CN103149591B - Method for automatically picking up seismic reflection event based on Kalman filtering - Google Patents
Method for automatically picking up seismic reflection event based on Kalman filtering Download PDFInfo
- Publication number
- CN103149591B CN103149591B CN201310065332.3A CN201310065332A CN103149591B CN 103149591 B CN103149591 B CN 103149591B CN 201310065332 A CN201310065332 A CN 201310065332A CN 103149591 B CN103149591 B CN 103149591B
- Authority
- CN
- China
- Prior art keywords
- lineups
- point
- seismic
- kalman filtering
- kth
- 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
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method for automatically picking up a seismic reflection event based on Kalman filtering. According to the scheme, the method comprises the following steps of: establishing a state equation and a measurement equation of a Kalman filtering system for picking up seismic reflection event information on the basis of a seismic reflection hyperbolic time-distance equation according to movement rules of seismic reflection waves; detecting the time of arrival of seismic wavelets in each seismic signal, thereby obtaining measurement data for Kalman filtering; and automatically initiating the event according to data in a free measurement point set, and performing gradual iterative filtering according to a sequence of prediction, observation and correction. A data association technology is used for solving the problem of simultaneous tracking of a plurality of events.
Description
Technical field
The invention belongs to seismic prospecting signal processing technology field, be specifically related to seismic reflection lineups automatic Picking technology.
Background technology
Since early 1960s kalman filter method be suggested, it is widely used in the field such as target following, navigation, electric system, medical science, automatically control, meteorology, communication, seismic prospecting.In field of seismic exploration, Kalman filtering is mainly used in the deconvolution of seismic data, to improve resolution and the signal to noise ratio (S/N ratio) of seismic prospecting signal.And most information entrained in seismic signal is included in lineups substantially, as: the self excitation and self receiving time t of lineups
0reflect the depth information at interface to a certain extent, the form reflection seismic event velocity of propagation information in media as well of lineups, in lineups, seismic wavelet change is derived from the condition etc. of seismic wave propagation process.Therefore detection and the pickup of lineups are to the process of seismic data with explain most important.
The mode of usual acquisition reflected seismic information is: by man-made explosion (conventional explosive) earthquake-wave-exciting; utilize seismic event in its communication process, run into the different roch layer interface of medium character and have the character that part energy is reflected back ground; then received by each wave detector being arranged in distance different from shot point and obtain multitrace seismogram (signal), the corresponding wave detector of per pass seismic signal.When underground has multiple reflecting interface, per pass seismic signal comprises multiple reflection seismic wavelet.Wherein, the line of the extreme value (being commonly called as crest or trough) that seismologic record Shang Ge road vibration phase is identical is called lineups.In order to obtain the lineups information in seismic data, multiple different detection and pick-up method are developed both at home and abroad.Mainly comprise: (1) edge detection method.These class methods convert the amplitude of each for two-dimension earthquake signal sampled point to different gray-scale value, seismic channel set are seen as a width gray level image thus utilize the edge detection method in image procossing to detect lineups.The lineups that these class methods obtain are fuzzyyer, and resolution is not high, affect comparatively large, and the result detected are the envelopes of gray scale Sudden change region, directly as detection of phase axis result, can not also need some subsequent treatment, as thinning processing etc. by noise factor.(2) neural network.These class methods utilize known lineups to make master sample training network, utilize error to return method and progressively revise connection weights between neuron, can use the data that network processes is new after reaching best weight value distribution.Neural network training process needs abundant learning sample, and the selection of sample exists larger difficulty, and the indoctrination session of new samples affects the network trained, and a large amount of interative computations causes process consuming time longer.(3) cross-correlation method, High order statistics and coherent algorithm.The ultimate principle of these three kinds of methods be all utilize lineups between seismic trace waveform similarity feature to extract lineups, easily affected by noise, and decline along with the increase spatial resolution calculating road, simultaneously calculated amount is larger.(4) chain matching method.The method represents the chain that the crest of the some features of per pass waveform band and trough form, then the pickup of lineups is just converted into the Optimum Matching problem between each chain, namely finds in all paths and makes total cost be minimum link.The method is equally easily affected by noise and be difficult to solve and intersect lineups problems.(5) method of interpolation and fitting process.Method of interpolation is some reference mark (Seed Points) of seismic data first manually being picked up to seismic event, then by method of interpolation to reference mark data interpolating, then filter issuable high-frequency anomaly value in Interpolation Process with wave filter.The basic thought of fitting process is the curve being considered as lineups to have certain dynamic law, described by the time series of a broad sense, and go this time series of matching with certain model (as: AR model), the time series that wherein suitable model and ripple reach is still more doubt, and these are all by lineups fitting result follow-up for impact.(6) Tu etc. propose a kind of PICKING UP THE EVENTS AUTOMATICALLY system (Tu P., Mason I., Zisserman A.An automated system for picking seismic events.SEG63rdannual meeting, 1993:234 – 237).The method by two-dimentional matched filter, Kalman filter and Flexible formwork assembly three step form.But the concrete state equation do not provided in literary composition needed for Kalman filter and measurement equation, method describes simple, has no follow-up article and reports further simultaneously.(7) chaotic oscillator detection method.The method utilizes the Duffing establishing equation revised to detect the chaotic oscillator system of Weak event, afterwards according to the different time-speed of hypothesis to through scan process to lineups, forming new wavelet constant duration sequence sends in chaotic oscillator system, if there is when large change shows this m-speed to there are lineups in oscillator system phase, otherwise does not exist.The method still can detect lineups more exactly when seismic data signal to noise ratio (S/N ratio) is extremely low, but the method due to needs search time-speed pair, and all demand solution chaotic oscillator differential equation group during each search, operational efficiency is lower.
In sum, up to the present, domestic and international seismologic record lineups pickup technology has achieved many achievements, but the characteristics of motion of most method and seismic event is in conjunction with less, and picks up many lineups simultaneously and have any problem.
Summary of the invention
In view of this, the invention provides a kind of seismic reflection lineups automatic pick method based on Kalman filtering, give the Kalman filter of the characteristics of motion design of seismic wave in combination, utilize data association technique to realize picking up while many lineups simultaneously.
In order to solve the problems of the technologies described above, the invention provides a kind of seismic reflection lineups automatic Picking scheme based on Kalman filtering, the program is first according to the characteristics of motion of earthquake reflected wave, state equation and the measurement equation of the Kalman filtering system for picking up seismic reflection lineups information is set up based on seismic reflection hyperbolic time-distance equation, secondly detect the seismic wavelet time of arrival in each road seismic signal thus obtain Kalman filtering metric data used, according to the automatic initial lineups of the data in free gauge point set, and the river pagination progressive alternate filtering of foundation " prediction one observation one corrects ", data association technique is wherein utilized to solve many lineups tracking problem simultaneously, thus the automatic Picking technology of seismic reflection lineups in a set of new noise is provided.
First the Kalman filter that the present invention sets up according to the characteristics of motion of earthquake reflected wave is described below.
In reflection wave seismic exploration field, seismic reflection wave traveling T-X curve is generally approximately hyperbolic curve, and its equation of motion meets hyperbolic time-distance equation.Consider certain even level's layered medium, the mean propagation velocity of seismic wave on reflecting interface in medium is v hypothetically, and the self excitation and self receiving time of shot point is t
0, the geophone offset of N number of wave detector is respectively x
1, x
2..., x
k, x
k+1..., x
n, the ripple that each wave detector receives this reflecting interface reflection wave reaches the time and is respectively t
1, t
2..., t
k, t
k+1..., t
n, then according to earthquake reflected wave hyperbolic time-distance equation
have
As can be seen from formula (1), it is relevant that state and the ripple in a kth detection road in a kth detection road reach the average velocity that time and seismic event be transmitted in a kth wave detector process through boundary reflection by shot point, based on this, first set up state equation and the measurement equation of the Kalman filtering system for following the tracks of seismic reflection lineups.The state vector in a kth detection road is made to be
Wherein v
kthe average velocity that to be seismic event by shot point be transmitted to through boundary reflection in a kth wave detector process, then state equation can be write as
X
k+1＝F
kX
k+U
k， （2）
Wherein state-transition matrix
Consider the complicacy of underground medium and cause earthquake reflected wave T-X curve to depart from hyperbolic curve and many uncertain phenomenons, in above-mentioned state equation, introducing process noise U
k, this process noise portrays the error between the actual characteristics of motion in underground propagation process of seismic event and the state equation model set up.Consider the correlativity between state vector two component, make process noise U
k=Γ
ku
k, wherein process noise distribution matrix
U
kfor zero-mean gaussian process noise, its variance is σ
u 2.
The measurement vector making a kth detection road is Z
k=[z
k 2], wherein z
kfor in the kth detection road that extracts from seismologic record, the ripple of each seismic wavelet reaches the time, then measurement equation can be write as
Z
k+1＝H
K+1X
k+1+W
k+1， (2)
Wherein measurement matrix H
k+1=[1 0].Consider inevitably can there is noise in the reception and measuring process of seismic event, in above-mentioned measurement equation, introduce measurement noise W
k+1, such as: the neighbouring noise etc. that the rustle of leaves in the wind causes of the thermonoise of wave detector, wave detector all can be regarded measurement noise as and carry out modeling.
Based on the model of the above-mentioned Kalman filtering system for following the tracks of seismic reflection lineups set up, comprise the steps: from a N road shot record migration seismic reflection record pickup seismic reflection lineups information
Step S00, initialization.Put free gauge point set, lineups set for empty, iterative steps k=1.
Step S01, extracts ripple and reaches time point, i.e. gauge point.Take out seismologic record Zhongk road seismic signal, the ripple extracting seismic wavelet in this track data reaches time point { z
k(i), i=1,2 ..., m
k, m
krepresent the seismic wavelet number extracted in kth road seismic signal, record the amplitude { A of each time point position seismic wavelet simultaneously
k(i), i=1,2 ..., m
k.Extract the method that ripple reaches time point in this step and can adopt conventional Threshold detection method, Deconvolution etc.
Step S02, data correlation.According to predicting the outcome of Kalman filtering iterative process last time, namely step S04 predicts that in k-1 iteration the position of each lineups on the kth road seismic signal that obtains and ripple reach temporal predictive value z
k|k-1, the ripple extracted in step S01 is reached time point (gauge point) and associate with each lineups in lineups set, concrete interrelational form is:
For each gauge point z extracted
ki (), judges whether it falls into the z that predicts the outcome of any one lineups of lineups set
k|k-1centered by preset time in region, if so, then this gauge point belongs to this lineups, and namely this gauge point can be used as the candidate point of the lineups that it falls into and is associated, by this gauge point stored in candidate data set corresponding to association lineups.If gauge point does not all associate with any lineups in lineups set, then by this gauge point stored in free point set.If lineups set is empty, now the direct all gauge points extracted by step S01 in epicycle process are stored in free point set.
Same gauge point likely closes links multiple different existing lineups, and this phenomenon easily occurs when two lineups intersect or be very near each other; Simultaneously because the expansion of seismic wavelet or the impact of noise may have multiple gauge point to be associated with same lineups, but same lineups should have an only gauge point in per pass seismic signal, all data in corresponding kth road in the candidate data set of these lineups are now utilized to calculate a synthetic quantity measuring point, as the metric data participating in kth time Kalman filtering iteration.
Wherein, the calculating of synthetic quantity measuring point can adopt probability weighted method
or the maximum method of amplitude
deng.Wherein, p (i) belongs to the probability of corresponding lineups for i-th gauge point in candidate data set, and J is the quantity of gauge point in candidate data set.
Step S03, initial new lineups.According to data acquisition M/L(in free point set wherein M and L be positive integer, and M is less than or equal to L) the initial new lineups of direct-vision method, namely according to the velocity of wave of arbitrary continuation two-point estimate reflection seismic waves in continuous L road gauge point
the span of l is 1,2 ... L-1, if there is the velocity of wave of M point to meet given speed requirement, then can initial new lineups, these new lineups are added in lineups set, and deletes being used for M the gauge point that initial new lineups use from free point set.The initialization of Kalman filter corresponding to new lineups is carried out after initial success.
Step S04, carries out this Kalman filtering iteration to lineups all in lineups set one by one.Iterative process is identical with conventional Kalman Filtering Model, is summarized as follows.To certain lineups, the state vector prediction produced according to last iteration, state estimation covariance prediction and measure vector predictors and current associate after the metric data that obtains calculate new breath, new breath covariance and filter gain, thus obtain state estimation and the state estimation covariance matrix of current iteration, then provide the predicted value z of each lineups needed for next iteration
k+1|k.
Step S05, makes k=k+1, repeats step S01 ~ S04, until last track data iteration is complete.Contain the seismic event information that will pick up in a series of state vectors then finally obtained, comprise the positional information t of lineups
kwith velocity information v
k.
So far, this flow process terminates.
In lineups pick process, constantly having cancelling of new lineups initial, old lineups termination and false lineups, in order to improve the adaptability of lineups automatic Picking, between step S04 and S05, increasing the step of lineups management further.In this step, give the quality score of each lineups according to the lineups rule drafted according to the data correlation situation of step S02 in epicycle iteration, and upgrade lineups state according to scoring, comprise the maintenance of lineups, the cancelling of termination and false lineups.
Preferably, when after the initial new lineups of employing M/L logical approach in step S03, being defined as by new lineups may lineups.So in these lineups management process, when the mark of possibility lineups is higher than the first scoring line set, possible lineups confirm as reliable lineups; Possible lineups and reliable lineups all adopt and associate bonus point at every turn, the principle of not associated deduction carries out quality score; Cancelled when possibility lineups mark is less than the second scoring line of setting, and deleted from lineups set; Terminated when reliable lineups scoring is less than the 3rd scoring line of setting; When the scoring of reliable lineups is greater than scoring above in limited time, then do not continue upwards to add up.
The present invention, compared with existing seismic reflection lineups pick-up method, has following beneficial effect:
(1) state equation set up and measurement equation take full advantage of the characteristics of motion of earthquake reflected wave, and filter result will be made more accurate.
(2) the present invention can pick up many reflection line-ups in seismologic record simultaneously.
(3) the present invention adopts lineups management process to make the maintenance of lineups have certain adaptability.
Accompanying drawing explanation
Fig. 1 is seismic reflection lineups automatic pick method schematic flow sheet;
Fig. 2 is the seismic reflection record of synthesis;
Fig. 3 reaches time point for extracted seismic wavelet ripple;
Fig. 4 is picked up lineups position.
Embodiment
Below in conjunction with accompanying drawing and a specific embodiment, technical solution of the present invention is further explained.
First be the two lineups seismic reflection records in 64 roads by the total number of channels of parameter synthesis one shown in table 1, as shown in Figure 2, in figure, record has superposed that average is zero, variance is σ
2the white Gaussian noise of=0.05W, visible very noisy masks lineups, makes the identification of lineups become difficulty.
Table 1 composite traces parameter used
Below for utilizing the concrete implementation step of method automatic Picking two lineups proposed:
Step S00, initialization.Put free gauge point set, lineups set for empty, iterative steps k=1.
Step S01, take out 64 seismologic record Zhongk road, road seismic signals, the ripple utilizing Threshold detection method to extract seismic wavelet in this track data reaches time point { z
k(i), i=1,2 ..., m
k, in theory because comprise two lineups, so m in seismologic record
k=2, but affected by noisely may cause m
k≠ 2.For avoiding same wavelet to be extracted repeatedly because there being multiple spot to cross thresholding, the time corresponding to wave crest point only choosing given thresholding here reaches time point as seismic wavelet ripple, records the amplitude { A of each time point position seismic wavelet simultaneously
k(i), i=1,2 ..., m
k.
Step S02, data correlation.According to predicting the outcome of last time Kalman filtering iterative process (i.e. step S04), the time point extracted in step S01 is associated with each lineups in lineups set.If certain gauge point does not all associate with any lineups in lineups set or lineups set is empty, then by this point stored in free point set; If certain gauge point can associate with certain existing lineups, then by this gauge point stored in the candidate data set of these lineups.Utilize probability weighted method to calculate synthetic quantity measuring point simultaneously.
Step S03, according to the initial new lineups of 2/2 direct-vision method of data acquisition in free gauge point set, even continuous 2 reflection seismic waves velocities of wave calculated
be less than 5000m/s, then initial new lineups.Initial success adopts 2 method of difference to carry out the initialization of Kalman filter afterwards, then Initial state estimation and initial covariance matrix are respectively:
Wherein suppose measurement noise W
kbe average be 0, variance is
white Gaussian noise.
Step S04, carries out Kalman filtering iteration to lineups all in lineups set one by one.Identical with conventional Kalman Filtering Model, repeat no more here.
Step S05, lineups manage.Give a mark rule according to data correlation situation in current iteration to the quality score of each lineups according to the lineups drafted in table 2, and upgrade lineups states according to scoring, comprise may the confirmation (possible the lineups of scoring >3 transfer reliable lineups to) of lineups, the maintenance of reliable lineups, the termination of the reliable lineups of scoring <0, the cancelling of false lineups (the possible lineups of the <1 that marks).
Table 2 lineups marking rule
| Be successfully associated | Uncorrelated | Minimum score | Top score | |
| Possible lineups | 1 | -3 | 1 | 3 |
| Reliable lineups | 2 | -3 | 2 | 8 |
| Termination lineups | - | - | - | 0 |
The use-pattern of lineups marking rule is:
When new one possibility lineups, its marking is 1, and the gauge point often associating a upper track data all adds 1 point, but once have a track data not associated on, then directly subtract 3 points, as the scoring >3 of possibility lineups, possible lineups transfer reliable lineups to, therefore possibility lineups is so minimum that to be divided into 1, and top score is 3.If but possibility lineups scoring <1, be considered to begin false lineups and cancelled.Therefore when may lineups have 1 point not associated upper time, these possibility lineups will be cancelled.
For reliable lineups, often associates the gauge point of a upper track data and all add 1 point, but best result is 8, but once have a track data not associated on, then directly subtract 3 points, as the scoring <0 that reliable lineups turn, reliable lineups are terminated.Therefore when reliable lineups have continuous 3 points not associated upper time, these reliable lineups will be terminated.
Step S06, makes k=k+1, repeats step S01 ~ S05, until last track data iteration is complete.The seismic event information that will pick up is contained in a series of state vectors then finally obtained.
For above-described embodiment condition and implementation step, Fig. 3 gives all seismic wavelet ripples extracted and reaches time point, Fig. 4 gives the positional information of last the lineups picked up, visible through this automatic Picking system the position of lineups of picking up be in the main true, compared with Fig. 3, eliminate false seismic wavelet point, to supplement on lineups the undetected seismic wavelet point because of noise effect, lineups continuity, slickness improve after filtering simultaneously.
In sum, these are only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (4)
1. based on a seismic reflection lineups automatic pick method for Kalman filtering, it is characterized in that, the method sets up the Kalman's system model based on the earthquake reflected wave characteristics of motion, and utilizes Kalman filtering to carry out progressive alternate pickup seismic reflection lineups;
The modeling procedure of described Kalman's system model is: set up the state equation and measurement equation that are used for Kalman filtering according to the earthquake reflected wave characteristics of motion; The state vector in a kth detection road is made to be
Wherein v
kthe average velocity that to be seismic event by shot point be transmitted to through boundary reflection in a kth wave detector process, t
kthat the ripple of a kth wave detector reception interface reflection wave reaches the time; Then state equation is write as
X
k+1＝F
kX
k+U
k， (2)
Wherein state-transition matrix
X
kit is the geophone offset of a kth wave detector; U
k=Γ
ku
kfor process noise, wherein process noise distribution matrix
U
kfor zero-mean gaussian process noise;
The measurement vector making a kth detection road is Z
k=[z
k 2], wherein z
kfor in the kth detection road that extracts from seismologic record, the ripple of each seismic wavelet reaches the time, then measurement equation is write as
Z
k+1＝H
k+1X
k+1+W
k+1， (2)
Wherein measurement matrix H
k+1=[1 0], W
k+1for the measurement noise introduced in above-mentioned measurement equation;
The described step utilizing Kalman filtering to carry out progressive alternate pickup seismic reflection lineups comprises:
Step S00, initialization; Put free gauge point set, lineups set for empty, iterative steps k=1;
Step S01, extracts ripple and reaches time point;
Take out seismologic record Zhongk road seismic signal, the ripple extracting seismic wavelet in this track data reaches time point and gauge point { z
k(i), i=1,2 ..., m
k, m
krepresent the seismic wavelet number extracted in kth road seismic signal, record the amplitude { A that each ripple reaches time point position seismic wavelet simultaneously
k(i), i=1,2 ..., m
k;
Step S02, data correlation;
In upper once Kalman filtering iterative process, predict that according to step S04 the position of each lineups on the kth road seismic signal obtained and ripple reach temporal predictive value z
k ︱ k-1, the ripple extracted in step S01 is reached time point z
ki () associates with each lineups in lineups set, concrete interrelational form is:
For each gauge point z extracted
ki (), judges whether it falls into the z that predicts the outcome of any one lineups of lineups set
k ︱ k-1centered by preset time in region, if so, then this gauge point belongs to this lineups, i.e. the candidate point of lineups that falls into as it of this gauge point and being associated, by this gauge point stored in candidate data set corresponding to association lineups; If gauge point does not all associate with any lineups in lineups set, then by this gauge point stored in free point set; If lineups set is empty, now the direct all gauge points extracted by step S01 in epicycle process are stored in free point set;
Step S03, initial new lineups;
According to the initial new lineups of data acquisition M/L direct-vision method in free point set, namely according to the velocity of wave of arbitrary continuation two-point estimate reflection seismic waves in continuous L road gauge point
the span of l is 1,2 ..., L-1, if there is the velocity of wave of M point to meet given speed requirement, then these new lineups are added in lineups set by initial new lineups, and delete being used for M the gauge point that initial new lineups use from free point set; The initialization of Kalman filter corresponding to new lineups is carried out after initial success; Wherein M and L is positive integer, and M is less than or equal to L;
Step S04, carries out this Kalman filtering iteration to lineups all in lineups set one by one, obtains state estimation and the state estimation covariance matrix of current iteration, then provides the predicted value z of each lineups position needed for next iteration
k+1 ︱ k;
Step S05, makes k=k+1, repeats step S01 ~ S04, until last road seismic signal iteration is complete; Contain the seismic event information that will pick up in a series of state vectors then finally obtained, comprise the positional information t of lineups
kwith velocity information v
k.
2. the method for claim 1, is characterized in that, in step S02, by gauge point z
ki () is stored in when associating in candidate data set corresponding to lineups, the situation of same lineups is associated with if there is multiple gauge point, then each gauge point being associated with same lineups is synthesized a synthetic quantity measuring point, as the metric data participating in kth time Kalman filtering iteration.
3. the method for claim 1, it is characterized in that, between step S04 and step S05, comprise the step of lineups management further: according to the lineups rule drafted according to the data correlation situation of step S02 in epicycle iteration to the quality score of each lineups, and upgrade lineups states according to scoring, comprise the maintenance of lineups, the cancelling of termination and false lineups.
4. method as claimed in claim 3, is characterized in that, in step S03, when after the initial new lineups of employing M/L direct-vision method, new lineups is defined as possibility lineups;
Then in described lineups management process, when the mark of possibility lineups is higher than the first scoring line set, possible lineups confirm as reliable lineups; Possible lineups and reliable lineups all adopt and associate bonus point at every turn, the principle of not associated deduction carries out quality score; Cancelled when possibility lineups mark is less than the second scoring line of setting, and deleted from lineups set; Terminated when reliable lineups scoring is less than the 3rd scoring line of setting; When the scoring of reliable lineups is greater than scoring above in limited time, then do not continue upwards to add up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310065332.3A CN103149591B (en) | 2013-03-01 | 2013-03-01 | Method for automatically picking up seismic reflection event based on Kalman filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310065332.3A CN103149591B (en) | 2013-03-01 | 2013-03-01 | Method for automatically picking up seismic reflection event based on Kalman filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103149591A CN103149591A (en) | 2013-06-12 |
| CN103149591B true CN103149591B (en) | 2015-07-15 |
Family
ID=48547773
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310065332.3A Expired - Fee Related CN103149591B (en) | 2013-03-01 | 2013-03-01 | Method for automatically picking up seismic reflection event based on Kalman filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103149591B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104714251B (en) * | 2015-03-24 | 2017-02-01 | 西安交通大学 | Oblique stacking peak amplitude portion edge detecting method for event automatic picking |
| JP6980199B2 (en) * | 2020-02-21 | 2021-12-15 | 株式会社東京測振 | Estimator, vibration sensor system, method performed by the estimator, and program |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5173880A (en) * | 1989-12-26 | 1992-12-22 | Exxon Production Research Company | Method of generating seismic wavelets using seismic range equation |
-
2013
- 2013-03-01 CN CN201310065332.3A patent/CN103149591B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5173880A (en) * | 1989-12-26 | 1992-12-22 | Exxon Production Research Company | Method of generating seismic wavelets using seismic range equation |
Non-Patent Citations (3)
| Title |
|---|
| 地震剖面图同相轴的AR自动追踪方法;周冠雄 胡志成;《自动化学报》;19910531;第17卷(第3期);359-362 * |
| 基于能量比与互相关法的地震剖面反射同相轴交互自动拾取研究;丁维凤 等;《海洋学报》;20120531;第34卷(第3期);87-91 * |
| 自动拾取同相轴系统;P.TU* et al.陈炳文 译;《美国勘探地球物理学家学会第63届年会论文集》;19930930;127-130 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103149591A (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102227650B (en) | Continuous adaptive surface wave analysis for three-dimensional seismic data | |
| CN101980054B (en) | Method for establishing near-surface velocity model in high-density seismic static correction processing | |
| CN101630016B (en) | Method for improving imaging quality of vertical seismic profile | |
| CN106646598B (en) | A kind of FAST-AIC methods microseism picking up signal method | |
| CN102313901B (en) | Iteration acquisition method for first arrival wave | |
| CN104714225B (en) | Dynamic programming tracking-before-detection method based on generalized likelihood ratios | |
| CN105512635A (en) | Category attribute fused deep network underground target identification method and system | |
| CN101937099B (en) | The method that the dynamic aperture of the Multiple attenuation of being correlated with for three-dimensional surface is determined | |
| CN104335072A (en) | Integrated passive and active seismic surveying using multiple arrays | |
| CN108196305B (en) | Mountain land static correction method | |
| CN102508293A (en) | Pre-stack inversion thin layer oil/gas-bearing possibility identifying method | |
| CN109001801B (en) | Fault variable-scale identification method based on multiple iteration ant colony algorithm | |
| CN104166161A (en) | Method and device for predicating fractures based on elliptical velocity inversion of anisotropism | |
| CN102073067A (en) | Method for increasing automatic pickup efficiency of geological data first-motion waves | |
| CN104570076A (en) | Automatic seismic wave first-arrival picking method based on dichotomy | |
| CN104765064A (en) | Microseism interference imaging method | |
| CN102692648A (en) | Refraction wave residual static correction method based on cannon first arrival | |
| CN106468771A (en) | A kind of multi-target detection and tracking method under the high clutter conditions of low Observable | |
| CN105259570A (en) | Seaborne time shifting earthquake receiving point displacement correction method | |
| CN103576200A (en) | Low signal-to-noise ratio zone shallow wave impedance interface static correction method | |
| CN115905805A (en) | DAS data multi-scale noise reduction method based on global information judgment GAN | |
| CN103576191A (en) | Method for identifying fault through seismic attributes | |
| CN103149591B (en) | Method for automatically picking up seismic reflection event based on Kalman filtering | |
| CN105474048A (en) | Predicting interbed multiples in seismic data using beam decomposition | |
| US6980482B2 (en) | Seismic survey system |
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: 20150715 Termination date: 20160301 |