CN112965113A - Seismic data signal-to-noise ratio improving method - Google Patents
Seismic data signal-to-noise ratio improving method Download PDFInfo
- Publication number
- CN112965113A CN112965113A CN202110138136.9A CN202110138136A CN112965113A CN 112965113 A CN112965113 A CN 112965113A CN 202110138136 A CN202110138136 A CN 202110138136A CN 112965113 A CN112965113 A CN 112965113A
- Authority
- CN
- China
- Prior art keywords
- data
- seismic
- seismic data
- signal
- denoising
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 91
- 239000013307 optical fiber Substances 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 42
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 230000006872 improvement Effects 0.000 claims abstract description 4
- 230000003287 optical effect Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 238000012935 Averaging Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 7
- 230000008901 benefit Effects 0.000 abstract description 5
- 230000009467 reduction Effects 0.000 abstract description 4
- 238000009825 accumulation Methods 0.000 abstract 1
- 238000004140 cleaning Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 30
- 238000005516 engineering process Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009529 body temperature measurement Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 239000011800 void material Substances 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013507 mapping 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
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/30—Noise handling
- G01V2210/32—Noise reduction
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method for improving the signal-to-noise ratio of seismic data, which comprises the following steps: s1: collecting multiple data records of seismic data with noise for the same seismic signal by using a distributed optical fiber sensing seismograph; s2: training the neural network by using a noise2noise denoising method or a noise2void denoising method; s3: and denoising the plurality of data records by using the trained neural network to complete the improvement of the signal-to-noise ratio of the seismic data. The invention utilizes the advantage that the multi-channel, multi-wavelength, multi-frequency or multi-code optical fiber sensing system can record the same seismic signal for multiple times, combines the noise2noise and noise2void denoising methods without cleaning the signal, reduces the requirement on a data sample when the signal is denoised, and can be combined with the traditional accumulation average denoising method to realize the high-efficiency noise reduction of the seismic data.
Description
Technical Field
The invention belongs to the technical field of seismic data denoising, and particularly relates to a seismic data signal-to-noise ratio improving method.
Background
Due to the characteristics of the distributed acoustic sensing technology and the characteristics that the optical fiber is very sensitive and is easily interfered, the seismic data acquired by the technology is often accompanied by more noise, so that the denoising link of the seismic data is a problem faced by the technology and is an important bottleneck for further development of the technology.
Some problems exist with the traditional noise reduction of seismic data: the seismic data are huge, the algorithm has more iteration times and higher complexity; or the local analysis difference is obtained, and the effective signal is distorted after denoising, so that the original amplitude characteristic of the seismic signal is damaged, and the signal fidelity is reduced. However, due to the characteristic problem of the DAS technology, the clean data is very difficult to acquire, and thus, the implementation and the use of the method are limited.
Disclosure of Invention
The invention aims to solve the problems of poor universality and complex process of the existing denoising method, and provides a seismic data signal-to-noise ratio improving method.
The technical scheme of the invention is as follows: a seismic data signal-to-noise ratio improving method comprises the following steps:
s1: collecting multiple data records of seismic data with noise for the same seismic signal by using a distributed optical fiber sensing seismograph;
s2: training the neural network by using a noise2noise denoising method or a noise2void denoising method;
s3: and denoising the plurality of data records by using the trained neural network to complete the improvement of the signal-to-noise ratio of the seismic data.
The invention has the beneficial effects that:
(1) the invention utilizes the advantage that the multi-channel, multi-wavelength, multi-frequency or multi-code optical fiber sensing system can record the same seismic signal for multiple times, combines the advantages that the noise removal methods of noise2noise and noise2void do not need clean signals and the amount of required sample data is small, reduces the requirement on the data sample during signal noise removal, and can be combined with the traditional accumulative average noise removal method to realize the high-efficiency noise reduction of the seismic data. The method is not only suitable for denoising seismic data, but also suitable for denoising other data of the same type, and can be popularized to a distributed optical fiber temperature measurement and strain sensing system to improve the signal-to-noise ratio of sensing data.
(2) On the basis of the traditional data denoising method based on the neural network, the method changes the acquisition mode of seismic data, the data acquisition of a network training set and the selection of a loss function, improves the defect that the prior denoising method based on the neural network needs to acquire clean data in advance as a training target, simplifies the denoising steps, further enhances the universality and enables the denoising process of the seismic data to be more accurate, efficient and intelligent. The noise removing method using noise2noise realizes the noise removing work of the collected noisy signals of the determined target under different noise environments. The noise removal of the seismic data is carried out by using noise2void, and the noise removal work of signals changing at any time under different noise environments is realized.
Further, in step S1, the method of acquiring seismic data with noise includes: laying an optical cable along the optical fiber logging, and receiving seismic signals by using the optical cable; the distributed optical fiber sensing seismograph is adopted to acquire seismic data of seismic signals, and a group of seismic data pairs with noise are acquired and obtained for the same seismic signal.
Further, in step S1, the method for acquiring seismic data for each seismic signal includes: the method comprises the following steps that a double-channel optical fiber sensing seismograph is used for simultaneously collecting two core optical fibers in an optical cable; or a dual-wavelength, dual-frequency or dual-code distributed optical fiber sensing seismograph is used for injecting light with two wavelengths, two frequencies and two codes into the same optical fiber, and each wavelength is collected to obtain a pair of seismic data with noise; or a double-coding distributed optical fiber sensing seismograph is used, two coded lights are injected into the same optical fiber, each wavelength is collected, and a pair of seismic data with noise is obtained.
Further, in step S2, if the seismic data pairs corresponding to the seismic signals are both noisy, the method for training the neural network includes: using one of the seismic data (x)i,yi) And training the objective function, and performing average optimization on the output result.
Further, in step S3, a pair of seismic data acquired for the same seismic signal, one of which is used as input data and the other is used as target data, is denoised by using a noise2noise denoising method, or two seismic data are denoised by using a noise2void denoising method respectively, and the two processed data are averaged.
Further, in step S3, the method for acquiring seismic data for each seismic signal includes: the method comprises the steps that a multi-channel, multi-wavelength, multi-frequency or multi-code distributed optical fiber sensing seismograph is used for simultaneously collecting n seismic data for the same seismic signal, the collected n seismic data are divided into n/2 seismic data pairs, a noise2noise denoising method is used for denoising each seismic data pair, and denoised data are averaged;
for the condition of more channels, dividing n/2 data pairs processed by a noise2noise denoising method into n/4 data pairs, and carrying out n2n denoising processing and averaging.
For the n collected seismic data, denoising by using a noise2void denoising method respectively, and then averaging;
and for the seismic signals excited in the same position in the same mode, if the signal and the noise characteristics are basically the same, exciting the seismic signals at one position n times, carrying out data acquisition on the seismic signals excited at each time, carrying out denoising, and averaging the n denoised seismic data obtained after processing to obtain the final denoised seismic signals.
Further, in step S2, if image denoising is implemented by using a noise2void denoising method, the method for training the neural network is as follows: selecting N-dimensional-N-dimensional data blocks from the seismic data pair as input of a training neural network, replacing central data with random data in the central data field to obtain target data, and performing training optimization on a target function by using the target data to complete seismic data denoising.
Drawings
FIG. 1 is a flow chart of a seismic data denoising method.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings.
Before describing specific embodiments of the present invention, in order to make the solution of the present invention more clear and complete, the definitions of the abbreviations and key terms appearing in the present invention will be explained first:
noise2 noise: the abbreviation is noise to noise, which refers to training a neural network in the case where both the input and output are noisy signals, called noise2noise for short. A paper from ICML2018 published jointly by researchers from great india, alto university and MIT.
noise2 void: the abbreviation refers to noise to void, which refers to training a neural network in the case of only one noisy signal, called noise2void for short.
As shown in fig. 1, the present invention provides a method for improving signal-to-noise ratio of seismic data, comprising the following steps:
s1: collecting multiple data records of seismic data with noise for the same seismic signal by using a distributed optical fiber sensing seismograph;
s2: training the neural network by using a noise2noise denoising method or a noise2void denoising method;
s3: and denoising the plurality of data records by using the trained neural network to complete the improvement of the signal-to-noise ratio of the seismic data.
In the embodiment of the present invention, as shown in fig. 1, in step S1, the method for acquiring seismic data with noise is as follows: laying an optical cable along the optical fiber logging, and receiving seismic signals by using the optical cable; the distributed optical fiber sensing seismograph is adopted to acquire seismic data of seismic signals, and a group of seismic data pairs with noise are acquired and obtained for the same seismic signal.
In the embodiment of the present invention, as shown in fig. 1, in step S1, the method for acquiring seismic data for each seismic signal includes: the method comprises the following steps that a double-channel optical fiber sensing seismograph is used for simultaneously collecting two core optical fibers in an optical cable; or a dual-wavelength, dual-frequency or dual-code distributed optical fiber sensing seismograph is used for injecting light with two wavelengths, two frequencies and two codes into the same optical fiber, and each wavelength is collected to obtain a pair of seismic data with noise; or a double-coding distributed optical fiber sensing seismograph is used, two coded lights are injected into the same optical fiber, each wavelength is collected, and a pair of seismic data with noise is obtained.
In the embodiment of the present invention, as shown in fig. 1, in step S2, if the seismic data pairs corresponding to the seismic signals all have noise, the method for training the neural network includes: using one of the seismic data (x)i,yi) And training the objective function, and performing average optimization on the output result.
In the embodiment of the present invention, as shown in fig. 1, in step S3, a pair of seismic data acquired from the same seismic signal is used, one of the pair of seismic data is used as input data and the other is used as target data, and a noise removal method is used to perform noise removal processing, or a noise removal method 2void is used to perform noise removal processing on the two seismic data, and the two processed data are averaged.
In the embodiment of the present invention, as shown in fig. 1, in step S3, the method for acquiring seismic data for each seismic signal includes: the method comprises the steps that a multi-channel, multi-wavelength, multi-frequency or multi-code distributed optical fiber sensing seismograph is used for simultaneously collecting n seismic data for the same seismic signal, the collected n seismic data are divided into n/2 seismic data pairs, a noise2noise denoising method is used for denoising each seismic data pair, and denoised data are averaged;
for the condition of more channels, dividing n/2 data pairs processed by a noise2noise denoising method into n/4 data pairs, and carrying out n2n denoising processing and averaging.
For the n collected seismic data, denoising by using a noise2void denoising method respectively, and then averaging;
and for the seismic signals excited in the same position in the same mode, if the signal and the noise characteristics are basically the same, exciting the seismic signals at one position n times, carrying out data acquisition on the seismic signals excited at each time, carrying out denoising, and averaging the n denoised seismic data obtained after processing to obtain the final denoised seismic signals.
In the embodiment of the present invention, as shown in fig. 1, in step S3, if the seismic data pairs are both noisy and the optimal solution of the loss function is at the arithmetic mean, the training output result of the neural network is optimized averagely.
In the embodiment of the present invention, as shown in fig. 1, in step S2, image denoising is implemented by using a noise2void denoising method, and the method for training the neural network is as follows: selecting N-dimensional-N-dimensional data blocks from the seismic data pair as input of a training neural network, replacing central data with random data in the central data field to obtain target data, and performing training optimization on a target function by using the target data to complete seismic data denoising.
In the invention, an optical cable is arranged along the optical fiber logging, a seismic source generates seismic waves in a heavy hammer knocking mode, an excitation source is used for excitation, the optical cable senses seismic signals to generate optical signals, and data acquisition is carried out through a distributed optical fiber sensing seismograph. For each seismic signal, the distributed optical fiber sensing seismograph can acquire N channels of data at one time, directly acquire two-dimensional DAS seismic data and directly use the DAS seismic data without one-dimensional to two-dimensional processing. And the characteristics of multiple channels and multiple wavelengths of the distributed optical fiber sensing seismograph can directly acquire the pair of noisy seismic data. According to the characteristic, the noise data is not processed to obtain clean seismic data to be used as target data of deep learning, and an image denoising method noise2noise without the clean data is applied to denoising of the seismic data.
In the invention, the training of the denoising network can be realized without obtaining a clean data set as a target data set through a complex data processing process. The data set acquired in the seismic data acquisition step is divided into two parts, namely an output signal training set and a target signal training set.
The characteristics of the seismic signals excited for multiple times at the same position are basically the same, so that the seismic signals can be denoised after being excited for one time, then the seismic signals are excited for multiple times at the position, and then the average processing is carried out after the denoising is carried out respectively, so that the denoised seismic signals are obtained.
In the embodiment of the present invention, as shown in fig. 1, in step S3, if the seismic data pairs corresponding to the seismic signals all have noise, and the noise removal method of the image is implemented by using a noise2noise removal method, the method for training the neural network is as follows: using one of the seismic data (x)i,yi) Training the target function, and performing average optimization on an output result; the expression of the objective function is:
wherein argmin {. cndot } represents the value operation of the corresponding variable when the objective function reaches the minimum value,
the mathematical expectation of x is expressed as,
expressing conditional expectation for y, L (-) expressing a loss function, fθ(x) Representing the network function, theta representing the network parameter, x representing the input data, and y representing the output data.
In the embodiment of the present invention, as shown in fig. 1, in step S3, if the seismic data pairs are all noisy and the optimal solution of the loss function is at the arithmetic mean, the training output result of the neural network is optimized averagely, and the calculation formula is:
where z represents the noise-free data, y represents the output data,
indicating the condition expectation of y.
In the present invention, for a noisy signal, which corresponds to a noise-free signal, there is not only one possibility, but also the output of the neural network is different due to the difference used to train the loss function, and if the optimal solution of the loss function is at the arithmetic mean, the result of the neural network output is the arithmetic mean of all possible output results. In this case, since the value of z is constant as long as the mean value of y is constant, y expected to be 0 can be used as a noise2Instead of y, the output does not change.
The output data is ideally clean data as a result of the training of the network given an infinite number of data in the training set, and the expected value of the output data is also clean data in a finite number of training data. As the amount of training data increases, the output results will also be closer to noise-free data.
The above is the case when the optimal solution of the loss function is obtained at the mean value, and the specific choice of the loss function depends on the nature of the noise to be eliminated. The average value is zero when the general Gaussian white noise exists, and the loss function optimal quality has better denoising effect if the loss function optimal quality is obtained at the average value.
In the embodiment of the present invention, as shown in fig. 1, in step S3, when the seismic signal corresponds to a single seismic data, the expression of the objective function is:
wherein argmin {. cndot.) represents the value calculation of the corresponding variable when the objective function reaches the minimum value, L (·) represents the loss function, f (·) represents the network function, θ represents the network parameter, x represents the input data, y represents the output data, i represents the first summation value, j represents the second summation value, i and j represent 1, 2, 3 … …,
which represents the input data, is,
representing the target data.
In the invention, if data with the size of N x N is taken as input from two-dimensional data, the network is trained by taking the central data of the data block as a target, and the network learns to directly output the central value of the input data block. Therefore, in the use of the method, the central data of each input data block is replaced by random data in the neighborhood, and the network learning to the identity mapping is effectively avoided. The noise signal is a random signal and has no correlation, and under the condition that the seismic signal has the correlation, the network cannot recover the noise signal at the central position from the noise in the neighborhood, but can recover the seismic data signal at the central position to a certain extent from the correlation of the seismic signal in the neighborhood. The input and target data may be derived from individual noisy seismic data. And extracting N x N data blocks from the two-dimensional seismic data, and replacing the data in the center of the data blocks by one random data at the periphery to be used as input data.
In the embodiment of the invention, when the seismic signals are detected on site, the acquired seismic data are all noisy seismic data, and in order to reduce the noise of the noisy data and effectively suppress the noise, a proper loss function is selected for different types of noise, and the loss function is optimized.
The gaussian noise belongs to common noise, and for the gaussian noise, since the average value of common additive white gaussian noise is 0, a loss function is selected: l (z, y) ═ z-y)2。
When the function is used as a loss function to train the neural network, the optimal solution of the function is obtained at the arithmetic mean value of y, for the neural network trained by the loss function, the network learns the arithmetic mean value of all possible results, and the mean value of Gaussian white noise is 0, so the noise of the output result of the neural network is 0 theoretically, and clean seismic data are obtained. The output data is ideally clean as a result of the network training when there is an infinite number of data in a given training set, and the output data is expected to be clean when there is a finite number of training data. As the amount of training data increases, the error of the output result is also smaller.
The working principle and the process of the invention are as follows: firstly, collecting a group of seismic data pairs with noise by using a distributed optical fiber sensing seismograph; repeatedly acquiring, acquiring multiple sets of seismic data pairs with noise, and using the seismic data pairs as a data set for training a neural network; and training the neural network by utilizing multiple sets of seismic data pairs with noise to finish seismic data denoising. Meanwhile, the seismic data signal-to-noise ratio improving method is not limited to a two-dimensional form.
The invention has the beneficial effects that:
(1) the invention utilizes the advantage that the multi-channel, multi-wavelength, multi-frequency or multi-code optical fiber sensing system can record the same seismic signal for multiple times, combines the advantages that the noise removal methods of noise2noise and noise2void do not need clean signals and the amount of required sample data is small, reduces the requirement on the data sample during signal noise removal, and can be combined with the traditional accumulative average noise removal method to realize the high-efficiency noise reduction of the seismic data. The method is not only suitable for denoising seismic data, but also suitable for denoising other data of the same type, and can be popularized to a distributed optical fiber temperature measurement and strain sensing system to improve the signal-to-noise ratio of sensing data.
(2) On the basis of the traditional data denoising method based on the neural network, the method changes the acquisition mode of seismic data, the data acquisition of a network training set and the selection of a loss function, improves the defect that the prior denoising method based on the neural network needs to acquire clean data in advance as a training target, simplifies the denoising steps, further enhances the universality and enables the denoising process of the seismic data to be more accurate, efficient and intelligent. The noise removing method using noise2noise realizes the noise removing work of the collected noisy signals of the determined target under different noise environments. The noise removal of the seismic data is carried out by using noise2void, and the noise removal work of signals changing at any time under different noise environments is realized.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (7)
1. A method for improving the signal-to-noise ratio of seismic data is characterized by comprising the following steps:
s1: collecting multiple data records of seismic data with noise for the same seismic signal by using a distributed optical fiber sensing seismograph;
s2: training the neural network by using a noise2noise denoising method or a noise2void denoising method;
s3: and denoising the plurality of data records by using the trained neural network to complete the improvement of the signal-to-noise ratio of the seismic data.
2. The method for improving signal-to-noise ratio of seismic data according to claim 1, wherein in step S1, the method for acquiring seismic data with noise comprises: laying an optical cable along the optical fiber logging, and receiving seismic signals by using the optical cable; the distributed optical fiber sensing seismograph is adopted to acquire seismic data of seismic signals, and a group of seismic data pairs with noise are acquired and obtained for the same seismic signal.
3. The method for improving the signal-to-noise ratio of seismic data according to claim 2, wherein in the step S1, the method for acquiring the seismic data for each seismic signal comprises: the method comprises the following steps that a double-channel optical fiber sensing seismograph is used for simultaneously collecting two core optical fibers in an optical cable; or a dual-wavelength, dual-frequency or dual-code distributed optical fiber sensing seismograph is used for injecting light with two wavelengths, two frequencies and two codes into the same optical fiber, and each wavelength is collected to obtain a pair of seismic data with noise; or a double-coding distributed optical fiber sensing seismograph is used, two coded lights are injected into the same optical fiber, each wavelength is collected, and a pair of seismic data with noise is obtained.
4. The method for improving the signal-to-noise ratio of seismic data according to claim 1, wherein in step S2, if the seismic data pairs corresponding to the seismic signals are all noisy, the method for training the neural network comprises: using one of the seismic data (x)i,yi) And training the objective function, and performing average optimization on the output result.
5. The method for improving signal-to-noise ratio of seismic data according to claim 1, wherein in step S3, a pair of seismic data acquired for a same seismic signal, one of which is used as input data and the other is used as target data, is denoised by a noise2noise denoising method, or two seismic data are denoised by a noise2void denoising method respectively, and the two processed data are averaged.
6. The method for improving signal-to-noise ratio of seismic data according to claim 1, wherein in step S3, the method for acquiring seismic data for each seismic signal comprises: the method comprises the steps that a multi-channel, multi-wavelength, multi-frequency or multi-code distributed optical fiber sensing seismograph is used for simultaneously collecting n seismic data for the same seismic signal, the collected n seismic data are divided into n/2 seismic data pairs, a noise2noise denoising method is used for denoising each seismic data pair, and denoised data are averaged;
for the condition of more channels, dividing n/2 data pairs processed by a noise2noise denoising method into n/4 data pairs, and carrying out n2n denoising processing and averaging.
For the n collected seismic data, denoising by using a noise2void denoising method respectively, and then averaging;
and for the seismic signals excited in the same position in the same mode, if the signal and the noise characteristics are basically the same, exciting the seismic signals at one position n times, carrying out data acquisition on the seismic signals excited at each time, carrying out denoising, and averaging the n denoised seismic data obtained after processing to obtain the final denoised seismic signals.
7. The method for improving the signal-to-noise ratio of seismic data according to claim 1, wherein in the step S2, image denoising is realized by a noise2void denoising method, and the method for training the neural network comprises: selecting N-dimensional-N-dimensional data blocks from the seismic data pair as input of a training neural network, replacing central data with random data in the central data field to obtain target data, and performing training optimization on a target function by using the target data to complete seismic data denoising.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110138136.9A CN112965113B (en) | 2021-02-01 | 2021-02-01 | Method for improving signal-to-noise ratio of seismic data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110138136.9A CN112965113B (en) | 2021-02-01 | 2021-02-01 | Method for improving signal-to-noise ratio of seismic data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112965113A true CN112965113A (en) | 2021-06-15 |
| CN112965113B CN112965113B (en) | 2024-07-02 |
Family
ID=76272941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110138136.9A Active CN112965113B (en) | 2021-02-01 | 2021-02-01 | Method for improving signal-to-noise ratio of seismic data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112965113B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113484913A (en) * | 2021-08-16 | 2021-10-08 | 成都理工大学 | Seismic data denoising method with multi-granularity feature fusion convolution neural network |
| CN113538260A (en) * | 2021-06-21 | 2021-10-22 | 复旦大学 | LDCT image denoising and classifying method for self-supervision and supervised combined training |
| GB2611874A (en) * | 2021-10-06 | 2023-04-19 | Cgg Services Sas | Method for seismic interference noise attenuation using DNN |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120035854A1 (en) * | 2010-08-09 | 2012-02-09 | Schlumberger Technology Corporation | Seismic acquisition system including a distributed sensor having an optical fiber |
| CN108932480A (en) * | 2018-06-08 | 2018-12-04 | 电子科技大学 | The study of distributing optical fiber sensing signal characteristic and classification method based on 1D-CNN |
| CN109782339A (en) * | 2019-01-14 | 2019-05-21 | 西安交通大学 | A kind of poststack three dimensional seismic data stochastic noise suppression method based on 3D-DnCNN network |
| CN110058305A (en) * | 2019-05-24 | 2019-07-26 | 吉林大学 | A kind of DAS seismic data noise-reduction method based on convolutional neural networks |
| CN110490823A (en) * | 2019-08-14 | 2019-11-22 | 北京大学深圳研究生院 | A kind of image de-noising method under true environment |
| CN110703316A (en) * | 2019-10-23 | 2020-01-17 | 电子科技大学 | Optical fiber ground seismic wave detection method and system |
| DE102020101525A1 (en) * | 2019-01-24 | 2020-07-30 | Nvidia Corporation | BLIND-SPOT FOLDING ARCHITECTURES AND BAYESE IMAGE RECOVERY |
| CN111539879A (en) * | 2020-04-15 | 2020-08-14 | 清华大学深圳国际研究生院 | Video blind denoising method and device based on deep learning |
| CN111580161A (en) * | 2020-05-21 | 2020-08-25 | 长江大学 | Earthquake random noise suppression method based on multi-scale convolution self-coding neural network |
| US20200284937A1 (en) * | 2019-03-04 | 2020-09-10 | Chevron U.S.A. Inc. | System and method for displaying seismic events in distributed acoustic sensing data |
| WO2020242448A1 (en) * | 2019-05-24 | 2020-12-03 | Halliburton Energy Services, Inc. | Distributed acoustic sensing to geophone seismic data processing |
| CN112130195A (en) * | 2020-10-13 | 2020-12-25 | 中油奥博(成都)科技有限公司 | Time-shifting VSP data acquisition system and method based on distributed optical fiber acoustic sensing |
-
2021
- 2021-02-01 CN CN202110138136.9A patent/CN112965113B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120035854A1 (en) * | 2010-08-09 | 2012-02-09 | Schlumberger Technology Corporation | Seismic acquisition system including a distributed sensor having an optical fiber |
| CN108932480A (en) * | 2018-06-08 | 2018-12-04 | 电子科技大学 | The study of distributing optical fiber sensing signal characteristic and classification method based on 1D-CNN |
| CN109782339A (en) * | 2019-01-14 | 2019-05-21 | 西安交通大学 | A kind of poststack three dimensional seismic data stochastic noise suppression method based on 3D-DnCNN network |
| DE102020101525A1 (en) * | 2019-01-24 | 2020-07-30 | Nvidia Corporation | BLIND-SPOT FOLDING ARCHITECTURES AND BAYESE IMAGE RECOVERY |
| US20200284937A1 (en) * | 2019-03-04 | 2020-09-10 | Chevron U.S.A. Inc. | System and method for displaying seismic events in distributed acoustic sensing data |
| CN110058305A (en) * | 2019-05-24 | 2019-07-26 | 吉林大学 | A kind of DAS seismic data noise-reduction method based on convolutional neural networks |
| WO2020242448A1 (en) * | 2019-05-24 | 2020-12-03 | Halliburton Energy Services, Inc. | Distributed acoustic sensing to geophone seismic data processing |
| CN110490823A (en) * | 2019-08-14 | 2019-11-22 | 北京大学深圳研究生院 | A kind of image de-noising method under true environment |
| CN110703316A (en) * | 2019-10-23 | 2020-01-17 | 电子科技大学 | Optical fiber ground seismic wave detection method and system |
| CN111539879A (en) * | 2020-04-15 | 2020-08-14 | 清华大学深圳国际研究生院 | Video blind denoising method and device based on deep learning |
| CN111580161A (en) * | 2020-05-21 | 2020-08-25 | 长江大学 | Earthquake random noise suppression method based on multi-scale convolution self-coding neural network |
| CN112130195A (en) * | 2020-10-13 | 2020-12-25 | 中油奥博(成都)科技有限公司 | Time-shifting VSP data acquisition system and method based on distributed optical fiber acoustic sensing |
Non-Patent Citations (1)
| Title |
|---|
| 王函等: "采用无干净标签的深度学习方法衰减地震噪声", 《2020年中国地球科学联合学术年会论文集（十四）》, pages 1336 - 1337 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538260A (en) * | 2021-06-21 | 2021-10-22 | 复旦大学 | LDCT image denoising and classifying method for self-supervision and supervised combined training |
| CN113538260B (en) * | 2021-06-21 | 2022-04-12 | 复旦大学 | LDCT image denoising and classifying method for self-supervision and supervised combined training |
| CN113484913A (en) * | 2021-08-16 | 2021-10-08 | 成都理工大学 | Seismic data denoising method with multi-granularity feature fusion convolution neural network |
| CN113484913B (en) * | 2021-08-16 | 2023-06-16 | 成都理工大学 | Seismic data denoising method for multi-granularity feature fusion convolutional neural network |
| GB2611874A (en) * | 2021-10-06 | 2023-04-19 | Cgg Services Sas | Method for seismic interference noise attenuation using DNN |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112965113B (en) | 2024-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112965113B (en) | Method for improving signal-to-noise ratio of seismic data | |
| CN109858408B (en) | Ultrasonic signal processing method based on self-encoder | |
| CN113484257B (en) | Water quality concentration calculating system and method based on neural network and full spectrum absorbance | |
| CN108550116A (en) | The self-adapting random resonant denoising method of silicon monocrystal growth image under low signal-to-noise ratio | |
| Wang et al. | Rapid response DAS denoising method based on deep learning | |
| CN113033796A (en) | Image identification method of all-optical nonlinear diffraction neural network | |
| CN116576956A (en) | Multisource vibration signal separation method based on distributed optical fiber acoustic wave sensing | |
| CN116662738A (en) | Distributed vibration sensing signal denoising method, electronic device and storage medium | |
| CN112634159A (en) | Hyperspectral image denoising method based on blind noise estimation | |
| Mrozek | Simultaneous monitoring of chromatic dispersion and optical signal to noise ratio in optical network using asynchronous delay tap sampling and convolutional neural network (deep learning) | |
| CN117788379A (en) | Domain transformation-based end-to-end heterogeneous remote sensing image change detection method | |
| CN113624447A (en) | Composite denoising method for fiber bragg grating sensor network | |
| CN114598577B (en) | Multi-band signal fusion filtering method for 5G communication system | |
| Dumont et al. | Deep learning on real geophysical data: A case study for distributed acoustic sensing research | |
| CN113465676B (en) | Double-stage Brillouin frequency shift extraction method | |
| CN115470863A (en) | Domain generalized electroencephalogram signal classification method based on double supervision | |
| CN111856559B (en) | Multi-channel seismic spectrum inversion method and system based on sparse Bayes learning theory | |
| US20210150401A1 (en) | Entanglement-enhanced machine learning with quantum data acquisition | |
| CN110275157B (en) | Vector sound orientation method based on genetic algorithm self-adaptive random resonance | |
| CN112241024B (en) | Method for improving signal-to-noise ratio of seismic data, computer storage medium and system | |
| CN116046893B (en) | Ultrasonic guided wave signal enhancement method and system based on multi-mode recognition-fusion | |
| CN116755151B (en) | Anti-diffraction wave field separation method based on iterative focusing optimization | |
| CN116577829B (en) | Automatic extraction method based on background noise dispersion curve | |
| CN113126163B (en) | Five-dimensional seismic data noise attenuation method and device | |
| Yang et al. | Deep learning with soft attention mechanism for small-scale ground roll attenuation |
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 |