CN106529104B - The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water - Google Patents
The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water Download PDFInfo
- Publication number
- CN106529104B CN106529104B CN201611236912.4A CN201611236912A CN106529104B CN 106529104 B CN106529104 B CN 106529104B CN 201611236912 A CN201611236912 A CN 201611236912A CN 106529104 B CN106529104 B CN 106529104B
- Authority
- CN
- China
- Prior art keywords
- value
- phase
- point
- phase screen
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to the phase screen emulation modes closely propagated in a kind of light under water turbulent flow, belong to Underwater Target Detection field, correlation between phase screen is not accounted in order to solve the multilayer phase screen of the prior art, the shortcomings that being unable to real simulation turbulence characteristic so as to cause phase screen model, and propose the phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water, comprising: input initial pictures；The pixel value of current base point is calculated using the datum mark of given value；Calculate generic pixel difference；The absolute value of calculating benchmark point；It repeats the above process until the left and right datum mark calculating of all phase screens finishes；Several processes are generated, each process contains 1 frame；Use the absolute value of each point in the method while Computational frame of parallel computation；The absolute value for being averaging final each point is computed repeatedly to each process；It repeats the above steps until all the points calculating finishes.The analogue simulation of propagation model of the present invention suitable for light under water turbulent flow.
Description
Technical field
The present invention relates to the phase screen emulation modes closely propagated in a kind of light under water turbulent flow, belong to submarine target spy
Survey field.
Background technique
The detection of underwater object is one of important branch of hydrospace detection.For the submarine target of movement, moving
Turbulent flow will necessarily be generated in the process.It converts the direct detection of object to the indirect detection of the field of turbulent flow of its generation, is a kind of
Important detection means.It therefore, is to carry out the main problem of the researchs such as undersea detection to the characteristic research of underwater field of turbulent flow.In benefit
When carrying out the properties study of underwater turbulent flow with laser, for light propagation, turbulent flow is on its main influence precisely due to water is situated between
The phase random fluctuation of light wave caused by the refractive index fluctuation of matter.Therefore, establishing propagation model of the light in turbulent flow is to carry out water
Lower turbulent flow research, carries out the premise and basis of Underwater Target Detection.
The algorithm model for establishing underwater turbulent flow multilayer phase screen is the main problem for carrying out underwater turbulent flow research.It carries out underwater
The research of turbulence characteristic can lay the foundation for researchs such as underwater turbulence detecting, submarine optical communications.Traditional phase screen generation side
Method has Zernike polynomial method, divides shape phase-screen method, geometrical approximation, covariance matrix predicted phase screen method.These methods are all
The characteristic of energy preferably simulation turbulent flow, but it is not suitable for the light propagation simulation of short distance.The short distance propagation model of light needs
Consider turbulent flow continuity bring multilayer phase screen between correlation, and think in conventional model be between multilayer phase screen
It is mutually independent.
Therefore in the prior art, although the light propagation algorithm model of underwater turbulent flow has very much, and it is all more mature,
In the actual conditions for considering closely to propagate, need to propose these methods amendment appropriate to meet truth.
Summary of the invention
The purpose of the present invention is to solve the multilayer phase screens of the prior art not to account for the correlation between phase screen
Property, the shortcomings that being unable to real simulation turbulence characteristic so as to cause phase screen model, and propose a kind of light low coverage in turbulent flow under water
Phase screen emulation mode from propagation.
The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water, includes the following steps:
Step 1) inputs initial pictures, and initial pictures are used to generate phase screen each other with correlation；Initial pictures
The pixel value of middle pixel is for characterizing phase value；The pixel value of each pixel and relative difference are labeled in initial pictures
For given value；The pixel of specific quantity and position in initial pictures and phase screen is chosen to be datum mark, and every a line
The datum mark of identical quantity is all had with each column；
Step 2) calculates b-th of benchmark in a-th of phase screen to be generated using all datum marks for being marked as given value
The pixel value of point, and its pixel value is labeled as given value；A is positive integer from 1 to N, and b is the positive integer from 1 to M, N be to
The phase screen number of generation, M are the number of datum mark in each phase screen；
It is pre- that step 3) using the location information of all datum marks for being marked as given value is input to three-dimensional covariance matrix
Model is surveyed, the probability distribution value of b-th of datum mark of a-th of phase screen is obtained, by probability distribution value multiplied by relevant to position
Random fluctuation value obtains generic pixel difference.And its absolute pixel differences value is labeled as given value；
Pixel value and generic pixel difference value are obtained b-th of datum mark of a-th of phase screen to be generated by step 4)
Absolute value；
B value is added 1 by step 5)；
Step 6) repeats step 2) to step 5), finishes until M datum mark of a-th of phase screen to be generated all calculates；
A value is added 1 by step 7)；
Step 8) repeats step 2 to 7, finishes until all datum marks in all N number of phase screens all calculate；
Step 9) generates a process of (K-P+1) × (K-P+1) for the 1st phase screen to be generated, and each process contains 1
A frame, each frame contain P × P pixel in the 1st phase screen to be generated；Each process is for calculating its frame contained
The absolute value of all pixels point in frame, different, the mutual relationship of frame in each process are as follows: the 1st process
The point in the frame upper left corner is the point of current the 1st row the 1st of phase screen column, and the point in the frame upper left corner of the 2nd process is current phase
The point for shielding the frame upper left corner of a process of point ... (K-P+1) × (K-P+1) of the 1st row the 2nd column is the K of current phase screen
The point of row K column.Wherein K is the number of the datum mark of phase screen a line.It include Q datum mark in each frame；
Step 10) uses the absolute value of each point in the method for parallel computation while Computational frame, specifically:
The point that any one is not labeled as given value is calculated using the datum mark for being marked as given value in frame
Absolute value, until all the points in frame all calculate completion；
After the completion of all processes of step 11) calculate, absolute value is calculated to identical point in each process and is averaging, is obtained
To the absolute value of final each point；
Step 12) repeats step 9) to step 11), until the institute in all N number of phase screens for next phase screen
It a little all calculates and finishes, and export N number of phase screen.
The invention has the benefit that being constructed using the information of all known points new when constructing phase screen model
Phase screen, so that having correlation between the phase screen formed；It is more theoretical in the analog result of turbulent flow refractive index structures function
Curve has preferable degree of conformity；Compared to other mature algorithms, the present invention can make turbulent flow have in the direction of propagation of light
There is certain correlation.
Detailed description of the invention
Fig. 1 is the flow chart for the phase screen emulation mode closely propagated in light of the invention under water turbulent flow；
Fig. 2 is the stream of the one embodiment for the phase screen emulation mode that light of the invention is closely propagated in turbulent flow under water
Cheng Tu；
Fig. 3 is showing for the one embodiment for the phase screen emulation mode that light of the invention is closely propagated in turbulent flow under water
It is intended to；
Fig. 4 is the schematic diagram of Random Incident light unit；
Fig. 5 is phase distortion situation schematic diagram when light passes through turbulent flow；
Fig. 6 is the analog result and theory for making the turbulent flow refractive index structures function for the phase screen being obtained by the present invention
The comparison diagram of curve；Star discrete point is the point that simulation result obtains in figure；Full curve in figure is the corresponding song of theoretical value
Line；
Fig. 7 is the related coefficient schematic diagram between the phase screen for the different interval being obtained by the present invention；It is horizontal in figure
Coordinate is the spacing of two phase screens；Ordinate is related coefficient in figure.
Specific embodiment
Specific embodiment 1: the phase screen emulation mode that the light of present embodiment is closely propagated in turbulent flow under water,
As shown in Figure 1, including the following steps:
Step 1) inputs initial pictures, and initial pictures are used to generate phase screen each other with correlation；Initial pictures
The pixel value of middle pixel is for characterizing phase value；The pixel value of each pixel and relative difference are labeled in initial pictures
For given value；The pixel of specific quantity and position in initial pictures and phase screen is chosen to be datum mark, and every a line
The datum mark of identical quantity is all had with each column.
Initial pictures are Fig. 3 (a), and each grid is a pixel, and initial pictures also belong to phase screen, each in figure
What three points were labeled is exactly datum mark, there is 9 × 9=81 datum mark in Fig. 3 (a).The pixel value of these pixels and opposite
Difference is marked as given value.Relative difference is amount relevant to the position of pixel, only it is to be understood that this pixel is the
The point of which column of which row in several phase screens, so that it may uniquely determine a relative difference.
Since step 2), the present invention will generate a series of phase screen, and calculate wherein pixel value of each point and
Relative difference.In each phase screen to be generated, the number of pixel and the arrangement of datum mark with initial pictures be it is identical,
Difference be only that phase screen to be generated each point pixel value and relative difference be at the beginning it is unknown, need to lead to
Method of the invention is crossed to be calculated.
Step 2) calculates b-th of benchmark in a-th of phase screen to be generated using all datum marks for being marked as given value
The pixel value of point, and its pixel value is labeled as given value；A is positive integer from 1 to N, and b is the positive integer from 1 to M, N be to
The phase screen number of generation, M are the number of datum mark in each phase screen.
Specifically, being exactly to be calculated using known all datum marks in current new new phase screen to be generated not
Know a little.For example, calculating for the first time is calculated in first phase screen to be generated with 81 datum marks in initial pictures
First (the first row first row) datum mark, the content specifically calculated are to calculate pixel value and relative difference.Point calculating finishes
Afterwards, the point being had been calculated using 81 points of initial pictures and just, totally 82 points, to calculate in first phase screen to be generated
Second (column of the first row the 5th) datum mark.And so on, each current calculative point is using all elder generations
Preceding calculated known point does so the correlation that can guarantee between phase screen come what is determined.
It is pre- that step 3) using the location information of all datum marks for being marked as given value is input to three-dimensional covariance matrix
Model is surveyed, the probability distribution value of b-th of datum mark of a-th of phase screen is obtained, by probability distribution value multiplied by relevant to position
Random fluctuation value obtains generic pixel difference.And its absolute pixel differences value is labeled as given value；
Three-dimensional covariance matrix prediction model can be indicated with following formula:
WithRepresent any two spatial position vector；x1,y1,z1ForThree coordinate components；x2,y2,z2For's
Three coordinate components；xn,yn,znFor three coordinate components of nth pixel point in current phase screen image；To removeInstitute's generation
Outside the point of table, the spatial position vector of remaining all the points；To removeOutside representative point, the space of remaining all the points
Position vector；x′1,y′1,z′1ForThree coordinate components；x′2,y′2,z′2ForCoordinate components；x′n,y′n,z′nTo work as
It is removed outside nth pixel point in preceding phase screen image, corresponding three coordinate components of the spatial position vector of remaining all the points；Dφ
For phase indirect detection；T is weighting function, and it is 0 outside region that value, which is 1, in zone of dispersion；C is phase covariance function.
Pixel value and generic pixel difference value are obtained b-th of datum mark of a-th of phase screen to be generated by step 4)
Absolute value.
B value is added 1 by step 5).
Step 6) repeats step 2) to step 5), finishes until M datum mark of a-th of phase screen to be generated all calculates.
A value is added 1 by step 7).
Step 8) repeats step 2 to 7, finishes until all datum marks in all N number of phase screens all calculate.
Step 9) generates a process of (K-P+1) × (K-P+1) for the 1st phase screen to be generated, and each process contains 1
A frame, each frame contain P × P pixel in the 1st phase screen to be generated；Each process is for calculating its frame contained
The absolute value of all pixels point in frame, different, the mutual relationship of frame in each process are as follows: the 1st process
The point in the frame upper left corner is the point of current the 1st row the 1st of phase screen column, and the point in the frame upper left corner of the 2nd process is current phase
The point for shielding the frame upper left corner of a process of point ... (K-P+1) × (K-P+1) of the 1st row the 2nd column is the K of current phase screen
The point of row K column.Wherein K is the number of the datum mark of phase screen a line.It include Q datum mark in each frame.
Specifically, step 9) calculates the pixel value in each frame using concurrent process.The schematic diagram of frame such as Fig. 3
(b) shown in, heavy black line frame is frame above-mentioned, there is 3 × 3=9 datum mark in the frame of Fig. 3 (b).The frame of Fig. 3 (b)
The frame of the 1st process is shown, the frame of second process is the frame in first process to 4 pixel lists of right translation
What position was formed.And so on.
Step 10) uses the absolute value of each point in the method for parallel computation while Computational frame, specifically:
The point that any one is not labeled as given value is calculated using the datum mark for being marked as given value in frame
Absolute value, until all the points in frame all calculate completion.
It should be noted that the calculating of step 10) refers to that all (K-P+1) × (K-P+1) formed in step 9) are same
The calculating content of this step of Shi Jinhang.Multi-process calculating in this way can greatly reduce operation time.After this step calculates, i.e.,
It indicates that all the points in the 1st phase screen to be generated calculate to finish.
After the completion of all processes of step 11) calculate, absolute value is calculated to identical point in each process and is averaging, is obtained
To the absolute value of final each point.
The point that i.e. all processes are calculated has overlapping, and step 11) is that the point that will be computed repeatedly seeks an average value.
Step 12) repeats step 9) to step 11), until the institute in all N number of phase screens for next phase screen
The N number of phase screen for a little all calculating and finishing, and export.
Above-mentioned is main process of the invention.Being briefly summarized is: meeting turbulent flow first with fractal method acquisition
The single layer two-dimensional random phase screen of statistical property.Using initial phase screen, obtained by three-dimensional covariance matrix prediction technique
Multistory frame with certain correlation.Followed by three-dimensional covariance matrix prediction technique, pass through the prediction side of this three-dimensional
The frame that method generates, the phase distribution of phase distribution and interlayer in layer can better conform to the characteristic of underwater turbulent flow.
The speed of calculating is significantly promoted on the basis of guaranteeing the accuracy of result followed by the method for parallel computation.Finally,
The result that parallel computation is obtained makees certain weighted average calculation, obtains the continuous multilayer turbulent flow phase with layer dependencies
Screen.
Fig. 2 is the stream of the one embodiment for the phase screen emulation mode that light of the invention is closely propagated in turbulent flow under water
Cheng Tu.Wherein random screen refers to phase screen, and allocation matrix refers to that datum mark arranges the matrix to be formed in each phase screen, enriches
Matrix refers to the matrix that frame and its interior pixels point are formed, and as can be seen from Figure 2 utilizes currently known datum mark meter
Calculate the substantially process of unknown datum mark and remaining point.
Mentality of designing and remarkable result of the invention is explained further below:
Using for three-dimensional covariance matrix makes when generating new phase, and the present invention is by traditional only consideration two-dimensional surface
Interior phase value has been extended to the phase value considered in three-dimensional space.And due to the expansion of dimension, so that computationally intensive amplitude mentions
Rise, therefore in algorithm optimization, need to be used to reference to known phase make necessarily required selection and reach reduction calculation amount
Purpose.In view of the reduction of fixed phase quantity, the predicted phase value that is calculated can due to sample size deficiency and
The deviation compared with theoretical value occurs, is optimized therefore, it is necessary to the selection method to fixed phase and obtains optimal scheme.
In the building process of algorithm, the selection of frame and the selection of parallel computation unit mainly influence the big of calculation amount
It is small, and the selection of weight factor then affects calculated phase values and theoretical value obtains extent of deviation.When considering the selection of weight factor,
Due to adjacent two o'clock degree of correlation with it is between the two at a distance from be inversely proportional, the inverse of distance can be taken as weight factor
To embody influence size of the different zones phase to zoning.And due to smaller apart from biggish phase effect degree, then
It, can the opposite quantity for reducing the phase apart from larger position when selecting fixed phase.Specific weight factor needs specific
Situation concrete analysis, selects suitable representation to obtain the optimal case of algorithm.It is parallel to count by adjusting the size of frame
The size of the computing unit of calculation and average weighted weight factor, the turbulence model of available difference statistical nature, for reality
Border situation needs to choose suitable parameter.
Specific embodiment 2: the present embodiment is different from the first embodiment in that: three-dimensional covariance matrix prediction
Model is used to calculate the phase fluctuation situation between any two points of space, its calculation formula is:
Phase covariance function describes the phase fluctuation situation of space any two points, and its characterization is opposite wave
Emotionally condition, it is unrelated with true phase value.
R1, r2 represent any two spatial position vector；DφFor phase indirect detection；T is weighting function, in discrete regions
Value is 1 in domain, is 0 outside region；C is phase covariance function.
Other steps and parameter are same as the specific embodiment one.
Specific embodiment 3: the present embodiment is different from the first and the second embodiment in that:
In step 2), the calculation formula of the pixel value of j-th of datum mark in i-th of phase screen to be generated are as follows:
xi,yj,zkComponent of any point on the direction x, the direction y and the direction z respectively in space；Wherein A is prediction
Matrix, φmFor known phase column matrix, σ is that interpolation criteria is poor, and R is that variance is the Gaussian random variable that 1 mean value is 0.Wherein
The form of prediction matrix A are as follows:
The form of interpolation criteria difference are as follows:
Specific derivation process are as follows:
For the random screen group of k layers of i × j size, random screen can be expressed as to the matrix of k i × j size, such as
Kth layer shields at random to be indicated are as follows:
Column vector processing is carried out to k layers of random screen, available one one column i × j × k row column matrix:
φm=(φ1,1,1φ2,1,1…φ(i-1),j,kφi,j,k)Τ
It then can use known random screen and provide the known covariance matrix shielded at random, form are as follows:
When needing to predict the phase of a new unknown phase place value position, the known covariance square shielded at random can use
Battle array carries out calculating prediction.Such as needing to obtain coordinate is (xi,yj,zk) phase value, then calculation formula are as follows:
Wherein A is prediction matrix, φmFor known phase column matrix, σ is that interpolation criteria is poor, and it is 1 mean value that R, which is variance,
For 0 Gaussian random variable.The wherein form of the form of prediction matrix and interpolation criteria difference are as follows:
Other steps and parameter are the same as one or two specific embodiments.
Specific embodiment 4: unlike one of present embodiment and specific embodiment one to three: initial pictures and
Each phase screen has 9 × 9 datum marks, and each frame has 9 × 9 pixels, and each frame has 3 × 3 datum marks.
Just as being previously mentioned, if slower, frame when will lead to very much pixel in each frame of calculating greatly that frame is chosen
It is excessive to choose the too small process for needing to generate that will lead to, consumes more system resources.Therefore the present invention, which chooses, contains 3 × 3
The frame size of datum mark can achieve optimal computational efficiency.
Other steps and parameter are identical as one of specific embodiment one to three.
Specific embodiment 5: unlike one of present embodiment and specific embodiment one to four:
The absolute value for calculating the geometric center point for the minimum rectangle that any four known point is constituted, if geometry can not be found
Central point then calculates the absolute value of all the points according to the sequence of pixel.
This process can first calculate the light point that all arrows are directed toward as shown in Fig. 3 (c), and such point has 4, this
The characteristics of a little points is the geometric center point for being all the minimum rectangle that the datum mark in frame can be constituted.And so on, until meter
The all the points in frame are calculated.Present embodiment is only to provide a kind of method that all the points comparatively fast have been calculated, the technology of this field
Personnel unambiguously determination can also can use other methods to calculate these points, such as calculate by ranks sequence, and difference is only
It is arithmetic speed.
Other steps and parameter are identical as one of specific embodiment one to four.
As shown in figure 4, the present invention regards incident light by many energy as when constructing the model that light is propagated in turbulent flow under water
Measure equal light unit composition.When light unit is propagated in the multilayer phase screen for characterizing turbulent flow statistical property, phase screen is utilized
Gradient magnitude determines the deviation angle and propagation path of light unit.When the light unit quantity of calculating reaches enough, so that it may
The light distribution and phase distribution of final emergent light are obtained with approximation, as shown in figure 5, i.e. last light wave passes through after turbulent flow
As a result.
From fig. 6 it can be seen that the present invention phase screen spacing be less than external measurement in the range of very close to theoretical curve.
As can be seen from Figure 7 there is the correlation significantly changed with phase screen spacing between the phase screen that the present invention generates
Coefficient.
The present invention can also have other various embodiments, without deviating from the spirit and substance of the present invention, this field
Technical staff makes various corresponding changes and modifications in accordance with the present invention, but these corresponding changes and modifications all should belong to
The protection scope of the appended claims of the present invention.
Claims (4)
1. the phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water, which comprises the steps of:
Step 1) inputs initial pictures, and initial pictures are used to generate phase screen each other with correlation；Picture in initial pictures
The pixel value of vegetarian refreshments is for characterizing phase value；The pixel value of each pixel and relative difference are marked as in initial pictures
Know value；The pixel of specific quantity and position in initial pictures and phase screen is chosen to be datum mark, and every a line and every
One column all have the datum mark of identical quantity；
Step 2) calculates the pixel of b-th of datum mark in a-th of phase screen using all datum marks for being marked as given value
Value, and its pixel value is labeled as given value；A is the positive integer from 1 to N, and b is the positive integer from 1 to M, and N is phase to be generated
Position screen number, M are the number of datum mark in each phase screen；
Step 3) is input to three-dimensional covariance matrix using the location information of all datum marks for being marked as given value and predicts mould
Type obtains the probability distribution value of b-th of datum mark of a-th of phase screen, by probability distribution value multiplied by relevant to position random
Undulating value obtains generic pixel difference；And its absolute pixel differences value is labeled as given value；
Pixel value and generic pixel difference value are obtained the absolute value of b-th of datum mark of a-th of phase screen by step 4)；
B value is added 1 by step 5)；
Step 6) repeats step 2) to step 5), finishes until M datum mark of a-th of phase screen all calculates；
A value is added 1 by step 7)；
Step 8) repeats step 2 to 7, finishes until all datum marks in all N number of phase screens all calculate；
Step 9) generates a process of (K-P+1) × (K-P+1) for the 1st phase screen, and each process contains 1 frame, each
Frame contains P × P pixel in the 1st phase screen；Each process is for calculating all pixels point in the frame that it contains
Absolute value, different, the mutual relationship of frame in each process are as follows: the point in the frame upper left corner of the 1st process is
1st datum mark of the 1st row of current phase screen, the point in the frame upper left corner of the 2nd process are the 2nd of current the 1st row of phase screen the
The point in a datum mark ... (K-P+1) × (K-P+1) a process frame upper left corner is the K of the line k of current phase screen
A datum mark；Wherein K is the number of the datum mark of phase screen a line；It include Q datum mark in each frame；
Step 10) uses the absolute value of each point in the method for parallel computation while Computational frame, specifically:
The absolute of the point that any one is not labeled as given value is calculated using the datum mark for being marked as given value in frame
Value, until all the points in frame all calculate completion；
After the completion of all processes of step 11) calculate, absolute value is calculated to identical point in each process and is averaging, is obtained most
The absolute value of each point eventually；
Step 12) repeats step 9) to step 11), until all the points in all N number of phase screens for next phase screen
It all calculates and finishes, and export N number of phase screen；
The three-dimensional covariance matrix prediction model is used to calculate the phase fluctuation situation between any two points of space, calculation formula
Are as follows:
WithRepresent any two spatial position vector；x1,y1,z1ForThree coordinate components；x2,y2,z2ForThree
Coordinate components；xn,yn,znFor three coordinate components of nth pixel point in current phase screen image；To removeIt is representative
Point outside, the spatial position vector of remaining all the points；To removeOutside representative point, the space bit of remaining all the points
Set vector；x′1,y′1,z′1ForThree coordinate components；x′2,y′2,z′2ForCoordinate components；x′n,y′n,z′nIt is current
It is removed outside nth pixel point in phase screen image, corresponding three coordinate components of the spatial position vector of remaining all the points；DφFor
Phase indirect detection；T () is weighting function, and it is 0 outside region that value, which is 1, in zone of dispersion；C is phase covariance function.
2. the method according to claim 1, wherein in step 2), j-th datum mark in i-th of phase screen
Pixel valueCalculation formula are as follows:
xi,yj,zkComponent of any point on the direction x, the direction y and the direction z respectively in space；Wherein A is prediction square
Battle array, φmFor known phase column matrix, σ is that interpolation criteria is poor, and R is that variance is the Gaussian random variable that 1 mean value is 0；It is wherein pre-
Survey the form of matrix A are as follows:
The form of interpolation criteria difference are as follows:
3. the method according to claim 1, wherein initial pictures and each phase screen have 9 × 9 benchmark
Point, each frame have 9 × 9 pixels, and each frame has 3 × 3 datum marks.
4. according to the method described in claim 3, it is characterized in that, step 10) specifically:
The absolute value of the geometric center point for the minimum rectangle that any four known point is constituted is calculated, if geometric center point is greater than 1
It is a, then the absolute value of these points is successively calculated according to ranks sequence；Calculate the several of the minimum rectangle that any four known point is constituted
The absolute value of what central point calculates the absolute value of all the points according to the sequence of pixel if geometric center point can not be found.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611236912.4A CN106529104B (en) | 2016-12-28 | 2016-12-28 | The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611236912.4A CN106529104B (en) | 2016-12-28 | 2016-12-28 | The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106529104A CN106529104A (en) | 2017-03-22 |
| CN106529104B true CN106529104B (en) | 2019-08-23 |
Family
ID=58338130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611236912.4A Expired - Fee Related CN106529104B (en) | 2016-12-28 | 2016-12-28 | The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106529104B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107145667B (en) * | 2017-05-08 | 2020-07-03 | 中国科学院长春光学精密机械与物理研究所 | Rapid calculation method of wave front structure function |
| CN111664937B (en) * | 2020-06-08 | 2021-04-30 | 西安电子科技大学 | Method for determining fluctuation time characteristic of partially coherent Gaussian beam irradiance |
| CN113225130B (en) * | 2021-03-25 | 2022-09-09 | 中国人民解放军国防科技大学 | Atmospheric turbulence equivalent phase screen prediction method based on machine learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102183354A (en) * | 2011-02-25 | 2011-09-14 | 中国科学院上海光学精密机械研究所 | Turbulence simulation device based on optical addressing liquid crystal spatial light modulator |
| CN103605860A (en) * | 2013-11-28 | 2014-02-26 | 中国科学院软件研究所 | Non-ideal light source simulation method based on equivalent phase screen method |
| CN104374546A (en) * | 2014-09-25 | 2015-02-25 | 太原理工大学 | Universal atmosphere turbulence phase screen design method based on diffractive optical element |
-
2016
- 2016-12-28 CN CN201611236912.4A patent/CN106529104B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102183354A (en) * | 2011-02-25 | 2011-09-14 | 中国科学院上海光学精密机械研究所 | Turbulence simulation device based on optical addressing liquid crystal spatial light modulator |
| CN103605860A (en) * | 2013-11-28 | 2014-02-26 | 中国科学院软件研究所 | Non-ideal light source simulation method based on equivalent phase screen method |
| CN104374546A (en) * | 2014-09-25 | 2015-02-25 | 太原理工大学 | Universal atmosphere turbulence phase screen design method based on diffractive optical element |
Non-Patent Citations (5)
| Title |
|---|
| Fast simulation of a Kolmogorov phase screen;Cressida M. Harding等;《Applied Optics》;19990410;第38卷(第11期);第2161-2170页 |
| Fractal phase screen generation algorithm for atmospheric turbulence;HUILI ZHAI等;《Applied Optics》;20150501;第53卷(第13期);第4023-4032页 |
| 一种空间激光通信中模拟湍流效应的新方法;向劲松等;《红外与激光工程》;20150930;第44卷(第9期);第2721-2725页 |
| 基于自适应光学的激光精跟踪技术研究;王保峰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150415;第2015年卷(第04期);第I135-20页 |
| 相位差波前探测技术及其在图像恢复中的应用研究;李斐;《中国博士学位论文全文数据库 信息科技辑》;20120715;第2012年卷(第07期);第I138-58页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106529104A (en) | 2017-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106529104B (en) | The phase screen emulation mode that a kind of light is closely propagated in turbulent flow under water | |
| CN106646645B (en) | A kind of gravity forward modeling accelerated method | |
| CN107607942B (en) | Based on the large scale electromagnetic scattering of deep learning model and the prediction technique of back scattering | |
| CN103414853A (en) | Device and method for stabilizing video image sequence capable of doing multi-degree of freedom movement in real time | |
| CN104463970B (en) | Method for determining three-dimensional gravity center of city based on remote-sensing image and application thereof | |
| Zhu et al. | Error estimation of 3D reconstruction in 3D digital image correlation | |
| CN109086663A (en) | The natural scene Method for text detection of dimension self-adaption based on convolutional neural networks | |
| Kim et al. | Adaptive weighted sum method for multiobjective optimization | |
| CN116774292A (en) | Seismic wave travel time determining method, system, electronic equipment and storage medium | |
| CN108828669A (en) | A kind of two-dimensional intersection survey line static corrections processing method, apparatus and system | |
| Du et al. | New cross-matching algorithm in large-scale catalogs with ThreadPool technique | |
| CN107544098A (en) | Generation method, device, storage medium and the processor of roughness of ground surface | |
| Weinbub et al. | Shared-memory parallelization of the fast marching method using an overlapping domain-decomposition approach | |
| CN103716639B (en) | Search algorithm of frame image motion estimation | |
| Zhang et al. | Stereo matching cost computation based on nonsubsampled contourlet transform | |
| CN107976714A (en) | A kind of channel set computational methods of complicated earth surface classification spatial distance weighting | |
| CN117473877B (en) | Lightning three-dimensional radiation source position inversion method based on stationary satellite data | |
| Xu et al. | An optimization method for ITER radial x-ray camera line-of-sight design basing on Bayesian uncertainty analysis | |
| Arnold et al. | Vision-Assisted Digital Twin Creation for mmWave Beam Management | |
| CN105628342B (en) | A kind of optical flow visualization technical optics are laid out method of value solving | |
| Ertl et al. | Three-dimensional topography simulation using advanced level set and ray tracing methods | |
| CN106291549B (en) | A kind of the diameter radar image emulation mode and its application of metal polygon | |
| Wang et al. | Visual-perception-driven Urban Three-dimensional Scene Data Scheduling Method. | |
| CN113486772B (en) | Human body posture estimation model training method, estimation method, device and equipment | |
| Lan et al. | A cluster-analysis and convex hull-based fast large-scale phase unwrapping method for single-and multibaseline SAR interferograms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190823 Termination date: 20211228 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |