CN105445801B - A kind of processing method for eliminating 2-d seismic data random noise - Google Patents
A kind of processing method for eliminating 2-d seismic data random noise Download PDFInfo
- Publication number
- CN105445801B CN105445801B CN201410441753.6A CN201410441753A CN105445801B CN 105445801 B CN105445801 B CN 105445801B CN 201410441753 A CN201410441753 A CN 201410441753A CN 105445801 B CN105445801 B CN 105445801B
- Authority
- CN
- China
- Prior art keywords
- data
- frequency
- noise
- domain
- processing
- 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
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The present invention provides a kind of processing methods for eliminating 2-d seismic data random noise, belong to oil-gas seismic exploration field.This method includes：A, seismic data is inputted：Input earthquake-capturing single-shot data to be treated or road set information and altogether receiving point data；B, frequency analysis：It chooses representative single-shot or trace gather or receives point data altogether, carry out the frequency range that frequency analysis obtains noise distribution；C, two dimension fitting is three-dimensional：The two-dimension earthquake data of input are fitted to 3D data volume；D, time-domain is transformed into frequency domain：The step c 3D data volumes obtained are transformed by frequency domain from time-domain by Fourier transformation；E, frequency-division section Wiener filtering：In the frequency range of the noise distribution obtained in step b, Wiener filtering processing is carried out to the 3D data volume that step d is obtained；F, frequency domain is transformed into time-domain：To pass through Wiener filtering using fourier reconstructed formula, treated that data transform to time-domain.
Description
Technical field
The invention belongs to oil-gas seismic exploration fields, and in particular to a kind of processing for eliminating 2-d seismic data random noise
Method.
Background technology
Attenuation for random noise, both at home and abroad existing experts and scholars carried out many researchs, have developed large number of rows it
Effective method and technology, such as bandpass filtering method, FXCNS pressings, small echo sub-band forecast.All these methods are in practical application
In all have received good effect, but since method is all based on greatly the exploitation of poststack model, thus be generally possible in poststack application
Obtain satisfied as a result, although some methods can be generalized to prestack, but due to its theoretical foundation or assumptions condition
Limitation, application effect can be restricted significantly.
1st, bandpass filtering method.According to the noise range in data, a bandpass filter is designed, by the frequency beyond noise
It is designed as by and exports, noise components will be filtered at this time.This method is simple, but method is filtering off the same of noise
When, some active ingredients are also had lost into some useful information to filtering.
2nd, FXCNS pressings.FXCNS prediction noises are carried out in the data of given frequency, are missed according to least square
Poor criterion estimates the noise of each frequency respectively.Using frequency and speed two parameters limitation noise ranges, to specified range with
Outer effect of signals is smaller, and energy press-space samples irregular noise, applied widely.Shortcoming is received in cross line direction
Point is very sparse, is unfavorable for random noise prediction and compacting.
3rd, Wavelet transformation processing method.It is that spectrum analysis is carried out to the noise of data in temporal-spatial field first, really
The frequency range of vertical noise；Then separation sound attenuation processing is carried out in selected frequency range, improves data signal-to-noise ratio.Frequency dividing point
From that can identify simultaneously rejecting abnormalities amplitude according to the statistical law of amplitude in the frequency band of restriction, so as to protect useful signal.Cause
There is the difference of radio-frequency component for earthquake useful signal and abnormal sound, the seismic data of different radio-frequency components has different letters
It makes an uproar and compares, to reach high s/n ratio and high-resolution as far as possible, different specific aims is carried out to different radio-frequency components and is handled.
During oil-gas seismic exploration, seismic data does not only exist relevant noise in receive process are gathered, but also deposits
There is random noise.Due to random noise irregular distribution, elimination and attenuation difficulty bigger are more serious on the quality influence of data,
So attenuation random noise, extremely important to improving data signal-to-noise ratio and imaging precision.
The content of the invention
It is an object of the invention to solve above-mentioned problem in the prior art, a kind of elimination 2-d seismic data is provided
The processing method of random noise effectively eliminates the random noise of seismic data, so as to improve the quality of seismic imaging.
