CN114200525B - Self-adaptive multi-channel singular spectrum analysis seismic data denoising method - Google Patents
Self-adaptive multi-channel singular spectrum analysis seismic data denoising method Download PDFInfo
- Publication number
- CN114200525B CN114200525B CN202111508457.XA CN202111508457A CN114200525B CN 114200525 B CN114200525 B CN 114200525B CN 202111508457 A CN202111508457 A CN 202111508457A CN 114200525 B CN114200525 B CN 114200525B
- Authority
- CN
- China
- Prior art keywords
- singular
- seismic data
- matrix
- values
- singular value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000010183 spectrum analysis Methods 0.000 title claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims abstract description 47
- 230000001131 transforming effect Effects 0.000 claims abstract description 8
- 230000003044 adaptive effect Effects 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 8
- 238000009795 derivation Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 12
- 238000013016 damping Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000009467 reduction Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 230000001629 suppression Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000005452 bending Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Computing Systems (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The application discloses a self-adaptive multi-channel singular spectrum analysis seismic data denoising method, which comprises the following steps: step 1: presetting time domain seismic data, and converting the time domain seismic data into frequency domain seismic data through DFFT; and 2, step: generating a block Hankel matrix based on the frequency domain seismic data; and 3, step 3: decomposing the Hankel matrix to obtain a singular value matrix, and combining singular values in the singular value matrix into a singular value sequence; and 4, step 4: based on Akaike information criterion, transforming the singular value sequence to obtain the number of the singular values; and calculating and denoising the number of singular values to obtain denoised seismic data. The method and the device automatically determine the number of the proper singular values based on the Akaike information criterion, and are beneficial to the industrial realization of multi-channel singular spectrum analysis.
Description
Technical Field
The invention belongs to the technical field of seismic data denoising processing, and particularly relates to a self-adaptive multi-channel singular spectrum analysis seismic data denoising method.
Background
The multi-channel singular spectrum analysis (MSSA) is a denoising method based on rank reduction, which decomposes original data into a signal subspace and a noise subspace through singular value decomposition, then sets the energy of the noise subspace to be zero (truncation), and achieves the purpose of denoising through inverse transformation. MSSA was developed from univariate Singular Spectral Analysis (SSA), which was widely used as an unconstrained model method for singular spectral analysis of one-dimensional time series trace matrices. Read (1993) rates the first extension of SSA to MSSA for the study of multivariate MSSA methods. The low-rank Hankel matrix is formed by the frequency spectrum similarity and predictability of adjacent seismic channels on the basis of the assumption of linear homodyne axes, the low-rank structure of the Hankel matrix of the data frequency slice is damaged by the existence of noise, and a truncated singular value decomposition method is commonly used for solving the low-rank approximation problem. Trickett (2008) applies the method to seismic data noise suppression and popularizes the method to f-x-y three-dimensional data to attenuate random noise. Oropeza and Sacchi (2011) use MSSA to achieve simultaneous denoising and reconstruction in prestack three-dimensional data. Many numerical experiments show that random noise cannot be completely eliminated by using the MSSA algorithm, and the denoising effect of the MSSA algorithm has a great space for improvement. Huang et al (2015) propose a damped multi-channel singular spectral analysis (DMSSA) algorithm by introducing a damping operator into the conventional MSSA. By introducing a soft threshold moving average operator into a damped rank reduction framework, oboue et al (2020) use a method called robust damped rank reduction to fuse the advantages of the soft threshold moving average operator and the damped operator together, thereby improving the signal-to-noise ratio of the seismic data. The damping rank reduction method is an effective method, and can recover effective signals from observation data containing noise and incompleteness.
The existing rank reduction method is based on the basic assumption of seismic linear event, time domain signals are converted into frequency domain signals after discrete fast Fourier transform, and then rank reduction is achieved by truncating the same singular value number for all frequencies through a manual estimation method. For large-scale data, the method based on rank reduction needs to divide the seismic data into different blocks, however, the number of singular values corresponding to each block is different, the number of the singular values suitable for data estimation of each block needs to be manually estimated at present, the calculation efficiency is low, and industrialization cannot be realized.
Disclosure of Invention
The application provides a self-adaptive multi-channel singular spectrum analysis seismic data denoising method, which is used for automatically determining the number of proper singular values based on an Akaike information criterion and is beneficial to the industrial realization of multi-channel singular spectrum analysis.
In order to achieve the above purpose, the present application provides the following solutions:
a self-adaptive multi-channel singular spectrum analysis seismic data denoising method comprises the following steps:
step 1: presetting time domain seismic data, and converting the time domain seismic data into frequency domain seismic data through DFFT;
and 2, step: generating a block Hankel matrix based on the frequency domain seismic data;
and step 3: decomposing the Hankel matrix to obtain a singular value matrix, and combining singular values in the singular value matrix into a singular value sequence;
and 4, step 4: based on Akaike information criterion, the singular value sequence is transformed to obtain the number of the singular values; and calculating and denoising the number of singular values to obtain denoised seismic data.
Preferably, the method of step 2 comprises:
obtaining frequency slice data of the frequency domain seismic data for each frequency; generating a Hankel matrix based on each row of data of the frequency slice data; and arranging all the Hankel matrixes into one block Hankel matrix.
Preferably, the singular value matrix isWhereinIs composed ofThe rank of (c) is determined,is thatOf left singular value vector of (V) x ×L x )×(V x ×L x ) An orthogonal matrix of orders;is thatOf right singular value vector of (H) x ×L y )×(H x ×L y ) An orthogonal matrix of orders;is a diagonal matrix formed by singular values in descending order, the number of non-zero singular values is equal to the matrixIs the ith frequency.
Preferably, the method for transforming the singular value sequence based on Akaike information criterion to obtain the number of singular values includes:
performing second-order derivation on a singular value sequence curve to obtain the change rate of the slope of the singular value sequence curve; calculating Akaike information criterion values of the change rates of the slope of the singular value sequence curves of all the frequencies at a preset point based on the Akaike information criterion; and calculating to obtain the minimum value of the Akaike information criterion values based on all the Akaike information criterion values, wherein the minimum value is the number of singular values of the frequency domain seismic data.
Preferably, the Akaike information criterion value is
Preferably, the method for performing computational denoising on the number of singular values includes: and inversely transforming the singular value sequence to a frequency domain through the block Hankel matrix, and then inversely transforming to a time domain through Fourier transform to obtain the denoised seismic data.
The beneficial effect of this application does:
for the traditional MSSA denoising method, the determination of the number of effective singular values is the key of the method. The application provides a self-adaptive multi-channel singular spectrum analysis seismic data denoising method, which can automatically determine the number of reliable singular values for massive data based on an Akaike information criterion, and denoise by adopting a DMSSA method to obtain a denoising effect with a high signal-to-noise ratio. The validity and reliability of the automatic method provided by the application are proved by numerical verification, and the method has great potential in industrial application. In addition, in the new method provided by the application, the computer can automatically perform denoising processing only by determining the dominant frequency range of the wavelet.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings required to be used in the embodiments are briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic flow chart of a self-adaptive multi-channel singular spectrum analysis seismic data denoising method in an embodiment of the present application;
FIG. 2 (a) is a three-dimensional view of simulated seismic data with five seismic event axes, as shown in the example two of the present application;
FIG. 2 (b) is a three-dimensional view of simulated seismic data according to the second embodiment of the present application, wherein 10% of random noise is added to the simulated seismic data;
FIG. 3 is a diagram illustrating the first 30 singular value curves of a frequency of seismic data three-dimensional view data simulated in the second embodiment of the present application;
FIG. 4 (a) is a diagram illustrating the self-adaptive method for determining singular value quantity in the second embodiment of the present applicationA sequence graph diagram;
FIG. 4 (b) is a schematic diagram of a sequence curve in the determination of singular value quantity by the adaptive method according to the second embodiment of the present application;
fig. 5 (a) is a schematic diagram of a denoising result in a simulation data denoising effect in a second embodiment of the present application;
fig. 5 (b) is a schematic diagram of noise removed in a simulation data denoising effect according to a second embodiment of the present application;
FIG. 6 is a schematic diagram of an actual noisy two-dimensional seismic section in the third embodiment of the present application;
FIG. 7 is a schematic diagram illustrating a singular value estimation principle of two-dimensional seismic data according to a third embodiment of the present application;
fig. 8 (a) is a schematic diagram of a noise suppression result of actual data applied to the denoising effect of actual seismic data in the third embodiment of the present application;
fig. 8 (b) is a schematic diagram of noise removed in an actual seismic data denoising effect in the third embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, the present application is described in further detail with reference to the accompanying drawings and the detailed description.
The first embodiment is as follows: as shown in FIG. 1, assume that a noisy three-dimensional seismic datum is D (x, y, t), where x = (x) 1 ,…,x m ),y=(y 1 ,…,y n ),t=t 1 ,…t s Seismic data size of x m ×y n ×t s . According to the DMSSA theory, the data can be denoised using the following four steps.
In a first step, seismic data is transformed from the time domain to the frequency domain by DFFT, obtaining data F (x, y, w), where w = (w) 1 ,…,w j )。
Second step, within a given frequencyEach frequency slice data is arranged into a block Hankel matrix. When the frequency is equal to w i The data for the frequency slice is shown in equation (1):
the process of constructing the block Hankel matrix includes two steps. First, F (x, y, w) i ) Is constructed as a Hankel matrix:
wherein R is K Is of size V x ×H x The Hankel matrix of (1),H x =y n -V x +1, symbolRepresenting a rounding operation. R in formula (2) K Is represented by F (x, y, w) i ) The K-th row of (1) of the Hankel matrix. Then, all R's are combined K ,K=1,…,x m Arranged into a block Hankel matrixAs shown in equation (3):
third step ofSingular value decomposition is performed and singular values are selected and truncated, which is a key step of the multi-channel singular spectrum analysis-like method. If the number of the singular values corresponding to the effective signal is N, only the first N singular values are reserved in the singular value diagonal matrix, and the sizes of all other singular values are set to be zero. To pairSingular value decomposition is performed to obtain formula (4):
Wherein d isThe rank of (c) is determined,is thatOf left singular value vector of (V) x ×L x )×(V x ×L x ) An orthogonal matrix of orders;is thatOf right singular value vector of (H) x ×L y )×(H x ×L y ) An orthogonal matrix of orders;is a diagonal matrix formed by singular values in descending order, the number of non-zero singular values is equal to the matrixOf the rank of (c).
Specifically, when the acquired seismic data does not contain noise, the diagonal matrix of equation (5) contains only non-zero singular values associated with the significant signal. If the data contains noise, the magnitude of all singular values will change and the number of non-zero singular values will increase. The original MSSA method only reserves the number of singular values, and has no influence on the size of the singular values, so that the denoising result has a great improvement space.
The DMSSA method is to attenuate the singular value increments caused by noise by adding a damping factor. Equations (6) and (7) represent this process:
wherein T represents a damping operator, I is a unit matrix, N is the number of estimated singular values, D represents a damping factor, the smaller the value of D is, the stronger the damping effect is, otherwise, the weaker the damping effect is, and the formulas (6) and (7) are the cores of the DMSSA. The essence of DMSSA denoising is that the numerical value of the (N + 1) th singular value is amplified or reduced by using a damping factor D, then the numerical values are subtracted by using the first N singular values, and the singular values after the (N + 1) th singular value are set to be zero, so that the purpose of suppressing noise is achieved.
Determining N is the most critical step in DMSSA denoising, which will affect the effect of noise suppression and the degree of corruption to the effective signal. If N is chosen too small, the valid signal will be corrupted; if N is selected too large, the noise suppression effect is reduced, and for MSSA methods, how to automatically determine the number of singular values is the key. The derivation is carried out under the condition that N is a crucial parameter which is known, and the problem that how to determine the number of singular values to be reserved in a data block is worth researching when complex and variable actual seismic data have no other sufficient geological data is faced. The method automatically determines the number of proper singular values based on Akaike information criterion, and is favorable for industrialized realization of multi-channel singular spectrum analysis.
Specifically, the singular values in the singular value matrix are combined into a singular value sequenceFor singular value sequenceThe number of effective singular values can be determined according to the huge difference between the N-th singular value and the (N + 1) -th singular value and the severe bending phenomenon. In fact, the choice of the value of N is to detect the position of the inflection point in the sequence of singular values. An Akaike information criterion is given below to automatically determine the number of retained singular values. Firstly, singular value vectors are transformed as follows:
equation (8) is actually the second derivative of the singular value sequence curve,the rate of change of the slope of the singular value curve is described. Then at w i The Akaike information criterion value for calculating the L-th point under the frequency is as follows:
where var is the variance of the data sequence and d is the vectorOf the length of (c).Is a vector of length d.The position corresponding to the global minimum in the sequence is the inflection point position, and then the minimum in all frequencies is solved according to the formula (9)
The global minimum value in equation (10) is the number N of singular values that need to be preserved in the entire data block.
After the singular value is determined by using the method, the frame of the DMSSA method is adopted in the denoising process. The newly proposed algorithm is called an adaptive damped multichannel singular spectrum analysis (admsas) method. In the proposed new algorithm, the computer can automatically perform denoising processing only by determining the dominant frequency range of the wavelet.
And fourthly, calculating a denoised result based on the truncated singular value. And performing inverse transformation on the frequency domain through a Hankel matrix, and performing inverse Fourier transformation on the frequency domain and the time-space domain to obtain the de-noised seismic data. The process can be expressed as
And performing the operation on all the frequencies to obtain the denoised seismic data.
Example two: to verify the validity and feasibility of the present application, a simulation data experiment was used to illustrate.
FIG. 2 is a simulated three dimensional seismic data having five seismic event axes of different dip angles, a data dimension of 300X 60, 300 samples over time, a sampling rate of 2ms, 60 samples in both the inline and crossline directions, FIG. 2 (a) is data without noise, FIG. 2 (b) is data with 10% random noise added, and a signal-to-noise ratio (SNR) of-1.322 dB. The wavelets used for the analog data are Rake wavelets with a dominant frequency of 40Hz, and the signal-to-noise ratio (SNR) is defined as:
where d is noise-free data, r is denoised data, the symbol | | 2 Representing the L2 norm.
Fig. 3 is a singular value of a block Hankel matrix corresponding to a certain frequency in the simulation data, a solid line in fig. 3 corresponds to a singular value of noiseless data, a dotted line corresponds to a singular value of data after noise is added, and a dot-dash line is a singular value of data after de-noising by the DMSSA method; the solid line in fig. 3 shows that the seismic data only has a few large singular values in the absence of noise, the number of the singular values is equal to N, and other singular values are all small, so that the singular value curve has a remarkable buckling phenomenon. Theoretically, the singular values after N should all be zero in the absence of noise. After the noise is added to the data, the magnitude of the singular value is changed and a large number of non-zero singular values also appear, so that the curve of the singular value after the noise is added is relatively gentle in decline, but a huge drop still exists between the Nth item and the (N + 1) th item.
Fig. 4 (a) and 4 (b) are calculation results of analog data using Akaike information criterion. In FIG. 4 (a), the solid line represents a singular value curve at a certain frequency, and the dotted line represents the corresponding AIC wi A curve. The minima occur at the N =8 dashed line. Since the numerical result in the main frequency range is stable, and abnormal values often appear in other ranges, in order to improve the precision, the control frequency is within 10-90Hz, the number of the singular values determined in the frequency range is selected by the self-adaptive method in the step (b) in FIG. 4, and as can be seen from the step (b) in FIG. 4, the number of the singular values determined in the self-adaptive manner is 8, and the result close to the true value N can be automatically estimated when the noise-containing data is processed by the method based on the Akaike information criterion.
Fig. 5 (a) and 5 (b) show the processing results of fig. 2 (b) by using the admsas method, where the number of singular value reservations determined by the adaptive method shown in fig. 4 (a) and 4 (b) is 8, and the damping factor D =3, respectively show the results in fig. 5 (a), where the signal-to-noise ratio is 22.110dB; fig. 5 (b) shows the removed noise. The self-adaptive algorithm can obtain a result which is closer to an accurate N value, obtain higher signal-to-noise ratio data, have clear coherence of a same phase axis and reserve local details.
Example three: in order to verify the effectiveness and feasibility of the application, an actual data experiment is used for illustration.
In order to prove the denoising effect of the algorithm provided by the application on the actual seismic data, a two-dimensional post-stack data seismic data is used for verifying the denoising effect of the self-adaptive method. The two-dimensional actual seismic data is shown in fig. 6, the data has 200 channels, each channel comprises 270 sampling points, the sampling rate is 2ms, the signal-to-noise ratio of the seismic data is low, the continuity of seismic event axes is poor, the random noise energy is strong, faults exist in the middle part and the lower part of the section, and particularly, a plurality of fracture structures exist in the same depth of the lower part.
FIG. 7 is a process for determining the number of event axes of two-dimensional actual seismic data in an adaptive manner. The horizontal axis represents different frequencies, the vertical axis represents the number of corresponding singular values, and the number of singular values determined by the solid line of fig. 7 indicates 8.
FIG. 8 (a) is the result of denoising in an adaptive manner, and FIG. 8 (b) is the removed noise; the used damping factor is D =5, the edges of the seismic event axes are clearly depicted and the noise is completely removed from the result. Although some signal energy loss exists between the bending and breaking coaxial regions, the protection on construction details is good, faults are still well represented, and experiments show that the method has a good effect in denoising two-dimensional actual seismic data.
The above-described embodiments are merely illustrative of the preferred embodiments of the present application, and do not limit the scope of the present application, and various modifications and improvements made to the technical solutions of the present application by those skilled in the art without departing from the spirit of the present application should fall within the protection scope defined by the claims of the present application.
Claims (6)
1. A self-adaptive multi-channel singular spectrum analysis seismic data denoising method is characterized by comprising the following steps:
step 1: presetting time domain seismic data, and converting the time domain seismic data into frequency domain seismic data through DFFT;
step 2: generating a block Hankel matrix based on the frequency domain seismic data;
and step 3: decomposing the Hankel matrix to obtain a singular value matrix, and combining singular values in the singular value matrix into a singular value sequence;
and 4, step 4: based on Akaike information criterion, transforming the singular value sequence to obtain the number of the singular values; calculating and denoising the singular values of the quantity to obtain denoised seismic data;
the method for transforming the singular value sequence based on Akaike information criterion to obtain the number of the singular values comprises the following steps:
performing second-order derivation on a singular value sequence curve to obtain the change rate of the slope of the singular value sequence curve; calculating Akaike information criterion values of the change rates of the slope of the singular value sequence curves of all the frequencies at a preset point based on the Akaike information criterion; calculating to obtain a global minimum value of the Akaike information criterion values based on all the Akaike information criterion values, wherein the global minimum value is the number of singular values of the frequency domain seismic data; wherein all frequencies are the main frequencies of the waves;
the method determines the dominant frequency range of the wavelet, so that the computer can automatically perform denoising treatment;
the method for calculating and denoising the number of singular values comprises the following steps: and inversely transforming the singular value sequence to a frequency domain through the block Hankel matrix, and then inversely transforming to a time domain through Fourier transform to obtain the denoised seismic data.
2. The adaptive multi-channel singular spectral analysis seismic data denoising method of claim 1, wherein the step 2 method comprises:
obtaining frequency slice data of the frequency domain seismic data for each frequency; generating a Hankel matrix based on each row of data of the frequency slice data; and arranging all the Hankel matrixes into one block Hankel matrix.
3. The adaptive multi-channel singular spectral analysis seismic data denoising method of claim 2, wherein the singular value matrix isWhereinIs composed ofThe rank of (c) is determined,is thatOf left singular value vector of (V) x ×L x )×(V x ×L x ) An orthogonal matrix of orders;is thatOf right singular value vector of (H) x ×L y )×(H x ×L y ) An orthogonal matrix of orders;diagonal matrix composed of singular values in descending order, non-zero singular valuesIs equal to the matrixOf rank, w i Is the ith frequency.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111508457.XA CN114200525B (en) | 2021-12-10 | 2021-12-10 | Self-adaptive multi-channel singular spectrum analysis seismic data denoising method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111508457.XA CN114200525B (en) | 2021-12-10 | 2021-12-10 | Self-adaptive multi-channel singular spectrum analysis seismic data denoising method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114200525A CN114200525A (en) | 2022-03-18 |
CN114200525B true CN114200525B (en) | 2022-12-02 |
Family
ID=80652206
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111508457.XA Active CN114200525B (en) | 2021-12-10 | 2021-12-10 | Self-adaptive multi-channel singular spectrum analysis seismic data denoising method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114200525B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115061206A (en) * | 2022-04-15 | 2022-09-16 | 中国地质科学院地球物理地球化学勘查研究所 | Method and system for separating and imaging diffracted waves of remote detection acoustic logging |
CN115657125B (en) * | 2022-11-09 | 2024-03-26 | 河北地质大学 | Method and device for reconstructing and denoising seismic data and electronic equipment |
CN117114500B (en) * | 2023-09-18 | 2024-03-22 | 中孚信息股份有限公司 | Behavior baseline establishment method, system and medium based on data decomposition |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11061159B2 (en) * | 2017-10-20 | 2021-07-13 | Pgs Geophysical As | Electromagnetic response data inversion using singular value decomposition |
CN111881858B (en) * | 2020-07-31 | 2024-02-13 | 中南大学 | Microseismic signal multi-scale denoising method and device and readable storage medium |
-
2021
- 2021-12-10 CN CN202111508457.XA patent/CN114200525B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN114200525A (en) | 2022-03-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114200525B (en) | Self-adaptive multi-channel singular spectrum analysis seismic data denoising method | |
Chen | Fast dictionary learning for noise attenuation of multidimensional seismic data | |
Zhou et al. | Seismic noise attenuation using an online subspace tracking algorithm | |
CN111257929B (en) | Singular value attenuation rank reduction denoising method | |
EP1145046B1 (en) | Method of attenuating noise in three dimensional seismic data using a projection filter | |
CN110096956B (en) | Signal denoising method and device based on EEMD and permutation entropy second-order difference | |
CN102854533A (en) | Wave field separation principle based denoising method for increasing signal to noise ratio of seismic data | |
Cheng* et al. | Application of robust principal component analysis (RPCA) to suppress erratic noise in seismic records | |
Zhou et al. | Sparse dictionary learning for seismic noise attenuation using a fast orthogonal matching pursuit algorithm | |
Zhou et al. | A hybrid method for noise suppression using variational mode decomposition and singular spectrum analysis | |
Oropeza et al. | A randomized SVD for multichannel singular spectrum analysis (MSSA) noise attenuation | |
CN114418886B (en) | Robust denoising method based on depth convolution self-encoder | |
CN103176946A (en) | Sparse decomposition and denoising method facing block sparse signals | |
CN111399057B (en) | Seismic data noise suppression method based on non-convex sparse constraint | |
CN106680876A (en) | Joint denoising method for seismic data | |
CN110261912B (en) | Interpolation and denoising method and system for seismic data | |
Kuruguntla et al. | Erratic noise attenuation using double sparsity dictionary learning method | |
CN112363217A (en) | Random noise suppression method and system for seismic data | |
CN114676734A (en) | Micro-seismic signal noise reduction method based on WPD-EMD-WPD | |
Zhou et al. | Coherent noise attenuation by kurtosis-guided adaptive dictionary learning based on variational sparse representation | |
CN113341463A (en) | Pre-stack seismic data non-stationary blind deconvolution method and related components | |
CN110687605A (en) | Improved K-SVD algorithm-based algorithm analysis application in seismic signal processing | |
CN113009564A (en) | Seismic data processing method and device | |
Sud | Interference cancellation by repeated filtering in the fractional fourier transform domain using mean-square error convergence | |
CN117471529B (en) | Unsteady seismic wavelet self-adaptive extraction method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |