CN104166163A - Method for automatically extracting fault curved surface based on three-dimensional large-data-volume seismic data cube - Google Patents
Method for automatically extracting fault curved surface based on three-dimensional large-data-volume seismic data cube Download PDFInfo
- Publication number
- CN104166163A CN104166163A CN201410425325.4A CN201410425325A CN104166163A CN 104166163 A CN104166163 A CN 104166163A CN 201410425325 A CN201410425325 A CN 201410425325A CN 104166163 A CN104166163 A CN 104166163A
- Authority
- CN
- China
- Prior art keywords
- point
- data
- fault
- tomography
- curved surface
- 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
Abstract
The invention discloses a method for automatically extracting a fault curved surface based on a three-dimensional large-data-volume seismic data cube. The method for automatically extracting the fault curved surface based on the three-dimensional large-data-volume seismic data cube comprises the following steps of ant body data conversion, data binaryzation, data denoising, establishment of an initial seed point queue, establishment of a connected component, determination of a fault point, solving of the fault point normal vector, division of a fault surface, secondary division of the fault surface and dimension fitting. The method for automatically extracting the fault curved surface based on the three-dimensional large-data-volume seismic data cube has the advantages that the method based on a space lattice distance is adopted, intersected fracture surfaces can be well separated through an obtained fault surface, the integrity is better, and the obtained fault surface can be better matched with a fault in original data; meanwhile, the large-data-volume processing algorithm is adopted, and therefore the method can be implemented more conveniently and more efficiently.
Description
Technical field
The invention belongs to tomography curved surface extraction method technical field, relate in particular to a kind of tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume.
Background technology
Seismic data interpretation is a very important part in geologic prospecting, the explanation of its interrupting layer one of core especially.Earth's crust rock stratum or rock mass enough can break when large External Force Acting when being subject to intensity, then along the plane of fracture, obvious relative displacement occur, and have now just formed rift structure.Earthquake Faulting is very common also very important in the earth's crust, it is divided into joint and tomography two classes conventionally, wherein earthquake fault can affect the formation and distribution of oil and natural gas etc., and exploration and the exploitation of the distribution of therefore explaining and understand fully tomography to the resource such as oil, rock gas all has very important meaning.Before the eighties in last century, geologist makes an explanation to tomography by artificial mode, utilizes the two dimensional cross-section of three-dimensional data and slice information manually to explain tomography.When the trailing of faults, along any line direction perpendicular to fault strike, or follow the trail of by line along main profile direction, then along landing surface vertical direction or horizontal direction, tomography is contrasted, extended, thereby expand in three dimensions.This method has that workload is large, the cycle is long, subjectivity is strong and result such as can not verify repeatedly at the shortcoming, and its processing procedure is numerous and diverse but also easily cause larger error.Effective utilization of the method need to explanation personnel be grasped abundant geology knowledge, and interpretation process is carried out to manual intervention frequently, and therefore this process is too dependent on relevant knowledge and the rich experience of explanation personnel about geological sciences.Along with the innovation of computer hardware technique and the develop rapidly of image processing techniques, researchist attempts image processing techniques to apply in earthquake fault explanation, the attributive analysis of three-dimensional space mesosome detected from two dimension slicing Zhu road, from manual interpretation to automatic tracing, analyzing 3-D data volume from two dimensional cross-section explains, fault recognizing technology has obtained developing on an unprecedented scale, but still exist many difficult problems and need to solve, this just makes the identification of tomography explain and remains one of seismic prospecting research field heavy difficult point.And developing rapidly along with Computerized three-dimensional image processing techniques, engender high-resolution coherence analysis technology, the precision and the efficiency that make to improve the three-dimension disclocation identification under mass seismic data become possibility, also therefore become a very important tackling key problem field.Along with making constant progress of the continuous complicated and exploration engineering of geologic prospecting, vast geology investigation and prospecting person and scientific research personnel identification and the explanation to tomographic systems conducts in-depth research, and proposed more and more convenient, more practical, more careful descriptions and the method for explaining tomography.For three-dimension disclocation identification, in nineteen ninety-five, in the 65th SEG meeting, coherent body technique is formally proposed from Bahorich M. and the Farmer S of Amoco oil company.1999, Gersztenkorn
[3]deng having proposed a kind of variation coherent body method based on covariance matrix also for the development of coherent body technique afterwards provides technical support and theoretical foundation on existing coherent body technique basis.2002, Cohen etc. proposed to use high-level data statistical approach and by more accurate and effective extracting method such as the uncontinuities of layer position, also for the fast development of the interpretation technique of three-dimension disclocation identification is afterwards laid a good foundation.Randen etc. proposed to detect by " artificial ant " mode suppressing and automatically extract the tomography in seismic volume in 2002.First the method strengthens the fault attributes in 3-d seismic data set, comprise variance attribute, dip and azimuth attribute etc., then in conjunction with " artificial ant ", it is carried out to squelch for attributive character, finally moving towards information interactive and extract tomography in conjunction with tomography.The method can reasonablely play the object of compacting noise and non-faulting response.Gibson in 2003 etc. have proposed a kind of HCF tomography automatic identifying method, the method is weighed the uncontinuity of geological data with coherent body, from coherent body, obtain some tomography fettucelles by default Seed Points and threshold value, then from tomography fettucelle, obtain final tomography curved surface by preferential (highest confidence first, HCF) the merger strategy of maximum confidence.2005, Dorn and James, Tingdahl, Pierre Jacquemin philosophy proposed to realize by the method for signal processing technology, artificial neural network technology and two Hough transformation (double hough transform) the automatic and semi-automatic identification of tomography.And then in 2006, the semi-automatic tracking of tomography is resolved floor height the mode bright and associating of driving wheel profile extractive technique and has been realized in the propositions such as Admasu.In the same year, Won-ki Jeong etc. has proposed to use interactive operation to carry out the method for fault recognizing based on GPU (Graphics Processing Unit), and the method, in the process field of mass seismic data, has very large reference.2008, Benjamin J etc. proposed again the interactive tomography curved surface computing method based on level set (level Sets).The method is extracted tomography curved surface together with level set computing method, clustering technique and three-dimensional visualization technique triplicity are arrived.Above method is followed the trail of the fault surface that obtains and can not well cross section be separated, and the tomography in rain raw data is identical preferably, between tomography, may have local adhesion phenomenon.In the time carrying out large data processing, due to the restriction of calculator memory resource, can exert an influence to the extraction of tomography curved surface.
Summary of the invention
In order to overcome the above problems, the present invention proposes a kind of tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume.
Technical scheme of the present invention is: a kind of tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume, it is characterized in that, and comprise the following steps:
S1. utilize ant group algorithm that seismic amplitude volume data is converted into ant volume data;
S2. the ant volume data in step S1 is carried out to data binaryzation, specifically comprises the following steps:
S21. within the scope of the maximal value and minimum value of ant volume data, choose an intermediate value as attribute threshold value:
S22. spatial point corresponding to data that is greater than attribute threshold value in ant volume data is set as to the point on the tomography of space, and is labeled as 1;
S23. spatial point corresponding to data that is less than attribute threshold value in ant volume data is set as to background dot or noise spot, and is labeled as 0, obtain 0,1 two-value data body;
S3. utilize and open method of operating, the two-value data body obtaining in step S2 is carried out to denoising;
S4. choose in ant volume data the point on a tomography in space lattice as Seed Points, and this Seed Points is deposited in an array, obtain an initialized Seed Points queue;
S5. search near the neighbor point of Seed Points, and neighbor point added to the Seed Points queue in step S4, obtain a connected component corresponding with proximity relations, specifically comprise the following steps:
S51. setting search step-length is 1, according to space lattice distance and proximity relations, and the neighbor point of search Seed Points;
S52. add the neighbor point in step S51 as Seed Points in the Seed Points queue in step S4, generate new Seed Points queue, obtain a connected component corresponding with proximity relations, be the point on original fault surface;
S6. repeating step S4 and S5, is included into corresponding connected component by the point on all tomographies in space, determines the point on each precursor fault face;
S7. according to the point on precursor fault face in step S6, ask on precursor fault face normal vector a little, specifically comprise the following steps:
S71. using the point on precursor fault face as original point, ask for some fields point of each original point on precursor fault face;
S72. the field point in step S71 is fitted to a characteristic face, ask for the normal vector of characteristic face;
S73. the normal vector using the normal vector of characteristic face in step S72 as original point, ask on precursor fault face normal vector a little;
S8. poor according to the inclination angle between each point on precursor fault face, taking the co-hade threshold value set as standard, each precursor fault face is divided into two mutually disjoint fault surfaces;
S9. the fault surface obtaining in step S8 is carried out to secondary and divide processing;
S10. the fault surface after secondary in step S9 being divided carries out the processing of matching structure face, realizes tomography curved surface and extracts.
Further, above-mentioned steps S3 utilizes and opens method of operating, two-value data body is carried out to denoising and specifically comprise the following steps:
S31. set set for A, structural element is B, utilizes structural element B pair set A to carry out corrosion treatment, specifically comprises the following steps:
S311. the pixel in the initial point pair set A of structural element B is contrasted one by one;
If S312. all pixels of structural element B are all included in the scope of set A, the respective pixel point of set A is retained;
If S313. all pixels of structural element B are not included in the scope of set A, the respective pixel shop of set A is given up;
S32. utilize structural element B to carry out expansion process to the corrosion treatment result in step S21, specifically comprise the following steps:
S321. structural element B is done to reflection about initial point and process, obtain structural element
S322. by the structural element in step S221
initial point and the pixel of set A contrast one by one;
If S323. structural element
in pixel without any a point in the scope of set A, respective pixel point in set A is retained;
If S324. structural element
in pixel in any one point in the scope of set A, the respective pixel point in set A is given up.
Further, above-mentioned steps S9 by fault surface carry out secondary divide process specifically comprise the following steps:
S91. choose some initial points on fault surface, form precursor fault curved surface;
S92. in determining step S91, in precursor fault curved surface, whether also has unallocated complete tomography point;
If S93. there is no unallocated complete tomography point in precursor fault curved surface, complete secondary and divide processing, algorithm finishes;
If S94. there is unallocated complete tomography point in precursor fault curved surface, search for the some neighbor points on each existing tomography to be divided;
S95. neighbor point matching in step S94 is formed to micro-plane, and calculate the offset distance of point to be divided and the micro-plane of matching;
S96. judge that point to be divided is whether in the offset distance threshold range in micro-plane;
If S97., in the offset distance threshold range in micro-plane to be divided, point to be divided is added in corresponding tomography curved surface to repeating step S92;
If S98. point to be divided, not in the offset distance threshold range in micro-plane, obtains a new tomography curved surface, repeating step S92 by tomography point structure.
The invention has the beneficial effects as follows: the present invention adopts the method based on space lattice distance, the fault surface obtaining can well separate cross section, and has more integrality, can match with the tomography in raw data preferably; Adopt large data processing algorithm simultaneously, make of the present invention realize more convenient efficient.
Brief description of the drawings
Fig. 1 is the tomography curved surface extraction method schematic flow sheet based on three-dimensional big data quantity seismic data volume of the present invention.
Fig. 2 is local adhesion schematic diagram between tomography of the present invention.
Fig. 3 is the operation chart of opening of structural element B pair set A of the present invention.
Fig. 4 is the operational processes selecting structure element schematic diagram of opening of the present invention.
Fig. 5 is the two-value data body schematic diagram after opening operational processes of the present invention.
Fig. 6 is fault branch schematic diagram of the present invention.
Fig. 7 is that fault branch of the present invention generates unusual triangle schematic diagram.
Fig. 8 is that tomography secondary of the present invention is divided processing schematic diagram.
Fig. 9 is that tomography of the present invention is divided after treatment without bifurcated tomography curved surface schematic diagram through secondary.
Figure 10 is that tomography of the present invention is divided the overall schematic diagram of the loose point of tomography curved surface after treatment through secondary.
Figure 11 is that tomography of the present invention is divided the loose point of tomography curved surface after treatment partial schematic diagram through secondary.
Figure 12 is big data quantity piecemeal Processing Algorithm schematic flow sheet of the present invention.
Figure 13 is data equal-specification piecemeal schematic diagram of the present invention.
Figure 14 is index-mapping schematic diagram of the present invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
As shown in Figure 1, be the tomography curved surface extraction method schematic flow sheet based on three-dimensional big data quantity seismic data volume of the present invention.A tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume, is characterized in that, comprises the following steps:
S1. utilize ant group algorithm that seismic amplitude volume data is converted into ant volume data.
S2. the ant volume data in step S1 is carried out to data binaryzation, specifically comprises the following steps:
S21. within the scope of the maximal value and minimum value of ant volume data, choose an intermediate value as attribute threshold value.
According to the property distribution situation of " ant body " data, within the scope of the maximal value and minimum value of " ant body " data, need as the case may be to choose an intermediate value as an attribute threshold value is set, the attribute threshold value arranging more levels off to maximal value, and final to follow the trail of the tomography point obtaining fewer.
S22. spatial point corresponding to data that is greater than attribute threshold value in ant volume data is set as to the point on the tomography of space, and is labeled as 1.
S23. spatial point corresponding to data that is less than attribute threshold value in ant volume data is set as to background dot or noise spot, and is labeled as 0, obtain 0,1 two-value data body.
S3. utilize and open method of operating, the two-value data body obtaining in step S2 is carried out to denoising, specifically comprise the following steps:
S31. set set for A, structural element is B, utilizes structural element B pair set A to carry out corrosion treatment, specifically comprises the following steps:
S311. the pixel in the initial point pair set A of structural element B is contrasted one by one;
If S312. all pixels of structural element B are all included in the scope of set A, the respective pixel point of set A is retained;
If S313. all pixels of structural element B are not included in the scope of set A, the respective pixel shop of set A is given up;
S32. utilize structural element B to carry out expansion process to the corrosion treatment result in step S21, specifically comprise the following steps:
S321. structural element B is done to reflection about initial point and process, obtain structural element
S322. by the structural element in step S221
initial point and the pixel of set A contrast one by one;
If S323. structural element
in pixel without any a point in the scope of set A, respective pixel point in set A is retained;
If S324. structural element
in pixel in any one point in the scope of set A, the respective pixel point in set A is given up.
As shown in Figure 2, be local adhesion schematic diagram between tomography of the present invention.The present invention's employing is opened operational processes and is eliminated the local adhesion phenomenon between tomography.As shown in Figure 3, be the operation chart of opening of structural element B pair set A of the present invention.Image is held to operation and generally can disconnect narrower interruption, eliminate tapering burr, and image outline is carried out smoothly.As shown in Figure 4, be the operational processes selecting structure element schematic diagram of opening of the present invention.As shown in Figure 5, be the two-value data body schematic diagram after opening operational processes of the present invention.Opening the two-value data body of operational processes front and back can find out by contrast, and operation is opened in utilization can, in the situation that keeping the original configuration feature of image, remove noise really, eliminates the local adhesion phenomenon between tomography.
S4. choose in ant volume data the point on a tomography in space lattice as Seed Points, and this Seed Points is deposited in an array, obtain an initialized Seed Points queue.
S5. search near the neighbor point of Seed Points, and neighbor point added to the Seed Points queue in step S4, obtain a connected component corresponding with proximity relations, specifically comprise the following steps:
S51. setting search step-length is 1, according to space lattice distance and proximity relations, and the neighbor point of search Seed Points.
Determine step-size in search, step-size in search is defaulted as 1, by suitable adjusting step-size in search, can make the fault surface of tracking more complete.Search is until all Seed Points all cannot be found the Seed Points making new advances within the scope of step-size in search again.
S52. add the neighbor point in step S51 as Seed Points in the Seed Points queue in step S4, generate new Seed Points queue, obtain a connected component corresponding with proximity relations, be the point on original fault surface.
S6. repeating step S4 and S5, is included into corresponding connected component by the point on all tomographies in space, determines the point on each precursor fault face.
S7. according to the point on precursor fault face in step S6, ask on precursor fault face normal vector a little, specifically comprise the following steps:
S71. using the point on precursor fault face as original point, ask for some fields point of each original point on precursor fault face.
S72. the field point in step S71 is fitted to a characteristic face, ask for the normal vector of characteristic face.
S73. the normal vector using the normal vector of characteristic face in step S72 as original point, ask on precursor fault face normal vector a little.
S8. poor according to the inclination angle between each point on precursor fault face, taking the co-hade threshold value set as standard, each precursor fault face is divided into two mutually disjoint fault surfaces.
Threshold value≤45, the inclination angle degree here.
S9. the fault surface obtaining in step S8 is carried out to secondary and divides processing, specifically comprise the following steps:
S91. choose some initial points on fault surface, form precursor fault curved surface.
S92. in determining step S91, in precursor fault curved surface, whether also has unallocated complete tomography point.
If S93. there is no unallocated complete tomography point in precursor fault curved surface, complete secondary and divide processing, algorithm finishes.
If S94. there is unallocated complete tomography point in precursor fault curved surface, search for the some neighbor points on each existing tomography to be divided.
S95. neighbor point matching in step S94 is formed to micro-plane, and calculate the offset distance of point to be divided and the micro-plane of matching.
S96. judge that point to be divided is whether in the offset distance threshold range in micro-plane.
If S97., in the offset distance threshold range in micro-plane to be divided, point to be divided is added in corresponding tomography curved surface to repeating step S92.
If S98. point to be divided, not in the offset distance threshold range in micro-plane, obtains a new tomography curved surface, repeating step S92 by tomography point structure.
As shown in Figure 6, be fault branch schematic diagram of the present invention.As shown in Figure 7, for fault branch of the present invention generates unusual triangle schematic diagram.As shown in Figure 8, divide and process schematic diagram for tomography secondary of the present invention.As shown in Figure 9, for tomography of the present invention is divided after treatment without bifurcated tomography curved surface schematic diagram through secondary.As shown in figure 10, for tomography of the present invention is divided the overall schematic diagram of the loose point of tomography curved surface after treatment through secondary.As shown in figure 11, for tomography of the present invention is divided the loose point of tomography curved surface after treatment partial schematic diagram through secondary.What the present invention carried out adopting when secondary is divided to tomography is the mode of part plan matching, and those complicated tomography curved surfaces are split into some tomography fettucelles, makes each fettucelle be approximately a space plane, separates the bifurcation problem of phantom with this.
S10. the fault surface after secondary in step S9 being divided carries out the processing of matching structure face, realizes tomography curved surface and extracts.
The present invention adopts big data quantity piecemeal Processing Algorithm.As shown in figure 12, be big data quantity piecemeal Processing Algorithm schematic flow sheet of the present invention.Whole piecemeal Processing Algorithm can be divided into two large steps, and first the first step is the blocking process to big data quantity, and then second step is exactly that the data after piecemeal are carried out to scheduling memory, specifically comprises the following steps:
Step 1. is divided into work area the little rectangular block of equal size.
The present invention adopts equal-specification dividing mode, big data quantity is carried out to the division of same size size, ensures that each the little rectangular block size of data after dividing is identical.As shown in figure 13, be data equal-specification piecemeal schematic diagram of the present invention.
Step 2. is set up identification document taking rectangular block as base unit.
Step 3. is set up rectangular block index information.
Step 4. is opened up buffer area at internal memory.
Step 5. judges whether required identification information reads in buffer memory.
If the required identification information of step 6. has read in buffer memory, directly read cache information, EO.
If the required identification information of step 7. does not read in buffer memory, judge that whether buffer memory is full;
If step 8. buffer memory less than, directly identification information piece is called in to buffer memory, read cache information, EO;
If step 9. buffer memory is full, find the rectangular block information not used at most recently;
Step 10. updating file, calls in buffer memory by information needed, replaces old rectangular block, EO.
Untapped replacement algorithm at most recently that what the present invention adopted is in the paged memory management mechanism of computer operating system is all replaced the data block not being used at most in nearest a period of time at every turn while carrying out data block displacement.In algorithm operational process, in the time need to accessing the label information of certain net point, first to the mark data block of finding this net point place in memory cache district, if found, directly access, need be set to 0 the nearest service time of accessed data block simultaneously, otherwise just corresponding mark data block need to be called in internal memory.While carrying out data block scheduling memory, first judge that whether memory cache district is full, if less than, directly desired data piece is called in internal memory, be set to 0 the nearest service time of this data block simultaneously, and certainly increased to 1 the nearest service time of other data with existing pieces in memory cache district.If buffer area is full, choose in buffer area and to be worth recently maximum data block service time, it has represented the data block not being accessed at most in nearest a period of time, by the information updating comprising in this data block after file, displaced by new data block again, be set to 0 the nearest service time of new data block simultaneously.
Those of ordinary skill in the art will appreciate that, embodiment described here is in order to help reader understanding's principle of the present invention, should be understood to that protection scope of the present invention is not limited to such special statement and embodiment.Those of ordinary skill in the art can make various other various concrete distortion and combinations that do not depart from essence of the present invention according to these technology enlightenments disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.
Claims (3)
1. the tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume, is characterized in that, comprises the following steps:
S1. utilize ant group algorithm that seismic amplitude volume data is converted into ant volume data;
S2. the ant volume data in step S1 is carried out to data binaryzation, specifically comprises the following steps:
S21. within the scope of the maximal value and minimum value of ant volume data, choose an intermediate value as attribute threshold value;
S22. spatial point corresponding to data that is greater than attribute threshold value in ant volume data is set as to the point on the tomography of space, and is labeled as 1;
S23. spatial point corresponding to data that is less than attribute threshold value in ant volume data is set as to background dot or noise spot, and is labeled as 0, obtain 0,1 two-value data body;
S3. utilize and open method of operating, the two-value data body obtaining in step S2 is carried out to denoising;
S4. choose in ant volume data the point on a tomography in space lattice as Seed Points, and this Seed Points is deposited in an array, obtain an initialized Seed Points queue;
S5. search near the neighbor point of Seed Points, and neighbor point added to the Seed Points queue in step S4, obtain a connected component corresponding with proximity relations, specifically comprise the following steps:
S51. setting search step-length is 1, according to space lattice distance and proximity relations, and the neighbor point of search Seed Points;
S52. add the neighbor point in step S51 as Seed Points in the Seed Points queue in step S4, generate new Seed Points queue, obtain a connected component corresponding with proximity relations, be the point on original fault surface;
S6. repeating step S4 and S5, is included into corresponding connected component by the point on all tomographies in space, determines the point on each precursor fault face;
S7. according to the point on precursor fault face in step S6, ask on precursor fault face normal vector a little, specifically comprise the following steps:
S71. using the point on precursor fault face as original point, ask for some fields point of each original point on precursor fault face;
S72. the field point in step S71 is fitted to a characteristic face, ask for the normal vector of characteristic face;
S73. the normal vector using the normal vector of characteristic face in step S72 as original point, ask on precursor fault face normal vector a little;
S8. poor according to the inclination angle between each point on precursor fault face, taking the co-hade threshold value set as standard, each precursor fault face is divided into two mutually disjoint fault surfaces;
S9. the fault surface obtaining in step S8 is carried out to secondary and divide processing;
S10. the fault surface after secondary in step S9 being divided carries out the processing of matching structure face, realizes tomography curved surface and extracts.
2. the tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume as claimed in claim 1, is characterized in that: described step S3 utilizes and opens method of operating, two-value data body is carried out to denoising and specifically comprise the following steps:
S31. set set for A, structural element is B, utilizes structural element B pair set A to carry out corrosion treatment, specifically comprises the following steps:
S311. the pixel in the initial point pair set A of structural element B is contrasted one by one;
If S312. all pixels of structural element B are all included in the scope of set A, the respective pixel point of set A is retained;
If S313. all pixels of structural element B are not included in the scope of set A, the respective pixel shop of set A is given up;
S32. utilize structural element B to carry out expansion process to the corrosion treatment result in step S21, specifically comprise the following steps:
S321. structural element B is done to reflection about initial point and process, obtain structural element
S322. by the structural element in step S221
initial point and the pixel of set A contrast one by one;
If S323. structural element
in pixel without any a point in the scope of set A, respective pixel point in set A is retained;
If S324. structural element
in pixel in any one point in the scope of set A, the respective pixel point in set A is given up.
3. the tomography curved surface extraction method based on three-dimensional big data quantity seismic data volume as claimed in claim 1, is characterized in that: described step S9 carries out secondary division processing by fault surface and specifically comprises the following steps:
S91. choose some initial points on fault surface, form precursor fault curved surface;
S92. in determining step S91, in precursor fault curved surface, whether also has unallocated complete tomography point;
If S93. there is no unallocated complete tomography point in precursor fault curved surface, complete secondary and divide processing, algorithm finishes;
If S94. there is unallocated complete tomography point in precursor fault curved surface, search for the some neighbor points on each existing tomography to be divided;
S95. neighbor point matching in step S94 is formed to micro-plane, and calculate the offset distance of point to be divided and the micro-plane of matching;
S96. judge that point to be divided is whether in the offset distance threshold range in micro-plane;
If S97., in the offset distance threshold range in micro-plane to be divided, point to be divided is added in corresponding tomography curved surface to repeating step S92;
If S98. point to be divided, not in the offset distance threshold range in micro-plane, obtains a new tomography curved surface, repeating step S92 by tomography point structure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410425325.4A CN104166163B (en) | 2014-08-27 | 2014-08-27 | Tomography curved surface extraction method based on three-dimensional big data quantity seismic data cube |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410425325.4A CN104166163B (en) | 2014-08-27 | 2014-08-27 | Tomography curved surface extraction method based on three-dimensional big data quantity seismic data cube |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104166163A true CN104166163A (en) | 2014-11-26 |
| CN104166163B CN104166163B (en) | 2016-08-24 |
Family
ID=51910062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410425325.4A Active CN104166163B (en) | 2014-08-27 | 2014-08-27 | Tomography curved surface extraction method based on three-dimensional big data quantity seismic data cube |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104166163B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104570096A (en) * | 2015-01-22 | 2015-04-29 | 中国海洋石油总公司 | Granite buried hill top surface fault interpretation method |
| CN105182415A (en) * | 2015-09-02 | 2015-12-23 | 中国石油化工股份有限公司 | Tomography identification method based on frequency division processing |
| CN105425296A (en) * | 2015-10-28 | 2016-03-23 | 中国石油天然气集团公司 | Geological body recognition method and apparatus |
| CN107368839A (en) * | 2017-06-22 | 2017-11-21 | 南京师范大学 | A kind of extraction method of the fault plane based on DEM |
| CN108037527A (en) * | 2017-12-07 | 2018-05-15 | 中国石油大学(华东) | A kind of complicated fault identification and detection method based on magic square operator |
| CN106199727B (en) * | 2016-08-30 | 2018-07-10 | 中国华能集团清洁能源技术研究院有限公司 | A kind of method using seismic interval velocity identification shale gas dessert |
| CN109086773A (en) * | 2018-08-29 | 2018-12-25 | 电子科技大学 | Fault plane recognition methods based on full convolutional neural networks |
| CN110174700A (en) * | 2019-05-16 | 2019-08-27 | 中海石油(中国)有限公司 | A kind of seismic properties boundary line Enhancement Method for simulating root growth |
| CN112130200A (en) * | 2020-09-23 | 2020-12-25 | 电子科技大学 | Fault identification method based on grad-CAM attention guidance |
| CN112505758A (en) * | 2020-11-17 | 2021-03-16 | 中国石油集团工程咨询有限责任公司 | Method for processing seismic fault image of complex geological structure based on fault bifurcation structure model |
| CN113267815A (en) * | 2021-07-07 | 2021-08-17 | 中海油田服务股份有限公司 | Filtering method and device for repeated broken edge data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222366A (en) * | 2011-07-29 | 2011-10-19 | 电子科技大学 | Method for fitting complex space curved surfaces |
| CN102222365A (en) * | 2011-07-29 | 2011-10-19 | 电子科技大学 | Method for reconstructing curved surface of complex space |
| CN102867330A (en) * | 2012-08-29 | 2013-01-09 | 电子科技大学 | Region-division-based spatial complex horizon reconstruction method |
| CN103412331A (en) * | 2013-08-30 | 2013-11-27 | 电子科技大学 | Automatic extraction method for three-dimensional earthquake fault |
| CN103489222A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | Target body surface reconstruction method in three-dimensional image |
-
2014
- 2014-08-27 CN CN201410425325.4A patent/CN104166163B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222366A (en) * | 2011-07-29 | 2011-10-19 | 电子科技大学 | Method for fitting complex space curved surfaces |
| CN102222365A (en) * | 2011-07-29 | 2011-10-19 | 电子科技大学 | Method for reconstructing curved surface of complex space |
| CN102867330A (en) * | 2012-08-29 | 2013-01-09 | 电子科技大学 | Region-division-based spatial complex horizon reconstruction method |
| CN103412331A (en) * | 2013-08-30 | 2013-11-27 | 电子科技大学 | Automatic extraction method for three-dimensional earthquake fault |
| CN103489222A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | Target body surface reconstruction method in three-dimensional image |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104570096A (en) * | 2015-01-22 | 2015-04-29 | 中国海洋石油总公司 | Granite buried hill top surface fault interpretation method |
| CN105182415A (en) * | 2015-09-02 | 2015-12-23 | 中国石油化工股份有限公司 | Tomography identification method based on frequency division processing |
| CN105182415B (en) * | 2015-09-02 | 2018-02-13 | 中国石油化工股份有限公司 | A kind of fault recognizing method based on scaling down processing |
| CN105425296A (en) * | 2015-10-28 | 2016-03-23 | 中国石油天然气集团公司 | Geological body recognition method and apparatus |
| CN105425296B (en) * | 2015-10-28 | 2018-04-06 | 中国石油天然气集团公司 | Geologic body recognition methods and device |
| CN106199727B (en) * | 2016-08-30 | 2018-07-10 | 中国华能集团清洁能源技术研究院有限公司 | A kind of method using seismic interval velocity identification shale gas dessert |
| CN107368839B (en) * | 2017-06-22 | 2020-05-12 | 南京师范大学 | Automatic fault plane extraction method based on DEM |
| CN107368839A (en) * | 2017-06-22 | 2017-11-21 | 南京师范大学 | A kind of extraction method of the fault plane based on DEM |
| CN108037527A (en) * | 2017-12-07 | 2018-05-15 | 中国石油大学(华东) | A kind of complicated fault identification and detection method based on magic square operator |
| CN109086773A (en) * | 2018-08-29 | 2018-12-25 | 电子科技大学 | Fault plane recognition methods based on full convolutional neural networks |
| CN110174700A (en) * | 2019-05-16 | 2019-08-27 | 中海石油(中国)有限公司 | A kind of seismic properties boundary line Enhancement Method for simulating root growth |
| CN110174700B (en) * | 2019-05-16 | 2021-01-15 | 中海石油(中国)有限公司 | Seismic attribute boundary line enhancement method for simulating root growth |
| CN112130200A (en) * | 2020-09-23 | 2020-12-25 | 电子科技大学 | Fault identification method based on grad-CAM attention guidance |
| CN112130200B (en) * | 2020-09-23 | 2021-07-20 | 电子科技大学 | Fault identification method based on grad-CAM attention guidance |
| CN112505758A (en) * | 2020-11-17 | 2021-03-16 | 中国石油集团工程咨询有限责任公司 | Method for processing seismic fault image of complex geological structure based on fault bifurcation structure model |
| CN113267815A (en) * | 2021-07-07 | 2021-08-17 | 中海油田服务股份有限公司 | Filtering method and device for repeated broken edge data |
| CN113267815B (en) * | 2021-07-07 | 2022-05-10 | 中海油田服务股份有限公司 | Method and device for filtering repeated broken edge data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104166163B (en) | 2016-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104166163A (en) | Method for automatically extracting fault curved surface based on three-dimensional large-data-volume seismic data cube | |
| Duncanson et al. | An efficient, multi-layered crown delineation algorithm for mapping individual tree structure across multiple ecosystems | |
| CN104049245B (en) | Urban building change detection method based on LiDAR point cloud spatial difference analysis | |
| CN102521273B (en) | High-resolution remote-sensing multifunctional urban land spatial information generation method | |
| CN104267431B (en) | The craven fault acquisition methods that a kind of seismotectonics is explained | |
| Zhou et al. | Building detection in Digital surface model | |
| CN109753707A (en) | A method of stratigraphic boundary, which is extracted, using section of exploration line carries out three-dimensional modeling | |
| CN103713316B (en) | A kind of speed predicting method and device based on blowhole digital representation | |
| CN105928957B (en) | A kind of construction method of fractured carbonate rock 3-dimensional digital rock core | |
| CN104570096B (en) | A kind of granite buried hill top surface fault interpretation method | |
| CN107679498A (en) | A kind of airborne laser point cloud downtown roads recognition methods | |
| CN106250895A (en) | A kind of remote sensing image region of interest area detecting method | |
| CN107271998A (en) | A kind of integrated D InSAR and the underground illegal mining recognition methods of GIS technology and system | |
| CN103530621A (en) | Coal and rock image identification method based on back propagation (BP) neural network | |
| CN109872311A (en) | A kind of Rock Mass Integrality sentences knowledge method | |
| CN101221246A (en) | Remote sensing and quantizing reconnaissance method of snowslide | |
| CN106842193A (en) | The processing method of Road Detection information, device and system | |
| CN104574512A (en) | Multi-scale DEM (digital elevation model) construction method considering topographical semantic information | |
| CN110706211A (en) | Convolutional neural network-based real-time detection method for railway roadbed disease radar map | |
| Zhou et al. | Automatic detection of lunar craters based on DEM data with the terrain analysis method | |
| CN111754623B (en) | Discrete crack modeling method based on multipoint geostatistics | |
| CN106709987B (en) | Dynamic construction method of three-dimensional geological profile model | |
| CN110672072A (en) | Information multivariate interpretation and characterization method for fractured rock mass structure | |
| CN103941287A (en) | Rapid three-dimensional fault interpretation method based on horizontal navigation | |
| Lee | Re-evaluation of factors controlling landslides triggered by the 1999 Chi–Chi earthquake |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201125 Address after: 610000 Sichuan city of Chengdu province Tianfu Tianfu Avenue South Huayang Street No. 846 Patentee after: CHENGDU AIWEI BEISI TECHNOLOGY Co.,Ltd. Address before: 611731 Chengdu province high tech Zone (West) West source Avenue, No. 2006 Patentee before: UNIVERSITY OF ELECTRONIC SCIENCE AND TECHNOLOGY |
|
| TR01 | Transfer of patent right |