and closing operations on the seismic data d, Represents a radial structuring element scale.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: According to the obtained interference noise trajectory, the scale and direction of the radial structure element are adaptively adjusted so that when the morphological filtering process is performed, the interference noise trajectory is always parallel to the large-scale direction of the radial structure element.

7. A seismic data denoising device based on mathematical morphology, characterized in that: The device comprises: A seismic data acquisition unit, used for acquiring seismic data; A radial structure element construction unit, used to construct a radial structure element, wherein the radial structure element is composed of two orthogonal conventional structure elements; An opening and closing calculation unit, used for performing opening and closing operations on the seismic data using the radial structural element to obtain an opening and closing operation result; The filtering unit is used to perform morphological filtering on the opening and closing operation results to obtain a denoised seismic signal.

8. The device according to claim 1, characterized in that The shape of the conventional structural element is linear, elliptical or triangular.

9. The device according to claim 7, characterized in that The radial structure element construction unit is specifically used to construct the radial structure element from the time and space domain according to the following formula: in,[] T represents transpose, column vector b i =[b i (τ)] and b j =[b j (τ)] are two radially orthogonal regular structure elements, following the information of reflection events in the horizontal direction and corresponding to normal reflectivity in the vertical direction.

10. The device according to claim 7, characterized in that The opening and closing calculation unit is specifically used to perform opening and closing operations on the seismic signal using the radial structural element based on the following formula to obtain the opening and closing operation result: Among them, d represents a seismic data, Structural elements Perform an open operation and then a closed operation on the seismic data d. Structural elements The seismic data d is first closed and then opened, and then the average value of the above two combined operations is used as the opening and closing operation result.

11. The device according to claim 7, characterized in that The filtering unit is specifically used to perform morphological filtering on the opening and closing operation result based on the following formula to obtain a denoised seismic signal d′: Among them, d represents a seismic data, Indicates the use of radial structural elements The opening and closing operation results obtained after performing opening and closing operations on the seismic data d, Represents a radial structuring element scale.

12. The device according to any one of claims 7 to 11, characterized in that The device also includes an adaptive adjustment unit, which is used to adaptively adjust the scale and direction of the radial structure element according to the acquired interference noise trajectory, so that when the morphological filtering process is performed, the interference noise trajectory is always parallel to the large-scale direction of the radial structure element.

13. An electronic device, characterized in that: The electronic device comprises: A memory storing executable instructions; A processor, wherein the processor runs the executable instructions in the memory to implement the method according to any one of claims 1 to 6.

14. A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 6 when executed by a processor.

Images