CN105825487A - Full well periphery electric imaging image generation method and system - Google Patents
Full well periphery electric imaging image generation method and system Download PDFInfo
- Publication number
- CN105825487A CN105825487A CN201610210700.2A CN201610210700A CN105825487A CN 105825487 A CN105825487 A CN 105825487A CN 201610210700 A CN201610210700 A CN 201610210700A CN 105825487 A CN105825487 A CN 105825487A
- Authority
- CN
- China
- Prior art keywords
- image
- repaired
- training
- pattern
- interpolation
- 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.)
- Pending
Links
- 238000003384 imaging method Methods 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 61
- 238000005457 optimization Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 238000012896 Statistical algorithm Methods 0.000 abstract 1
- 230000008439 repair process Effects 0.000 description 11
- 238000001914 filtration Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000001308 synthesis method Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a full well periphery electric imaging image generation method and system. The method comprises the following steps: loading electric imaging logging data obtained in advance, preprocessing the electric imaging logging data, and obtaining an image to be restored, wherein the image to be restored comprises a blank image to be restored; performing interpolation on the blank image to be restored by use of a preset interpolation algorithm and obtaining the image to be restored after the interpolation; defining the blank image to be restored before the interpolation as a first image in the image to be restored after the interpolation; and optimizing the first image by use of a preset statistical algorithm so as to obtain a final electric imaging image. Through the scheme provided by the invention, the consistency of the overall image after the blank image is restored can be ensured, no obvious noise is generated, the texture feature of the image is also maintained, and data after restoration is more reliable.
Description
Technical field
The present invention relates to well logging field, particularly relate to one full well week Electrical imaging image generating method and system.
Background technology
Electric imaging logging can obtain well week two dimensional image, can the most intuitively and clearly reflect the structure and features of the borehole wall.Utilize well logging image viewability and intuitive, the insoluble geological problem of conventional logging can be solved.But due to the reason on casing programme and electric imaging logging instrument structure, when measuring, instrument is in open configuration, cause when scanning along the borehole wall, having the part borehole wall to fail to measure, coverage rate can not reach 100%, there is blank image between each pole plate, electrical log picture produces white ribbon, relative to sound wave full well week image, quantity of information is relatively fewer, is unfavorable for the identification of follow-up image procossing and geological phenomenon.
Full well week Electrical imaging image generation belongs to image repair, it is exactly that information defect area on image is carried out the process of information filling, the image that its purpose is contemplated to there being information defect is repaired, more data more reliably, the inaccurate problem of explanation conclusion that minimizing factor data lacks and causes is provided for follow-up works such as explanations based on Electrical imaging image.At present, image repair method totally can be divided into restorative procedure based on structure and restorative procedure based on texture.Reparation algorithm based on structure is mainly interpolation algorithm, and its principle is simple, processes quickly, can well repair line structure and the objective contour of image, but deficiency is significantly to process vestige and make overall visual effect poor because existing, and the grain details of layer irreparably.Restorative procedure based on texture mainly includes textures synthesis and multiple spot Geo-statistic Method.The grain details of texture synthesis method energy well renwing stratum, due to the feature of algorithm own, especially when repairing window and being bigger, computationally intensive, speed is slow, can produce again many noises simultaneously, and repairing effect is the best；And method of based on multiple spot geological statistics, not only reconstruction effect is preferable, and when repairing window and being bigger, speed, but many noises can be produced equally.
Summary of the invention
In order to solve the problems referred to above, the present invention proposes one full well week Electrical imaging image generating method and system, it is possible to i.e. ensures that blank image is repaired the concordance of rear general image, will not produce obvious noise, maintain again the textural characteristics of image, make the data after reparation relatively reliable.
In order to achieve the above object, the present invention proposes a kind of well week Electrical imaging image generating method entirely, and the method includes:
Load the electric imaging logging data being obtained ahead of time, electric imaging logging data is carried out pretreatment, obtains image to be repaired；Wherein, image to be repaired includes: blank image to be repaired.
Use the interpolation algorithm preset that blank image to be repaired carries out interpolation, obtain the image to be repaired after interpolation；And the first image blank image to be repaired before interpolation being defined as in the image to be repaired after interpolation.
Use the statistic algorithm preset that the first image is optimized, obtain final Electrical imaging image.
Preferably,
Image to be repaired also includes: pole plate image.
Use the statistic algorithm preset that the first image is optimized, obtain final Electrical imaging image and include:
Pole plate image in image to be repaired is carried out image training and obtains one or more pattern models.
One or more contrast patterns of the first image are obtained according to one or more pattern models.
Calculate the distance of the node to be repaired of each on the first image and one or more contrast patterns according to default distance algorithm respectively, obtain the optimum pattern model of node each to be repaired on the first image.
Optimum pattern model is pasted on the node to be repaired corresponding with optimum pattern model.
Using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as final Electrical imaging image.
Preferably, the pole plate image in image to be repaired carries out the image training one or more pattern models of acquisition include:
Using the pole plate image in image to be repaired as training region.
Set training window and the parameter of training window is set.
The training window setting up parameter is moved on training region, obtains one or more different training pattern.
Obtaining n filter value of each training pattern, n is positive integer.
According to the gradient on training n direction of pattern and curvature, n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern.
Obtain gradient and the curvature of every height training pattern respectively.
By the error amount between gradient less than or equal to the first error threshold preset, and the error amount between curvature is less than or equal to the one or more sub-training pattern of the second default error threshold as a pattern model.
Preferably, obtain one or more contrast patterns of the first image according to one or more pattern models to include:
Calculate the average gray of one or more sub-training pattern in each pattern model；Using there is different average gray pattern model as a contrast patterns.
Preferably,
The interpolation algorithm preset is: inverse distance weighted interpolation algorithm.
The statistic algorithm preset is: multiple spot geological statistics algorithm.
In order to achieve the above object, the invention allows for a kind of well week Electrical imaging image generation system entirely, this system includes: pretreatment module, interpolating module and optimization module.
Pretreatment module, for loading the electric imaging logging data being obtained ahead of time, carries out pretreatment to this electric imaging logging data, obtains image to be repaired；Wherein, image to be repaired includes: blank image to be repaired.
Interpolating module, for using default interpolation algorithm that blank image carries out interpolation, obtains the image to be repaired after interpolation；And the first image blank image to be repaired before interpolation being defined as in the image to be repaired after interpolation.
Optimize module, for using default statistic algorithm that the first image is optimized, obtain final Electrical imaging image.
Preferably,
Image to be repaired also includes: pole plate image.
Optimizing module uses the statistic algorithm preset to be optimized the first image, obtains final Electrical imaging image and includes:
Pole plate image in image to be repaired is carried out image training and obtains one or more pattern models.
One or more contrast patterns of the first image are obtained according to one or more pattern models.
Calculate the distance of the node to be repaired of each on the first image and one or more contrast patterns according to default distance algorithm respectively, obtain the optimum pattern model of node each to be repaired on the first image.
Optimum pattern model is pasted on the node to be repaired corresponding with optimum pattern model.
Using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as final Electrical imaging image.
Preferably, optimize module the pole plate image in image to be repaired is carried out image training obtain one or more pattern models include:
Using the pole plate image in image to be repaired as training region.
Set training window and the parameter of training window is set.
The training window setting up parameter is moved on training region, obtains one or more different training pattern.
Obtaining n filter value of each training pattern, n is positive integer.
According to the gradient on training n direction of pattern and curvature, n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern.
Obtain gradient and the curvature of every height training pattern respectively.
By the error amount between gradient less than or equal to the first error threshold preset, and the error amount between curvature is less than or equal to the one or more sub-training pattern of the second default error threshold as a pattern model.
Preferably, one or more contrast patterns that optimization module obtains the first image according to one or more pattern models include:
Calculate the average gray of one or more sub-training pattern in each pattern model；Using there is different average gray pattern model as a contrast patterns.
Preferably,
The interpolation algorithm preset is: inverse distance weighted interpolation algorithm.
The statistic algorithm preset is: multiple spot geological statistics algorithm.
Compared with prior art, the present invention includes: loads the electric imaging logging data being obtained ahead of time, electric imaging logging data is carried out pretreatment, obtains image to be repaired；Wherein, this image to be repaired includes: blank image to be repaired.Image to be repaired after using the interpolation algorithm preset blank image to be repaired to carry out interpolation and obtains interpolation；And the first image blank image to be repaired before interpolation being defined as in the image to be repaired after interpolation.Use the statistic algorithm preset that the first image is optimized, obtain final Electrical imaging image.Pass through the solution of the present invention, it is possible to i.e. ensure that blank image is repaired the concordance of rear general image, obvious noise will not be produced, maintain again the textural characteristics of image, make the data after reparation relatively reliable.
Accompanying drawing explanation
Illustrating the accompanying drawing in the embodiment of the present invention below, the accompanying drawing in embodiment is for a further understanding of the present invention, is used for explaining the present invention, is not intended that limiting the scope of the invention together with description.
Fig. 1 is full well week Electrical imaging image generating method flow chart of the present invention；
Fig. 2 is full well week Electrical imaging image generation system composition frame chart of the present invention.
Detailed description of the invention
For the ease of the understanding of those skilled in the art, the invention will be further described below in conjunction with the accompanying drawings, can not be used for limiting the scope of the invention.
In order to overcome the deficiency of existing monotechnics, it is an object of the invention to provide a kind of full well week Electrical imaging image generating method being combined based on multiple spot geological statistics with interpolation, it is provided that full well week exact image of electric imaging logging image.
The present invention is based on EGPS log interpretation software platform, when electric imaging logging image is repaired, by interpolation method be combined based on Multiple-Point Geostatistics restorative procedure, interpolation algorithm can ensure that white space (i.e. blank image) is repaired the concordance of rear general image, obvious noise will not be produced, but cannot keep the textural characteristics of image, being repaired region has obvious repairing mark.Multiple-Point Geostatistics restorative procedure can be good at keeping the textural characteristics of image, but when the white space of repair data disappearance, easily produce noise, affect the accuracy of data.Both approaches being effectively combined, the region that just can reach to be repaired i.e. will not produce obvious noise, maintains again the textural characteristics of image, makes the data after reparation relatively reliable.
In order to achieve the above object, the present invention proposes a kind of well week Electrical imaging image generating method entirely, as it is shown in figure 1, the method includes:
The electric imaging logging data that S101, loading are obtained ahead of time, carries out pretreatment to electric imaging logging data, obtains image to be repaired；Wherein, image to be repaired includes: blank image to be repaired.
In embodiments of the present invention, need first to load the electric imaging logging data being obtained ahead of time, carry out electric imaging logging pre-processing of the information.This pretreatment includes pole plate alignment and equalization.Generating image to be repaired through pretreated electric imaging logging data, this image to be repaired is dynamic image or still image.This dynamic image or still image include pole plate image-region and blank image (that is, the white space on image to be repaired, this white space is part to be repaired on image), and the purpose of patent of the present invention carries out image repair to this white space exactly.
The interpolation algorithm that S102, employing are preset carries out interpolation to blank image to be repaired, obtains the image to be repaired after interpolation；And the first image blank image described to be repaired before interpolation being defined as in the image to be repaired after described interpolation.
Preferably, the interpolation algorithm preset is: inverse distance weighted interpolation algorithm.
Inverse distance weighted interpolation InverseDistancetoaPower, it is also possible to be referred to as inverse distance and take advantage of method.
The method of inverse distance power gridding is an interpolation by weighted average method, can by carry out definite or round and smooth in the way of interpolation.Degree state modulator how weight coefficient declines along with the increase leaving a grid node distance.For a bigger degree, the given higher weight share of nearer data point, for a less degree, weight ratio relatively evenly distributes to each data point.
When calculating a grid node, giving the weights of a particular data point and the appointment degree from node to observation station, this node is proportional to inverse distance.When calculating a grid node, the weight of dispensing is a mark, and the summation of all weights is equal to 1.0.When an observation station overlaps with a grid node, this observation station be given one actual be 1.0 weight, other observation stations all are given the weight of almost 0.0.In other words, this node is assigned to the value consistent with observation station.Here it is an accurate interpolation.
In embodiments of the present invention, the first image is obtained after using the interpolation algorithm preset that blank image is carried out interpolation, at each interpolation point on first image, there is certain gray value, the image to be repaired with this gray value is entered as the image to be repaired after interpolation the optimization process of step S103.
Wherein, the process using inverse distance weighted interpolation algorithm that blank image carries out interpolation is exactly the process obtaining the gray value at each interpolation point, including:
Obtain each data point in pole plate image-region to the distance of interpolation point and inverse distance.
Respectively to the distance of each data point to interpolation point divided by the sum of whole described inverse distances, it is thus achieved that the weight coefficient of each data point.
Using the gray value of each data point in described pole plate image-region and the sum of products of described weight coefficient as the gray value at described interpolation point.
First image is optimized by the statistic algorithm that S103, employing are preset, and obtains final Electrical imaging image.
Preferably, the statistic algorithm preset is: multiple spot geological statistics algorithm.
In embodiments of the present invention, in order to improve the reparation image effect produced in step S102, utilize multiple spot Geo-statistic Method to optimize image, be allowed to more mate with actual formation.
Preferably, image to be repaired also includes: pole plate image.
Preferably, use the statistic algorithm preset that the first image is optimized, obtain final Electrical imaging image and include:
S201, the pole plate image in image to be repaired is carried out image training obtain one or more pattern models.
Preferably, the pole plate image in image to be repaired carries out the image training one or more pattern models of acquisition include:
S301, using the pole plate image in image to be repaired as training region.
S302, setting training window also arrange the parameter training window.
In embodiments of the present invention, this training window is the window of an x*x, and this x is positive integer.
S303, move setting up the training window of parameter on training region, obtain one or more different training pattern.
S304, obtaining n filter value of each training pattern, n is positive integer.
In embodiments of the present invention, the wave filter of given multiple (such as 6) different directions, acting on training pattern, training image carries out level and the multiple filtering on vertical two kinds of directions respectively, this multiple filtering includes filtering, first derivative filtering, second dervative filtering.Obtain the multiple filter value including meansigma methods, gradient or curvature, i.e. n the filter value (such as 6) of the present invention.These filtering numerical value embody training pattern gradient on given multiple (such as 6) direction and curvature feature, the excursion the biggest explanation texture variations of gradient and curvature the feature instantiation texture complexity degree of image, gradient and curvature is the biggest.
Illustrate with gradient and curvature for embodiment below.
S305, according to training n direction of pattern on gradient and curvature, n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern.
S306, the gradient obtaining every height training pattern respectively and curvature.
S307, by the error amount between gradient less than or equal to the first error threshold preset, and the error amount between curvature less than or equal to the one or more sub-training pattern of the second error threshold preset as a pattern model.The son training pattern of multiple gradient and curvature feature similarity is had in each pattern model.
In embodiments of the present invention, this first error threshold and the second error threshold can be arranged by the most different application scenarios voluntarily.
S202, obtain one or more contrast patterns of the first image according to one or more pattern models.
Preferably, obtain one or more contrast patterns of the first image according to one or more pattern models to include:
Calculate the average gray of one or more sub-training pattern in each pattern model；Using there is different average gray pattern model as a contrast patterns.
S203, calculate the distance of the node to be repaired of each on the first image and one or more contrast patterns respectively according to the distance algorithm preset, obtain the optimum pattern model of node each to be repaired on the first image.
Here, the distance calculating the node to be repaired of each on the first image and one or more contrast patterns refers to calculate the Centroid distance of the node to be repaired of each on the first image and one or more contrast patterns.Need exist for explanation be, the in the same size of pattern is trained with above-mentioned son owing to repairing size of node, here, each height training pattern comprised in this contrast patterns is trained pattern as a node, the son being in center in Centroid i.e. contrast patterns by each contrast patterns respectively.
The optimum pattern model obtaining the node each to be repaired on the first image includes:
The over the ground pattern minimum with the distance of a node to be repaired is repaired as this band the optimum contrast patterns of node, obtain the son training pattern of the distance minimum repairing node in this optimum contrast patterns with this band, this sub-training pattern will be obtained as the optimum pattern model repairing node.
In embodiments of the present invention, just complete, by step S201 (including that step S301 is to step S307), the process being modeled in the embodiment of the present invention according to substrate image to step S202.
The first image described in this step be exactly step S102 obtains after inverse distance weighted interpolation algorithm carries out interpolation, there is the former blank image of certain gray value.After completing above-mentioned modeling, needing to arrange one or more reparation node on this first image, respectively each is repaired node and be optimized, wherein, this reparation size of node is in the same size with above-mentioned son training pattern.
Here, the distance algorithm preset can be below equation: (illustrating as a example by the Centroid of the first image here)
In formula, u0It it is the position of contrast patterns Centroid；N is the number of all nodes in contrast patterns；J is the position of present node in contrast patterns；ωjIt it is the weight of each node in contrast patterns；U is the Centroid of data event；hjIt it is the offset distance in contrast patterns.
S204, optimum pattern model is pasted on the node to be repaired corresponding with optimum pattern model.
In embodiments of the present invention, having been illustrated by step S203, here, paste on node to be repaired is to repair with this band to repair the closest son training region of node with this band in the closest optimum contrast patterns of node.
S205, using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as final Electrical imaging image.
In embodiments of the present invention, the substrate image on image to be repaired not being changed, by above step, each node to be repaired is pasted, until pasting complete the first image, just completing the full well week Electrical imaging image repair generation scheme of the present invention.
In order to achieve the above object, the invention allows for one full well week Electrical imaging image generation system 01, as in figure 2 it is shown, this system includes: pretreatment module 02, interpolating module 03 and optimization module 04.
Pretreatment module 02, for loading the electric imaging logging data being obtained ahead of time, carries out pretreatment to this electric imaging logging data, obtains image to be repaired；Wherein, image to be repaired includes: blank image to be repaired.
Interpolating module 03, the image to be repaired after using default interpolation algorithm blank image to carry out interpolation and obtains interpolation；And the first image blank image to be repaired before interpolation being defined as in the image to be repaired after interpolation.
Optimize module 04, for using default statistic algorithm that the first image is optimized, obtain final Electrical imaging image.
Preferably, image to be repaired also includes: pole plate image.
Optimizing module 04 uses the statistic algorithm preset to be optimized the first image, obtains final Electrical imaging image and includes:
Pole plate image in image to be repaired is carried out image training and obtains one or more pattern models.
One or more contrast patterns of the first image are obtained according to one or more pattern models.
Calculate the distance of the node to be repaired of each on the first image and one or more contrast patterns according to default distance algorithm respectively, obtain the optimum pattern model of node each to be repaired on the first image.
Optimum pattern model is pasted on the node to be repaired corresponding with optimum pattern model.
Using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as final Electrical imaging image.
Preferably, optimize module 04 the pole plate image in image to be repaired is carried out image training obtain one or more pattern models include:
Using the pole plate image in image to be repaired as training region.
Set training window and the parameter of training window is set.
The training window setting up parameter is moved on training region, obtains one or more different training pattern.
Obtaining n filter value of each training pattern, n is positive integer.
According to the gradient on training n direction of pattern and curvature, n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern.
Obtain gradient and the curvature of every height training pattern respectively.
By the error amount between gradient less than or equal to the first error threshold preset, and the error amount between curvature is less than or equal to the one or more sub-training pattern of the second default error threshold as a pattern model.
Preferably, one or more contrast patterns that optimization module 04 obtains the first image according to one or more pattern models include:
Calculate the average gray of one or more sub-training pattern in each pattern model；Using there is different average gray pattern model as a contrast patterns.
Preferably,
The interpolation algorithm preset is: inverse distance weighted interpolation algorithm.
The statistic algorithm preset is: multiple spot geological statistics algorithm.Compared with prior art, the present invention includes: loads the electric imaging logging data being obtained ahead of time, electric imaging logging data is carried out pretreatment, obtains image to be repaired；Wherein, this image to be repaired includes: blank image to be repaired.Use the interpolation algorithm preset that blank image to be repaired carries out interpolation, obtain the image to be repaired after interpolation；And the first image blank image to be repaired before interpolation being defined as in the image to be repaired after interpolation.Use the statistic algorithm preset that the first image is optimized, obtain final Electrical imaging image.Pass through the solution of the present invention, it is possible to i.e. ensure that blank image is repaired the concordance of rear general image, obvious noise will not be produced, maintain again the textural characteristics of image, make the data after reparation relatively reliable.
The present invention has the advantages that compared with prior art
Relative to based on multiple spot geological statistics restorative procedure, the method makes full use of the fireballing feature of interpolation method, quickly repairs electric imaging logging white space, it is utilized tentatively to repair result at white space, it is filtered, mates in filtering territory, searching modes match block.Due in heterogeneity than more serious white space, make use of preliminary interpolation result to be filtered, according to filter value find optimal mode class, i.e. during optimum pattern model, it is easier to find correct pattern, decreases the probability that abnormal patterns occurs, make reparation reliable results.Therefore, this restorative procedure has the advantages that speed is fast, precision is high, highly reliable.Solve complex stratum condition hypograph and repair problem.
It should be noted that; embodiment described above is for only for ease of those skilled in the art and understands; it is not limited to protection scope of the present invention; on the premise of without departing from the inventive concept of the present invention, any obvious replacement that the present invention is made by those skilled in the art and improvement etc. are all within protection scope of the present invention.
Claims (10)
1. one kind full well week Electrical imaging image generating method, it is characterised in that described method includes:
Load the electric imaging logging data being obtained ahead of time, described electric imaging logging data is carried out pretreatment, obtains image to be repaired；Wherein, described image to be repaired includes: blank image to be repaired；
Use the interpolation algorithm preset that described blank image to be repaired carries out interpolation, obtain the image to be repaired after interpolation；And the first image blank image described to be repaired before interpolation being defined as in the image to be repaired after described interpolation；
Use the statistic algorithm preset that described first image is optimized, obtain final Electrical imaging image.
2. well week Electrical imaging image generating method entirely as claimed in claim 1, it is characterised in that
Described image to be repaired also includes: pole plate image；
Described first image is optimized by the statistic algorithm that described employing is preset, and obtains final Electrical imaging image and includes:
Pole plate image in described image to be repaired is carried out image training and obtains one or more pattern models；
One or more contrast patterns of described first image are obtained according to the one or more pattern model；
Calculate the distance of each node to be repaired on described first image and the one or more contrast patterns according to default distance algorithm respectively, obtain the optimum pattern model of node each to be repaired on described first image；
Described optimum pattern model is pasted on the node to be repaired corresponding with described optimum pattern model；
Using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as described final Electrical imaging image.
3. well week Electrical imaging image generating method entirely as claimed in claim 2, it is characterised in that the described image training one or more pattern models of acquisition that carry out the pole plate image in described image to be repaired include:
Using the pole plate image in described image to be repaired as training region；
Set training window and the parameter of described training window is set；
The training window setting up described parameter is moved on described training region, obtains one or more different training pattern；
Obtaining n filter value of each training pattern, n is positive integer；
According to the gradient on described n direction of training pattern and curvature, described n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern；
Obtain gradient and the curvature of each described sub-training pattern respectively；
By the error amount between described gradient less than or equal to the first error threshold preset, and the error amount between described curvature is less than or equal to the one or more sub-training pattern of the second default error threshold as a pattern model.
4. well week Electrical imaging image generating method entirely as claimed in claim 3, it is characterised in that described one or more contrast patterns according to the one or more pattern model described blank image to be repaired of acquisition include:
Calculate the average gray of one or more sub-training pattern in each described pattern model；Using there is different average gray pattern model as a contrast patterns.
5. well week Electrical imaging image generating method entirely as claimed in claim 1, it is characterised in that
Described default interpolation algorithm is: inverse distance weighted interpolation algorithm；
Described default statistic algorithm is: multiple spot geological statistics algorithm.
6. one kind full well week Electrical imaging image generation system, it is characterised in that described system includes: pretreatment module, interpolating module and optimization module；
Described pretreatment module, for loading the electric imaging logging data being obtained ahead of time, carries out pretreatment to described electric imaging logging data, obtains image to be repaired；Wherein, described image to be repaired includes: blank image to be repaired；
Described interpolating module, for using default interpolation algorithm that described blank image carries out interpolation, obtains the image to be repaired after interpolation；And the first image blank image described to be repaired before interpolation being defined as in the image to be repaired after described interpolation；
Described optimization module, for using default statistic algorithm to be optimized described first image, obtains final Electrical imaging image.
7. well week Electrical imaging image generation system entirely as claimed in claim 6, it is characterised in that
Described image to be repaired also includes: pole plate image；
Described optimization module uses the statistic algorithm preset to be optimized described first image, obtains final Electrical imaging image and includes:
Pole plate image in described image to be repaired is carried out image training and obtains one or more pattern models；
One or more contrast patterns of described first image are obtained according to the one or more pattern model；
Calculate the distance of each node to be repaired on described first image and the one or more contrast patterns according to default distance algorithm respectively, obtain the optimum pattern model of node each to be repaired on described first image；
Described optimum pattern model is pasted on the node to be repaired corresponding with described optimum pattern model；
Using the image to be repaired of optimum pattern model that is pasted with on each node to be repaired corresponding thereto as described final Electrical imaging image.
8. well week Electrical imaging image generation system entirely as claimed in claim 7, it is characterised in that described optimization module carries out the image training one or more pattern models of acquisition and includes the pole plate image in described image to be repaired:
Using the pole plate image in described image to be repaired as training region；
Set training window and the parameter of described training window is set；
The training window setting up described parameter is moved on described training region, obtains one or more different training pattern；
Obtaining n filter value of each training pattern, n is positive integer；
According to the gradient on described n direction of training pattern and curvature, described n filter value is respectively classified into M decile, obtains the M of each training pattern6Individual unit, using each unit as a son training pattern；
Obtain gradient and the curvature of each described sub-training pattern respectively；
By the error amount between described gradient less than or equal to the first error threshold preset, and the error amount between described curvature is less than or equal to the one or more sub-training pattern of the second default error threshold as a pattern model.
9. well week Electrical imaging image generation system entirely as claimed in claim 8, it is characterised in that one or more contrast patterns that described optimization module obtains described blank image to be repaired according to the one or more pattern model include:
Calculate the average gray of one or more sub-training pattern in each described pattern model；Using there is different average gray pattern model as a contrast patterns.
10. well week Electrical imaging image generation system entirely as claimed in claim 6, it is characterised in that
Described default interpolation algorithm is: inverse distance weighted interpolation algorithm；
Described default statistic algorithm is: multiple spot geological statistics algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610210700.2A CN105825487A (en) | 2016-04-06 | 2016-04-06 | Full well periphery electric imaging image generation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610210700.2A CN105825487A (en) | 2016-04-06 | 2016-04-06 | Full well periphery electric imaging image generation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105825487A true CN105825487A (en) | 2016-08-03 |
Family
ID=56526747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610210700.2A Pending CN105825487A (en) | 2016-04-06 | 2016-04-06 | Full well periphery electric imaging image generation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105825487A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600535A (en) * | 2016-12-14 | 2017-04-26 | 中国科学院地质与地球物理研究所兰州油气资源研究中心 | Fullhole logging image generation method |
| CN108022209A (en) * | 2016-10-31 | 2018-05-11 | 北京东软医疗设备有限公司 | Acquisition methods and device, the method and apparatus of spatial compound imaging of color value |
| CN108734682A (en) * | 2017-04-20 | 2018-11-02 | 中国石油集团长城钻探工程有限公司 | A kind of adaptive restorative procedure of electric imaging logging image defect |
| CN109372497A (en) * | 2018-08-20 | 2019-02-22 | 中国石油天然气集团有限公司 | A kind of method of ultrasonic imaging dynamic equalization processing |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1687806A (en) * | 2005-06-10 | 2005-10-26 | 中油测井技术服务有限责任公司 | Full well wall restoring method for electric imaging logging map |
| CN103679650A (en) * | 2013-11-26 | 2014-03-26 | 四川大学 | Core three-dimension image repairing method |
| CN104376535A (en) * | 2014-11-04 | 2015-02-25 | 徐州工程学院 | Rapid image repairing method based on sample |
-
2016
- 2016-04-06 CN CN201610210700.2A patent/CN105825487A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1687806A (en) * | 2005-06-10 | 2005-10-26 | 中油测井技术服务有限责任公司 | Full well wall restoring method for electric imaging logging map |
| CN103679650A (en) * | 2013-11-26 | 2014-03-26 | 四川大学 | Core three-dimension image repairing method |
| CN104376535A (en) * | 2014-11-04 | 2015-02-25 | 徐州工程学院 | Rapid image repairing method based on sample |
Non-Patent Citations (2)
| Title |
|---|
| 孙建孟等: "电测井图像空白条带填充方法", 《测井技术》 * |
| 王俊华: "基于电成像测井图像处理关键技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108022209A (en) * | 2016-10-31 | 2018-05-11 | 北京东软医疗设备有限公司 | Acquisition methods and device, the method and apparatus of spatial compound imaging of color value |
| CN106600535A (en) * | 2016-12-14 | 2017-04-26 | 中国科学院地质与地球物理研究所兰州油气资源研究中心 | Fullhole logging image generation method |
| CN106600535B (en) * | 2016-12-14 | 2020-05-19 | 中国科学院地质与地球物理研究所兰州油气资源研究中心 | Whole-borehole logging image generation method |
| CN108734682A (en) * | 2017-04-20 | 2018-11-02 | 中国石油集团长城钻探工程有限公司 | A kind of adaptive restorative procedure of electric imaging logging image defect |
| CN109372497A (en) * | 2018-08-20 | 2019-02-22 | 中国石油天然气集团有限公司 | A kind of method of ultrasonic imaging dynamic equalization processing |
| CN109372497B (en) * | 2018-08-20 | 2022-03-29 | 中国石油天然气集团有限公司 | Ultrasonic imaging dynamic equalization processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105825487A (en) | Full well periphery electric imaging image generation method and system | |
| Alessandrini et al. | A comparison between the ECMWF and COSMO Ensemble Prediction Systems applied to short-term wind power forecasting on real data | |
| EP3249769B1 (en) | System and method for controlling power production from a wind farm | |
| CN104063894A (en) | Point cloud three-dimensional model reestablishing method and system | |
| CN110543649A (en) | fan arrangement method and device based on rapid evaluation fluid model and wake flow model | |
| EP3667852B1 (en) | Mesoscale data-based automatic wind turbine layout method and device | |
| CN106651007A (en) | Method and device for GRU-based medium and long-term prediction of irradiance of photovoltaic power station | |
| CN110580263B (en) | Method and system for generating wind resource data report | |
| CN102750722B (en) | The generation method and system of a kind of motion model track | |
| CN107817525B (en) | A kind of near surface constant speed fill method and device | |
| CN105718848A (en) | Quality evaluation method and apparatus of fingerprint images | |
| CN106056400A (en) | Method, apparatus and system of predicting number of new users | |
| CN115222792A (en) | Digital twin modeling method for railway bridge | |
| CN107783186A (en) | A kind of near-surface velocity modeling method and device | |
| CN109598047A (en) | A kind of phase in transformer equipment longevity prediction technique and system | |
| CN108051860B (en) | Method for generating error correction surface by utilizing trend surface form | |
| CN106443829A (en) | Method and apparatus for constructing near-surface model | |
| CN207008056U (en) | Difference cloud cluster extrapolation precipitation predicting device based on Hu squares | |
| CN102509445B (en) | Screening method and device for road condition prediction factor | |
| CN115544634A (en) | Oil and gas field level digital twin model space coordinate transformation method and system | |
| CN114004405A (en) | Photovoltaic power prediction method and system based on Elman neural network and satellite cloud picture | |
| CN110504677B (en) | Electric distance calculation method for power system | |
| CN111768045A (en) | Method for supplementing resident electricity consumption missing data by applying multiple interpolation in resident electricity consumption management | |
| Huang et al. | Probabilistic prediction intervals of wind speed based on explainable neural network | |
| CN110288124A (en) | The optimization method and device of land use pattern |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160803 |