CN102323619B - Linear denoising method based on multi-core processor - Google Patents
Linear denoising method based on multi-core processor Download PDFInfo
- Publication number
- CN102323619B CN102323619B CN 201110136526 CN201110136526A CN102323619B CN 102323619 B CN102323619 B CN 102323619B CN 201110136526 CN201110136526 CN 201110136526 CN 201110136526 A CN201110136526 A CN 201110136526A CN 102323619 B CN102323619 B CN 102323619B
- Authority
- CN
- China
- Prior art keywords
- centerdot
- geological data
- linear
- big gun
- single big
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a linear denoising method based on a multi-core processor, which comprises the steps that: multi-shot earthquake data is read into a main node of the multi-core processor; the main node distributes a plurality of pieces of single-shot earthquake data in the multi-shot earthquake data to a plurality of sub-nodes of the multi-core processor; all the sub-nodes carry out linear denoising to the single-shot earthquake data from the main node; and the main node collects the single-shot earthquake data which is linearly denoised from the sub-nodes, and outputs linearly denoised earthquake data. Wherein, when the sub-nodes carry out linear denoising, every time when the velocity of linear interference which is scanned from the single-shot earthquake data to be processed is consistent with any apparent velocity within a preset linear interference apparent velocity range, an optimized prediction filter operator is used for each frequency component in a frequency domain for prediction filtration along an X-direction after the same-phase shaft of the linear interference is flattened.
Description
Technical field
The present invention relates to a kind ofly many big guns geological data is being carried out the method for denoising, relate in particular to and a kind ofly based on polycaryon processor, many big guns geological data is carried out the method for linear denoising concurrently.
Background technology
In recent years, development along with the oil-gas exploration situation, the scope of seismic prospecting is also more and more extensive, more and more deep, seismic survey work also turns to western part from the east gradually, turned to the complicated terrain areas such as desert, hills, mountain region by the Plain, the simple easy-to-handle work of the sort of collection comes to an end basically, and the task of leaving the geophysical survey worker for is very arduous.In the seismic data that complicated earth surface area gathers, the signal to noise ratio (S/N ratio) of data is very low, and ubiquity quite serious line noise, generally this noise like severe jamming the usable reflection signal, greatly reduce the signal to noise ratio (S/N ratio) of seismic data.
Traditional linear denoising method is carried out the line noise removal to many big guns geological data in time domain or frequency domain.
General time domain linear denoising method is suitable for the linear disturbance noise in prestack two dimension or three-dimensional geological data is processed, and seismic signal is had good hi-fi of amplitude.The defective of these class methods is that speed is slower, and especially for the 3D seismic data that contains several thousand big guns even up to ten thousand, processing speed problem is even more serious, is difficult to satisfy actual needs.
Common frequency domain linear denoising method often causes obvious signal distortion, also filtered some effective compositions when eliminating interference, and smoothing effect is serious, make the entire profile seem stiff, so denoising effect is undesirable.
Summary of the invention
The object of the present invention is to provide and a kind ofly based on polycaryon processor, many big guns geological data is carried out the method for linear denoising concurrently, thereby improve the efficient that data are processed, and shorten the denoising time of geological data.
Another object of the present invention is to provide a kind of and based on polycaryon processor, many big guns geological data is carried out the method for linear denoising concurrently, optimize the algorithm that the frequency domain linear denoising is processed, thereby improve the fidelity effect through the geological data of linear denoising.
To achieve these goals, provide a kind of method of the linear denoising based on polycaryon processor, comprising: the host node of polycaryon processor reads in many big guns geological data; A plurality of single big gun geological data in many big guns geological data that host node will read in is distributed to a plurality of child nodes of polycaryon processor; Each child node of described a plurality of child nodes goes line noise to process to the single big gun geological data that comes autonomous node; Collect from described a plurality of child nodes single big gun geological data of processing through the past line noise with host node, and output is through the geological data of past line noise processing.Wherein, when each child node goes line noise to process, when the speed of the linear disturbance that scans from pending single big gun geological data is consistent with arbitrary apparent velocity in the apparent velocity scope of predetermined linear disturbance, after evening up the lineups of linear disturbance, to each frequency content, use the predictive filtering operator that draws according to following equation to carry out predictive filtering along directions X in frequency field:
Wherein, Y (x, w) carries out the input signal that Fast Fourier Transform (FFT) obtains, r to geological data A (x, t)
wwBe that frequency is the auto-correlation of the input signal Y (x, w) of w, m is the length of predictive filtering operator, c (0), c (1) ... c (m) is the predictive filtering operator.
Described each child node can be made described single big gun geological data the small echo frequency division and process before pending single big gun geological data is scanned, to isolate the geological data of linear disturbance place frequency band range.
Described host node can be distributed to a plurality of single big gun geological data corresponding to the child node number child node in batches and go line noise to process, and output is through the geological data of past line noise processing.
Described polycaryon processor can be heterogeneous multi-nucleus processor.
The master of heterogeneous multi-nucleus processor endorses with corresponding with described host node, and heterogeneous multi-nucleus processor is corresponding with described child node from core.
Description of drawings
By the description of carrying out below in conjunction with accompanying drawing, above and other purpose of the present invention and characteristics will become apparent, wherein:
Fig. 1 is the process flow diagram based on the processing of the linear denoising method of polycaryon processor that illustrates according to exemplary embodiment of the present invention; With
Fig. 2 illustrates each child node in Fig. 1 single big gun geological data to be carried out the process flow diagram of the step that linear denoising processes.
Embodiment
Below, describe embodiments of the invention in detail with reference to accompanying drawing.
Polycaryon processor is integrated two or more complete computing engines in one piece of processor, and described each computing engines is called as core.Polycaryon processor is divided into isomorphism and two kinds of structures of isomery.According to inventive concept of the present invention, use a plurality of cores of polycaryon processor to carry out the denoising of multinuclear geological data.In the present invention, take heterogeneous multi-nucleus processor as example, to the linear denoising that walks abreast of multinuclear geological data.But, through simply revising, also can realize in the isomorphism polycaryon processor.
For convenience of explanation, to be responsible in the present invention reading many big guns geological data, control other cores and carry out denoising, and output is called " host node " through the core of the geological data of denoising, and will be called " child node " by other cores that denoising is carried out in control.In heterogeneous multi-nucleus processor, the master of described " host node " and heterogeneous multi-nucleus processor check should, and " child node " and heterogeneous multi-nucleus processor should from checking.
Fig. 1 is the process flow diagram based on the processing of the linear denoising method of polycaryon processor that illustrates according to exemplary embodiment of the present invention.With reference to Fig. 1, at step S110, the host node of polycaryon processor reads in many big guns geological data.According to exemplary embodiment of the present invention, described many big guns geological data can be read from storer, hard disk, flash memory, CD or the network storage medium such as polycaryon processor.
At step S120, a plurality of single big gun geological data in many big guns geological data that host node will read in is distributed to a plurality of child nodes of polycaryon processor.In general, need many big guns geological data to be processed usually to comprise the geological data of several thousand or up to ten thousand shot points.According to exemplary embodiment of the present invention, can as required, use the parton node of polycaryon processor or whole child node to carry out denoising to the single big gun data that comprise.Single big gun geological data of the each processing of each child node.
At step S130, each child node of described a plurality of child nodes goes line noise to process to the single big gun geological data that comes autonomous node.Describe child node in detail with reference to Fig. 2 after a while the single big gun geological data that receives is carried out the processing of linear denoising.
At step S140, host node collects from described a plurality of child nodes single big gun geological data of processing through the past line noise, and output is through the geological data of past line noise processing.
In technical scheme of the present invention, many big guns geological data that host node can read in as the unit batch treatment take the number of child node, every batch will be distributed to each child node corresponding to a plurality of single big gun geological data of the number of child node, then after collecting the denoising result of whole child nodes, export whole denoising results, then begin the processing of a plurality of single big gun geological datas of next group.
Fig. 2 illustrates child node in Fig. 1 single big gun geological data to be carried out the process flow diagram of the step that linear denoising processes.
Single big gun geological data that need are processed, according to a preferred embodiment of the invention, at first child node can carry out the small echo frequency division and process, to isolate the geological data of linear disturbance place frequency band range.Yet it is not the step that must carry out, directly execution in step S210 that described frequency division is processed.
At step S210, child node is to single big gun geological data of need processing or as above carry out linear sweep through the isolated geological data of frequency division processing.In scanning process, at step S220, whether the arbitrary apparent velocity in the apparent velocity scope of the speed that child node is determined the linear disturbance that scans and predetermined linear disturbance is consistent.If at step S220, determine that the speed of the linear disturbance that scans and the apparent velocity in described apparent velocity scope are all inconsistent, return to step S210 and proceed linear sweep.
If at step S220, determine that the speed of the linear disturbance that scans is consistent with arbitrary apparent velocity in described apparent velocity scope, at step S230, child node is evened up the lineups of described linear disturbance.According to exemplary embodiment of the present invention, child node can be carried out the described processing of evening up according to following formula:
A(x，t)＝g(x，t)+s(x，t)
Wherein, x is road, space serial number, and t is the time-sampling point, and g (x, t) is the linear disturbance in geological data, and s (x, t) is the usable reflection in geological data, and A (x, t) is geological data.
Then, at step S230, the geological data that child node is evened up processing in frequency field to process carries out linear prediction.Be specially, will be through the geological data of evening up processing by being converted to frequency domain data as Fast Fourier Transform (FFT) (FFT); Then, to each frequency content, along the X-direction filtering that gives a forecast; The frequency domain data that to process through predictive filtering again is by converting back time domain data as inverse fast Fourier transform (IFFT).Wherein, in the processing of filtering that each frequency content is given a forecast along X-direction, selection can make the minimum predictive filtering operator of error energy.According to exemplary embodiment of the present invention, choose described predictive filtering operator c (s) according to following equation:
Wherein, Y (x, w) carries out the input signal that Fast Fourier Transform (FFT) obtains, r to geological data A (x, t)
wwBe that frequency is the auto-correlation of the input signal Y (x, w) of w, m is the length of predictive filtering operator, c (0), c (1) ... c (m) is the predictive filtering operator.
After the frequency-domain linear prediction filtering of completing steps S240, at step S250, check the linear sweep of whether having completed geological data.If determine to remain unfulfilled described linear sweep, return to step S210, proceed scan process.
If at step S250, determine to have completed described linear sweep, at step S260, deduct from the original single big gun geological data that receives the linear disturbance that obtains through linear prediction filtering, thereby remove the line noise of single big gun geological data.
Method according to the linear denoising based on polycaryon processor of the present invention, by using a plurality of cores of polycaryon processor, many big guns geological data is carried out linear denoising concurrently to be processed, and use the frequency-domain linear prediction filtering of optimizing to process, not only improved the efficient that geological data is processed, shorten the denoising time of geological data, also improved the fidelity effect through the geological data of linear denoising.
The invention is not restricted to above-described embodiment, without departing from the present invention, can carry out various changes and modifications.
Claims (4)
1. method based on the linear denoising of polycaryon processor comprises:
The host node of polycaryon processor reads in many big guns geological data;
A plurality of single big gun geological data in many big guns geological data that host node will read in is distributed to a plurality of child nodes of polycaryon processor;
Each child node of described a plurality of child nodes goes line noise to process to the single big gun geological data that comes autonomous node; With
Host node collects from described a plurality of child nodes single big gun geological data of processing through the past line noise, and exports the geological data of processing through the past line noise,
Wherein, when described each child node goes line noise to process, when the speed of the linear disturbance that scans from pending single big gun geological data is consistent with arbitrary apparent velocity in the apparent velocity scope of predetermined linear disturbance, after evening up the lineups of linear disturbance, to each frequency content, use the predictive filtering operator that draws according to following equation to carry out predictive filtering along directions X in frequency field:
Wherein, Y (x, w) carries out the input signal that Fast Fourier Transform (FFT) obtains, r to geological data A (x, t)
wwBe that frequency is the auto-correlation of the input signal Y (x, w) of w, m is the length of predictive filtering operator, c (0), c (1) ... c (m) is the predictive filtering operator.
2. the method for claim 1, is characterized in that, described each child node was made the small echo frequency division with described single big gun geological data and processed before pending single big gun geological data is scanned, to isolate the geological data of linear disturbance place frequency band range.
3. the method for claim 1, is characterized in that, described host node will be distributed to corresponding to a plurality of single big gun geological data of child node number child node in batches and go line noise to process, and output is through the geological data of past line noise processing.
4. the method for claim 1, is characterized in that, described polycaryon processor is heterogeneous multi-nucleus processor,
Wherein, the main core of heterogeneous multi-nucleus processor is corresponding with described host node, and heterogeneous multi-nucleus processor is corresponding with described child node from core.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110136526 CN102323619B (en) | 2011-05-25 | 2011-05-25 | Linear denoising method based on multi-core processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110136526 CN102323619B (en) | 2011-05-25 | 2011-05-25 | Linear denoising method based on multi-core processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102323619A CN102323619A (en) | 2012-01-18 |
| CN102323619B true CN102323619B (en) | 2013-06-12 |
Family
ID=45451398
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110136526 Expired - Fee Related CN102323619B (en) | 2011-05-25 | 2011-05-25 | Linear denoising method based on multi-core processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102323619B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459781B (en) * | 2014-12-09 | 2017-09-01 | 中国石油天然气集团公司 | The random noise attenuation method of three-dimensional earthquake data before superposition |
| CN106547025A (en) * | 2015-09-22 | 2017-03-29 | 中国石油化工股份有限公司 | A kind of mean value weighting denoising method |
| CN106201673B (en) * | 2016-06-24 | 2019-07-09 | 中国石油天然气集团公司 | A kind of seismic data processing technique and device |
| CN108230370B (en) * | 2017-12-29 | 2020-08-04 | 厦门市美亚柏科信息股份有限公司 | Tracking target speed prediction method based on holder and storage medium |
| CN108230371B (en) * | 2017-12-29 | 2020-08-18 | 厦门市美亚柏科信息股份有限公司 | Tracking target speed prediction method based on holder and storage medium |
| CN112230284A (en) * | 2019-07-15 | 2021-01-15 | 中国石油天然气集团有限公司 | Parallel random noise attenuation method, monitoring method and node for three-dimensional pre-stack data |
| CN111736224B (en) * | 2020-07-14 | 2021-04-20 | 西安交通大学 | Method, storage medium and equipment for suppressing linear interference of pre-stack seismic data |
| CN114895459B (en) * | 2022-05-17 | 2023-10-03 | 中国科学院光电技术研究所 | Surface layer self-adaptive optical wavefront real-time controller |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2105332C1 (en) * | 1994-11-23 | 1998-02-20 | Опытно-методическая сейсмологическая партия Института вулканологии Дальневосточного отделения РАН | Method testing stressed state of the earth's crust for prediction of heavy earthquakes |
| CN101487898A (en) * | 2009-02-27 | 2009-07-22 | 中国石油集团川庆钻探工程有限公司 | Method for oil gas water recognition by employing longitudinal wave seismic exploration post-stack data |
| CN101923175A (en) * | 2009-11-17 | 2010-12-22 | 中国科学院地质与地球物理研究所 | Method for directly generating angle gathers by using wave-equation migration |
| CN101930079A (en) * | 2009-06-26 | 2010-12-29 | 西安英诺瓦物探装备有限公司 | Method for processing relevant/stack data in seismic prospecting |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5715213A (en) * | 1995-11-13 | 1998-02-03 | Mobil Oil Corporation | High fidelity vibratory source seismic method using a plurality of vibrator sources |
-
2011
- 2011-05-25 CN CN 201110136526 patent/CN102323619B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2105332C1 (en) * | 1994-11-23 | 1998-02-20 | Опытно-методическая сейсмологическая партия Института вулканологии Дальневосточного отделения РАН | Method testing stressed state of the earth's crust for prediction of heavy earthquakes |
| CN101487898A (en) * | 2009-02-27 | 2009-07-22 | 中国石油集团川庆钻探工程有限公司 | Method for oil gas water recognition by employing longitudinal wave seismic exploration post-stack data |
| CN101930079A (en) * | 2009-06-26 | 2010-12-29 | 西安英诺瓦物探装备有限公司 | Method for processing relevant/stack data in seismic prospecting |
| CN101923175A (en) * | 2009-11-17 | 2010-12-22 | 中国科学院地质与地球物理研究所 | Method for directly generating angle gathers by using wave-equation migration |
Non-Patent Citations (2)
| Title |
|---|
| 平滑算子在地震叠前深度域成像中的应用;李振春等;《中国石油大学学报(自然科学版)》;20081220(第06期);全文 * |
| 李振春等.平滑算子在地震叠前深度域成像中的应用.《中国石油大学学报(自然科学版)》.2008,(第06期), |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102323619A (en) | 2012-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102323619B (en) | Linear denoising method based on multi-core processor | |
| CN102193107B (en) | Method for separating and denoising seismic wave field | |
| CN106597539B (en) | For the bent wave zone Radon converter noise drawing method of Huangtuyuan area | |
| CN102681014B (en) | Regular linear interference suppressing method based on polynomial fitting | |
| CN108549106B (en) | Aliasing noise drawing method and device | |
| CN102636811B (en) | Eliminating method of multiple waves in bidimensional seismic data on sea | |
| CN101382598B (en) | Pressing method for true 3-d seismics data linear noise | |
| CN104007469A (en) | Weak seismic signal reconstruction method based on curvelet transform | |
| CN102707314A (en) | Deconvolution method of multi-path double-spectral domain mixed phase wavelets | |
| CN105572723B (en) | The design method of controlled source scanning signal based on autocorrelation wavelet | |
| CN109738952A (en) | The direct offset imaging method in passive source based on full waveform inversion driving | |
| CN104614769A (en) | Beam-forming filtering method for suppressing seismic surface waves | |
| CN110261910A (en) | Seismic data surface wave minimizing technology based on adaptive sparse S-transformation | |
| CN104597502A (en) | Novel petroleum seismic exploration data noise reduction method | |
| CN104330826A (en) | A method for removing various noises under the condition of complex surface | |
| WO2021127382A1 (en) | Full waveform inversion in the midpoint-offset domain | |
| CN105044769B (en) | The method for improving the resolution ratio of seismic signal | |
| CN105319593A (en) | Combined denoising method based on curvelet transform and singular value decomposition | |
| CN104133248A (en) | High-fidelity sound wave interference suppression method | |
| CN105676292A (en) | 3D earthquake data de-noising method based on 2D curvelet transform | |
| CN103954993A (en) | Scale domain multichannel filtering method and device based on seismic signal continuous wavelet transformation | |
| CN103645504A (en) | Weak earthquake signal processing method based on generalized instantaneous phase and P norm negative norm | |
| CN105182414A (en) | Direct wave removing method based on wave equation forward modeling | |
| CN115993641B (en) | Method for extracting passive source surface wave dispersion curve | |
| AU2006237350B2 (en) | Seismic data processing method for RMO picking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130612 Termination date: 20180525 |