The present invention is achieved by the following technical solutions：
A kind of processing method for eliminating 2-d seismic data random noise, the described method includes：
A, seismic data is inputted：Input earthquake-capturing single-shot data to be treated, road set information or receiving point data altogether；
B, frequency analysis：It chooses representative single-shot or trace gather or receives point data altogether, carry out frequency analysis acquisition
The frequency range of noise distribution；
C, two dimension fitting is three-dimensional：The two-dimension earthquake data of input are fitted to 3D data volume；
D, time-domain is transformed into frequency domain：The 3D data volume that step c is obtained is turned from time-domain by Fourier transformation
Change to frequency domain；
E, frequency-division section Wiener filtering：In the frequency range of the noise distribution obtained in step b, the three-dimensional that is obtained to step d
Data volume carries out Wiener filtering processing；
F, frequency domain is transformed into time-domain：The conversion of Wiener filtering treated data will be passed through using fourier reconstructed formula
To time-domain；
G, three-dimensional is transformed into two dimension：The obtained data of step f are converted into 2-D data；
H, data export to form new data：Seismic data output after noise is separated.
In the step b, spectrum analysis is carried out to whole seismic data on the premise of noise is avoided as far as possible, by frequency spectrum
18db at cut-off frequency be set to the population frequency scope of seismic data；It chooses noise components and carries out spectrum analysis, by the 18db of frequency spectrum
Place's cut-off frequency is set to the frequency content of noise；
If seismic data is analyzed different types of noise, respectively really respectively there are polytype noise
The frequency content of fixed different noises is handled so as to the specific aim in later stage.
The step c is specific as follows：
For two-dimentional single-shot data, processing is as follows：It modifies, i.e., is designed as two-dimentional single-shot number with ascending order to road header value
Big gun road number number is designed as Y-coordinate sequence number by coordinate X sequence numbers with ascending order, and time t is designed as time coordinate Z, is so formed three-dimensional
The big gun data of mode；
For road set information, processing is as follows：Modify to the CMP header words of each trace gather, will every line CMP sequences
Number x coordinate axis sequence number is used as, it, will time conduct by using the single track sequence number of the CMP trace gathers of the big minispread of offset distance as y-axis sequence number
Z-axis so forms 3D data volume；
For common receiving point data, processing is as follows：It modifies, will each connect altogether to the header word for often having receiving point altogether
The point sequence number of sink, will be using the road sequence number in the common receiving point road of the big minispread of offset distance as Y-axis sequence as X-coordinate axle sequence number
Number, using time t as Z axis, so form 3D data volume.
The step g is specific as follows：
For two-dimentional single-shot data, processing is as follows：Road header value is changed into back initial value, that is, changes back to single-shot number and big gun road number
Number；
For road set information, processing is as follows：Change back to CMP sequence numbers (i.e. trace gather sequence number) and single track sequence number；
For common receiving point data, processing is as follows：Change back to the point sequence number of common receiving point and the road sequence number in common receiving point road.
Compared with prior art, the beneficial effects of the invention are as follows：By the fitting of the seismic data of two-dimensional approach it is three-dimensional after again into
Row processing, signal-to-noise ratio improvement effect are substantially reliable.Random noise attenuation is effective, and imaging precision improves.Two dimension can be effectively eliminated
The random noise of seismic data, original active ingredient are not destroyed.Low frequency noise has obtained preferably eliminating, after processing
Data noise it is higher, beneficial to the imaging in later stage.The imaging of overlap-add procedure section is clear, and construction is complete, and stratum reflection is strong
Weak relation and wave group feature substantially protrude；The section of migration processing, signal-to-noise ratio is high, and structure imaging understands that tomography breakpoint clearly may be used
It leans on, is easy to the geological research in later stage.
Description of the drawings
Fig. 1 two dimension fitting schematic three dimensional views
Fig. 2 is the step block diagram of this method
Fig. 3 a are the former data of a two dimensional model
Fig. 3 b are the data added in after random noise
Fig. 3 c are using two-dimensional approach treated data
Fig. 3 d are using this method treated data
Fig. 4 a initial data
Fig. 4 b decayed to Fig. 4 a using this method after data
Firsthand information in Fig. 5 a embodiments 1
Data after noise treatment in Fig. 5 b embodiments 1
The noise separated in Fig. 5 c embodiments 1
Conventional noise treatment section in Fig. 6 a embodiments 1
This method noise treatment section in Fig. 6 b embodiments 1
Initial data in Fig. 7 a embodiments 2
Data in Fig. 7 b embodiments 2 after noise treatment
The noise isolated in Fig. 7 c embodiments 2
The stacked section of conventional random noise processing in Fig. 8 a embodiments 2
The stacked section that this method random noise is handled in Fig. 8 b embodiments 2
Using the migrated section of conventional noise treatment in Fig. 9 a embodiments 2
The migrated section of this method noise treatment is used in Fig. 9 b embodiments 2.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings：
It is an object of the present invention to provide a kind of processing methods for effectively eliminating 2-d seismic data random noise.This method be by
2-D data is fitted in three-dimensional domain, then the filtering Processing for removing into Row noise, the three-dimensional of useful signal so in three-dimensional domain
Correlation properties are substantially beneficial to distinguish with noise, while carry out Wiener filtering processing using fourier-transform to frequency domain, are more advantageous to
Effect well has been seen in the separation of noise, real data processing.
As shown in Fig. 2, the present invention specifically includes following steps：
A, seismic data is inputted
Input earthquake-capturing single-shot data to be treated, road set information or receiving point data altogether.
B, frequency analysis
Input data is analyzed, the representative single-shot such as energy, frequency, noise is chosen at and (works as input data
Representative trace gather is chosen for when being set information；It is chosen when being common receiving point data and should be representative common reception
Point data), carry out frequency analysis.Spectrum analysis is carried out to overall data on the premise of noise is avoided as far as possible, by frequency spectrum
Cut-off frequency is set to the population frequency scope of data at 18db.Noise components progress spectrum analysis is chosen by cut-off frequency at the 18db of frequency spectrum to determine
For the frequency content of noise, such as data there are polytype noise, it is necessary to analyzed respectively different types of noise, point
Not Que Ding different noises frequency content, so as to the later stage specific aim handle.
C, two dimension fitting is three-dimensional
It modifies to two-dimentional single-shot data track header value, i.e., two-dimentional single-shot number is designed as coordinate X sequence numbers, big gun road with ascending order
Several numbers are designed as Y-coordinate sequence number with ascending order, and time t is designed as time coordinate Z.Thus the big gun data of two-dimensional approach are formd
The big gun data of one three dimensional constitution.
If in being handled in trace gather CMP, amending method is similar with single-shot；By the CMP header words of each trace gather
Modification, the CMP sequence numbers of every line are as x coordinate axis sequence number, using the single track sequence number of the CMP trace gathers of the big minispread of offset distance as y
Axis sequence number, time as z-axis, that is, constitute " three-dimensional " data volume.Three coordinate schematic diagrames after fitting are shown in Fig. 1, this is three-dimensional
Body is combined by multiple two-dimentional bodies, and the ellipsis in Fig. 1 represents two-dimentional ' single-shot ' either ' trace gather ' or ' receiving point altogether ' number
According to.
D, time-domain is transformed into frequency domain
The 3D data volume of processing is transformed by frequency domain from time-domain by Fourier transformation.
Fourier transform formula：
All data are handled using the formula, it becomes possible to data be made to be transformed into frequency domain from time-domain.
E, frequency-division section Wiener filtering
In front on the basis of b frequency analyses, it has been found that the frequency range of noise distribution, carries out in the frequency range
Wiener filtering processing (if noise in 2 or 2 frequencies above sections, it is necessary to carry out similar filtering in different frequency ranges
Processing).
Wiener filtering (being directly filtered to three-dimensional data) is special using the correlation properties and frequency spectrum of stationary random process
Property to mixing the processing method that is filtered of noisy signal, the filtering side for extracting useful signal can predict from blended data
Method.Wiener filtering principle is the mixture w (t) (step d obtain be exactly the mixture) for signal s (t) He noise n (t)
(w (t)=s (t)+n (t)) designs an optimum linear filtering factor, using filtering, smooth and prediction computational methods, according to
Minimum mean-squared error criterion, sub-argument goes out signal s (t) from w (t).Here it is to be separated noise using this method.
F, frequency domain is transformed into time-domain
Data after conversion are transformed to by time-domain using Fu Shi reconstructed formulas.
Fourier reconstructed formula：
G, three-dimensional is transformed into two dimension
It is specific as follows：When processing data are single-shots, change back to single-shot number and receive big gun Taoist monastic name；It is changeed back in this way during road set information
To trace gather sequence number and single track sequence number；Receiving point data is total in this way changes back to common reception period and altogether receiving point Taoist monastic name.
H, data export to form new data
By above serial calculating processing, the seismic data after noise is separated exports, in case the later stage is further applied.
Fig. 3 a- Fig. 3 d are model noise analysis design sketch, and Fig. 3 a are the former data of a two dimensional model, Fig. 3 b be add in
Data after machine noise, Fig. 3 c are using two-dimensional approach treated data, and Fig. 3 d are using this method treated data.Point
Analysis is learnt：1. after initial data is added in random noise, model substantially be can't see, as shown in Figure 3b；2. using two dimension
The data that normality mode is handled are not very clear, such as Fig. 3 c；3. using this method treated data shape substantially close to original
Data as shown in Figure 3d, illustrate that carrying out noise treatment effect using three dimensional constitution is better than two-dimensional approach.
Fig. 4 a, Fig. 4 b are analysis Contrast on effect, it can be seen that effective wave energy is strengthened after processing, noise
It is effectively eliminated.
Example 1：Oily one seismic acquisition single-shot data treatment effect of area.Fig. 5 a- Fig. 5 c are that oily one seismic prospecting of area is adopted
Collect single-shot data treatment effect figure, Fig. 5 a are firsthand information, Fig. 5 b are data after noise treatment, Fig. 5 c are making an uproar of separating
Sound.It is not difficult to find out：The noise of low frequency has obtained preferably eliminating, and treated, and data noise is higher, beneficial to the imaging in later stage
Processing；Fig. 6 a and Fig. 6 b are overlap-add procedure section design sketch, and Fig. 6 a are conventional noise treatment section, Fig. 6 b are after this method is handled
Section.It can be seen that using the section of this method treated single-shot data is overlapped processing, low frequency noise has obtained preferably disappearing
It removes, imaging is clear, and construction is complete, is easy to the geologic interpretation research in later stage.
Example 2：Oily two seismic acquisition single-shot data treatment effect of area.
Fig. 7 a to Fig. 7 c are oily two seismic acquisition single-shot data treatment effect figures of area, and Fig. 7 a are firsthand information, Fig. 7 b
It is the noise separated for data, Fig. 7 c after noise treatment.It is not difficult to find out：The noise of low frequency has obtained preferably eliminating, place
Data noise after reason is higher, beneficial to the imaging in later stage；Fig. 8 a, Fig. 8 b are the section design sketch of overlap-add procedure, Fig. 8 a
For section after using conventional noise treatment section, Fig. 8 b as this method processing.It can be seen that using this method treated single-shot data
The section of processing is overlapped, signal-to-noise ratio is apparently higher than routine, and structure imaging understands, cophase wave group continuity is good, after being easy to
The processing of phase.Fig. 9 a, Fig. 9 b are the section design sketch of migration processing, and Fig. 9 a are migrated section, the figure using conventional noise treatment
9b is the migrated section of this method processing.It can be seen that the section handled using this method, signal-to-noise ratio is high, and structure imaging understands, tomography
Breakpoint is clearly reliable, is easy to the geological research in later stage.
Above-mentioned technical proposal is one embodiment of the present invention, for those skilled in the art, at this
On the basis of disclosure of the invention application process and principle, it is easy to make various types of improvement or deformation, be not limited solely to this
Invent the described method of above-mentioned specific embodiment, therefore previously described mode is simply preferred, and and without limitation
The meaning of property.
Claims (4)
1. a kind of processing method for eliminating 2-d seismic data random noise, it is characterised in that：The described method includes：
A, seismic data is inputted：Input earthquake-capturing single-shot data to be treated, road set information or receiving point data altogether；
B, frequency analysis：It chooses representative single-shot or trace gather or receives point data altogether, carry out frequency analysis and obtain noise
The frequency range of distribution；
C, two dimension fitting is three-dimensional：The two-dimension earthquake data of input are fitted to 3D data volume；
D, time-domain is transformed into frequency domain：The step c 3D data volumes obtained are transformed into from time-domain by Fourier transformation
Frequency domain；
E, frequency-division section Wiener filtering：In the frequency range of the noise distribution obtained in step b, the three-dimensional data that is obtained to step d
Body carries out Wiener filtering processing；
F, frequency domain is transformed into time-domain：Using fourier reconstructed formula will pass through Wiener filtering treated data transform to when
Between domain；
G, three-dimensional is transformed into two dimension：The obtained data of step f are converted into 2-D data；
H, data export to form new data：Seismic data output after noise is separated.
2. the processing method according to claim 1 for eliminating 2-d seismic data random noise, it is characterised in that：The step
In rapid b, spectrum analysis is carried out to whole seismic data on the premise of noise is avoided as far as possible, is determined by cut-off frequency at the 18db of frequency spectrum
For the population frequency scope of seismic data；It chooses noise components and carries out spectrum analysis, be set to noise by cut-off frequency at the 18db of frequency spectrum
Frequency content；
If seismic data is analyzed different types of noise, is determined respectively or not there are polytype noise
With the frequency content of noise, handled so as to the specific aim in later stage.
3. the processing method according to claim 1 for eliminating 2-d seismic data random noise, it is characterised in that：The step
Rapid c is specific as follows：
For two-dimentional single-shot data, processing is as follows：It modifies to road header value, i.e., two-dimentional single-shot number is designed as coordinate with ascending order
Big gun road number number is designed as Y-coordinate sequence number by X sequence numbers with ascending order, and time t is designed as time coordinate Z, so forms three dimensional constitution
Big gun data；
For road set information, processing is as follows：Modify to the CMP header words of each trace gather, will the CMP sequence numbers of every line make
For X-coordinate axle sequence number, by using the single track sequence number of the CMP trace gathers of the big minispread of offset distance as Y-axis sequence number, will the time as Z axis,
So form 3D data volume；
For common receiving point data, processing is as follows：It modifies to the header word for often having receiving point altogether, will each be total to receiving point
Point sequence number as X-coordinate axle sequence number, will by using the road sequence number in the common receiving point road of the big minispread of offset distance as Y-axis sequence number
Time t so forms 3D data volume as Z axis.
4. the processing method according to claim 1 for eliminating 2-d seismic data random noise, it is characterised in that：The step
Rapid g is specific as follows：
For two-dimentional single-shot data, processing is as follows：Road header value is changed into back initial value, that is, changes back to single-shot number and big gun road number number；
For road set information, processing is as follows：Change back to CMP sequence numbers and single track sequence number；
For common receiving point data, processing is as follows：Change back to the point sequence number of common receiving point and the road sequence number in common receiving point road.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410441753.6A CN105445801B (en) | 2014-09-01 | 2014-09-01 | A kind of processing method for eliminating 2-d seismic data random noise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410441753.6A CN105445801B (en) | 2014-09-01 | 2014-09-01 | A kind of processing method for eliminating 2-d seismic data random noise |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105445801A CN105445801A (en) | 2016-03-30 |
| CN105445801B true CN105445801B (en) | 2018-06-01 |
Family
ID=55556213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410441753.6A Active CN105445801B (en) | 2014-09-01 | 2014-09-01 | A kind of processing method for eliminating 2-d seismic data random noise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105445801B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105954799B (en) * | 2016-04-27 | 2018-04-03 | 中国石油天然气股份有限公司 | Time-frequency domain seismic data processing method based on weighted stacking |
| CN106772610B (en) * | 2016-11-14 | 2020-05-01 | 中国石油化工股份有限公司 | Prestack gather data dimension-increasing attribute analysis method |
| CN106896409B (en) * | 2017-03-14 | 2018-12-07 | 中国海洋石油集团有限公司 | A kind of varying depth cable ghost reflection drawing method based on wave equation boundary values inverting |
| CN107133589A (en) * | 2017-05-04 | 2017-09-05 | 临沂大学 | Denoising algorithm based on Wiener filter |
| CN107728200B (en) * | 2017-09-29 | 2019-03-29 | 中国石油化工股份有限公司 | Ground micro-seismic fracturing fracture dynamic spread method of real-time |
| CN113589383B (en) * | 2020-04-30 | 2024-03-19 | 中国石油化工股份有限公司 | Seismic data linear interference noise elimination method based on deep learning |
| CN112255683A (en) * | 2020-10-26 | 2021-01-22 | 中国石油天然气集团有限公司 | Noise suppression method and device for seismic data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101598809A (en) * | 2008-06-04 | 2009-12-09 | 中国石油天然气集团公司 | A kind of self-adaptation is eliminated the method for linear programming noise and multiple reflection interference |
| CN103852786A (en) * | 2014-02-13 | 2014-06-11 | 中国石油天然气股份有限公司 | Reverse time migration imaging method and system applied to land seismic data |
| CN103995292A (en) * | 2014-06-09 | 2014-08-20 | 桂林电子科技大学 | Transient electromagnetic early signal reconstruction method |
-
2014
- 2014-09-01 CN CN201410441753.6A patent/CN105445801B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101598809A (en) * | 2008-06-04 | 2009-12-09 | 中国石油天然气集团公司 | A kind of self-adaptation is eliminated the method for linear programming noise and multiple reflection interference |
| CN103852786A (en) * | 2014-02-13 | 2014-06-11 | 中国石油天然气股份有限公司 | Reverse time migration imaging method and system applied to land seismic data |
| CN103995292A (en) * | 2014-06-09 | 2014-08-20 | 桂林电子科技大学 | Transient electromagnetic early signal reconstruction method |
Non-Patent Citations (2)
| Title |
|---|
| 基于核函数主分量的维纳滤波方法研究;李月等;《地球物理学报》;20100531;第53卷(第5期);1226-1233 * |
| 维纳滤波在地震资料噪声消除中的应用;高岩等;《石油仪器》;20110630;16-17 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105445801A (en) | 2016-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105445801B (en) | A kind of processing method for eliminating 2-d seismic data random noise | |
| Chen et al. | Empirical low-rank approximation for seismic noise attenuation | |
| CN107144880B (en) | A kind of seismic wave wave field separation method | |
| CN107144879B (en) | A kind of seismic wave noise-reduction method based on adaptive-filtering in conjunction with wavelet transformation | |
| Chen et al. | Random noise reduction using a hybrid method based on ensemble empirical mode decomposition | |
| CN102043165B (en) | Surface Wave Separation and Suppression Method Based on Basis Pursuit Algorithm | |
| CN108267784A (en) | A kind of seismic signal random noise compression process method | |
| CN103364832A (en) | Seismic attenuation qualitative estimation method based on self-adaptive optimal kernel time frequency distribution | |
| CN104297791A (en) | Inversion method and system based on seismic dominant frequency | |
| CN106646637A (en) | Method for removing peak noise in nuclear magnetism signal | |
| CN104849756A (en) | Method for improving resolution ratio of seismic data and enhancing energy of valid weak signals | |
| CN106707342B (en) | Common-shot-gather multistage surface wave pressing method and device | |
| CN105652322A (en) | T-f-k field polarization filtering method for multi-component seismic data | |
| Liu et al. | Irregularly sampled seismic data reconstruction using multiscale multidirectional adaptive prediction-error filter | |
| CN103576198A (en) | Method for rapidly predicting two-dimensional offshore earthquake data free surface multiple | |
| CN106199698A (en) | Frequency domain geological data reconstructing method based on many subwaves information | |
| Wu et al. | Fast principal component analysis for stacking seismic data | |
| CN105259579A (en) | A high-amplitude shielding layer rejecting method based on seismic data instantaneous attributes | |
| CN110554428A (en) | Seismic wave low-frequency energy change rate extraction method based on variational modal decomposition | |
| CN104849590A (en) | Method for detecting weak pulse signals under mixed noise interference | |
| CN104635264B (en) | Pre-stack seismic data processing method and device | |
| CN106950600B (en) | A kind of minimizing technology of near surface scattering surface wave | |
| CN103135133A (en) | Vector noise reduction method and device for multi-component seismic data | |
| Zhou et al. | Unsupervised machine learning for waveform extraction in microseismic denoising | |
| CN103645504A (en) | Weak earthquake signal processing method based on generalized instantaneous phase and P norm negative norm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |