CN110288564A - Binaryzation speckle quality evaluating method based on power spectrumanalysis - Google Patents

Binaryzation speckle quality evaluating method based on power spectrumanalysis Download PDF

Info

Publication number
CN110288564A
CN110288564A CN201910427975.5A CN201910427975A CN110288564A CN 110288564 A CN110288564 A CN 110288564A CN 201910427975 A CN201910427975 A CN 201910427975A CN 110288564 A CN110288564 A CN 110288564A
Authority
CN
China
Prior art keywords
speckle
district
binaryzation
error
sub
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
Application number
CN201910427975.5A
Other languages
Chinese (zh)
Other versions
CN110288564B (en
Inventor
吴文杰
刘聪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
Original Assignee
Nanjing University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology filed Critical Nanjing University of Science and Technology
Priority to CN201910427975.5A priority Critical patent/CN110288564B/en
Publication of CN110288564A publication Critical patent/CN110288564A/en
Application granted granted Critical
Publication of CN110288564B publication Critical patent/CN110288564B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of binaryzation speckle quality evaluating method based on power spectrumanalysis, method includes the following steps: camera acquires binaryzation speckle pattern；The extraction of gray value is carried out to the pattern of acquisition and does noise reduction process；It sets sub-district size and confines target sub-district；Picture power spectrum, speckle duty ratio, noise variance are obtained in selected sub-district；Power spectrum is subjected to inverse Fourier transform, and then solves the probability density function that the adjacent two o'clock of speckle is same speckle primitive；Error solving precision is set, speckle error amount is solved and is respectively worth corresponding probability, and finds out error mean and variance, and solve all target sub-district average means and variance, achievees the purpose that evaluate speckle with this.The present invention has carried out binaryzation speckle quality evaluation by the way of power spectrumanalysis, and method can have many advantages, such as that equipment is simple, convenient and practical according to experimenter for the required precision unrestricted choice error precision of experiment.

Description

Binaryzation speckle quality evaluating method based on power spectrumanalysis
Technical field
The present invention relates to flash ranging experimental solid mechanics field, especially a kind of binaryzation speckle matter based on power spectrumanalysis Measure evaluation method.
Background technique
Binaryzation speckle is widely used in digital picture and is concerned in the measurement of (DIC), big with shade of gray, and measurement misses The advantages that difference is small.
Now for the quality evaluation of binaryzation speckle mostly from speckle duty ratio, speckle radius angularly, lack pair In considering for speckle power spectrum.Also, existing speckle quality evaluation mostly uses the mode of numerical simulation.Such as, Pan Bing, Wu great Fang Et al. five visibly different speckle patterns are utilized in the quality evaluation of speckle pattern " in Digital Image Correlation Method research " text Accurate translation has been carried out, and measured value and predicted value have been compared, mean value error has been analyzed and standard deviation, such mode lacks Generality, and lack specific experimental embodiment.
Summary of the invention
The purpose of the present invention is to provide a kind of binaryzation speckle quality evaluating method based on power spectrumanalysis, with power Based on spectrum analysis, from experimental viewpoint, succinctly intuitively evaluation power spectrum is fine or not, easy to operate, it is easy to accomplish.
The technical solution for realizing the aim of the invention is as follows: a kind of binaryzation speckle quality evaluation based on power spectrumanalysis Method, this method experimental provision include camera, speckle pattern fixed plate, binaryzation speckle to be measured, white light source, optical platform, electricity Sub- computer；Method includes the following steps:
Step 1, camera acquire binaryzation speckle pattern；
Step 2 carries out gray value extraction to the pattern of acquisition, and is saved with a matrix type, extracts from matrix Noise component(s) simultaneously removes, and obtains speckle duty ratio, camera noise expression formula；
Step 3, setting sub-district size, confine several target sub-districts；
Step 4 chooses target sub-district, obtains the corresponding sub-district power spectrum of picture: being become using image subsection auto-correlation Fourier It changes or other modes acquires picture sub-district power spectrum, result is stored in matrix Sxx；
Power spectrum, is subtracted the definite value component determined by sub-district size and put down with speckle duty ratio by step 5, setting sub-district size The product of side, and then solve the probability density function that the adjacent two o'clock of speckle is same speckle primitive；
Step 6, setting error solving precision, solve speckle error amount and are respectively worth corresponding probability, and it is equal to find out error Value and variance；
Step 7 chooses target sub-district again, repeats step 4~step 6, until the selection of target sub-district finishes, utilizes each son Area's error mean and variance seek overall error mean value and variance.
Compared with prior art, the present invention its remarkable advantage is: (1) the method for the present invention is innovatively using power spectrum point The mode of analysis has carried out binaryzation speckle quality evaluation, and realizes the function from the angle of experiment；(2) method can be according to reality For the person of testing for the required precision unrestricted choice error precision of experiment, accuracy value is smaller than 0.001pixel；(3) equipment required for Simply, convenient and practical, the advantages that visual result is obvious.
Detailed description of the invention
Fig. 1 is measuring device schematic diagram of the present invention.
Fig. 2 is that the present invention is based on the binaryzation speckle quality evaluating method flow charts of power spectrumanalysis.
Fig. 3 is the quality analysis results schematic diagram of speckle in the embodiment of the present invention.
Specific embodiment
As shown in Figure 1 and Figure 2, the experimental provision of the binaryzation speckle quality evaluating method includes: an industrial camera 1, height Resolution ratio camera lens 2, speckle pattern fixed plate and its clamping device 3, binaryzation speckle 4 to be measured, white light source 5, optical platform 6, electricity Sub- computer 7；Method includes the following steps:
Step 1, camera acquire binaryzation speckle pattern: under the irradiation of white light source, utilizing the work with high-resolution lens Industry camera is acquired binaryzation speckle.
Step 2, characteristics of image and noise characteristic extract: carrying out the extraction of gray value, and to the pattern of acquisition with matrix Form is saved.And then gray matrix is divided into two major parts submatrix, the data between two pieces of submatrixs all have obvious number It is worth size gap.Then it averages respectively to two pieces of submatrixs, and records the difference of each pixel practical value and the mean value, The difference is recorded as each pixel spot noise figure.Finally regard expressions of noise as gaussian probability distribution and expression that mean value is zero Formula, pixel spot noise figure variance are the expression formula variance, then noise probability density distribution expression formula f (I) are as follows:
Wherein, I is noisy gray-value, and σ is noise variance；
Regard the ratio of the element number of the biggish submatrix of numerical value and entire gray matrix element number as speckle duty Than being denoted as δ.
Step 3, setting sub-district size, are the rectangular area of L*L, and user sets N number of target sub-district.
Step 4 chooses target sub-district, obtains the corresponding sub-district power spectrum of picture: being become using image subsection auto-correlation Fourier It changes or other modes acquires picture sub-district power spectrum, and then result is stored in matrix Sxx。
Step 5 first solves the definite value component determined by sub-district size, is expressed asAs a result as follows:
Wherein, u, v are power spectrum variable, and L is sub-district dimensional parameters.
Then, speckle duty ratio is multiplied with the definite value component determined by sub-district size, and ask speckle power spectrum and its Difference simultaneously stores in the matrix form, and gained difference is expressed asAs a result are as follows:
It solves againWith (δ-δ2) quotient inverse Fourier transform Fsame(x, y) is saved in the matrix form:
Finally, solving the probability density function f that the adjacent two o'clock of speckle is same speckle primitivesame(x, y):
Step 6 sets error solving precision as unit of pixel.
Speckle error amount and the solution procedure for being respectively worth corresponding probability are as follows:
Firstly, delimiting in 3 regions σ is error range that may be present, using the error solving precision ε of setting to the region It is divided, obtains each sample point coordinates:
I=-Ni,-Ni+1…0…Ni-1,Ni,
J=-Nj,-Nj+1…0…Nj-1,Nj
Wherein,
And then confirm the coherent value R between two o'clock gray scalei,j:
Wherein, I~N (1,2 σ2)。
The probability that each sample point obtains maximum value is solved again:
Wherein, erf (x) is error function, f (Ri,j) it is Ri,jThe probability density function of satisfaction, expression formula are as follows:
Wherein,
The above maximum value asks method using numerical solution.
After having solved speckle error amount and being respectively worth corresponding probability, error mean and variance are then asked.Error mean with Variance includes the single error δ of horizontal, vertical both directionU、δV, the composition error δ comprising two directionsUV.It calculates step such as Under:
Firstly, seeking the error delta δ of each pixelU,ΔδV,ΔδUVMean and variance is denoted as E (Δ δ respectivelyU)、D(Δ δU)、E(ΔδV)、D(ΔδV)、E(ΔδUV)、D(ΔδUV), calculation method is as follows:
In turn, it is re-introduced into influence of the sub-district size for error, calculates sub-district global error, mean value is denoted as respectively: E (δU)、E(δV)、E(δUV), variance is denoted as D (δ respectivelyU)、D(δV)、D(δUV), as a result are as follows:
Step 7 chooses target sub-district again, repeats step 4,5,6, until the selection of target sub-district finishes, records each son Area's error mean and variance.Overall error mean value and variance are acquired using each sub-district error mean and variance, mean value is denoted as respectively: EAlways(δU)、EAlways(δV)、EAlways(δUV), variance is denoted as respectively: DAlways(δU)、DAlways(δV)、DAlways(δUV), calculation is as follows:
Factor of the error mean finally obtained as primary concern speckle quality.
Using above-mentioned measurement method, actual speckle quality analysis is carried out for several speckles.Utilize numerical simulation software Having produced a series of radiuses is 2, the speckle of different duty.It is analyzed using quality of the above-mentioned calculation method to speckle, Analysis result is as shown in figure 3, can be seen that experimental result and theory analysis fitting well from result of implementation.

Claims (10)

1. a kind of binaryzation speckle quality evaluating method based on power spectrumanalysis, which is characterized in that the experimental provision of this method Including camera, speckle pattern fixed plate, binaryzation speckle to be measured, white light source, optical platform, electronic computer, this method includes Following steps:
Step 1, camera acquire binaryzation speckle pattern；
Step 2 carries out gray value extraction to the binaryzation speckle pattern of acquisition, and is saved with a matrix type, from matrix Middle extraction noise component(s) simultaneously removes, and obtains speckle duty ratio, camera noise expression formula；
Step 3, setting sub-district size, confine several target sub-districts；
Step 4 chooses target sub-district, obtains the corresponding sub-district power spectrum of picture: being asked using image subsection auto-correlation Fourier transformation Picture sub-district power spectrum is obtained, result is stored in matrix Sxx；
Power spectrum is subtracted the definite value component determined by sub-district size and speckle duty ratio square by step 5, setting sub-district size Product, and then solve the probability density function that the adjacent two o'clock of speckle is same speckle primitive；
Step 6, setting error solving precision, solve and speckle error amount and are respectively worth corresponding probability, and find out error mean with Variance；
Step 7 chooses target sub-district again, repeats step 4~step 6, until the selection of target sub-district finishes, is missed using each sub-district Poor mean value and variance seek overall error mean value and variance.
2. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step The specific method is as follows for noise component(s) removal and form of noise extraction in rapid 2:
Gray matrix is divided into two major parts submatrix by step 2-1, is averaged respectively to two pieces of submatrixs, and record each pixel The difference of point practical value and the mean value, records the difference as each pixel spot noise figure；
Step 2-2, noise regard the gaussian probability distribution that mean value is zero as, and pixel spot noise figure variance is the expression formula variance, Then noise probability density distribution expression formula f (I) are as follows:
Wherein, I is noisy gray-value, and σ is noise variance.
3. the binaryzation speckle quality evaluating method according to claim 2 based on power spectrumanalysis, which is characterized in that step Speckle duty ratio in rapid 2 is the element number of the biggish submatrix of numerical value and the ratio of entire gray matrix element number, note For δ.
4. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step The rectangular area that sub-district in rapid 3 is L*L, target sub-district number are N.
5. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that institute The definite value representation in components stated in step 5 isAs a result as follows:
Wherein, u, v are power spectrum variable, and L is sub-district dimensional parameters.
6. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step It is as follows to solve mode for probability density function in rapid 5:
Speckle duty ratio is multiplied with the definite value component determined by sub-district size, and ask speckle power spectrum and its difference and with square The storage of formation formula, gained difference are expressed asAs a result are as follows:
It solvesWith (δ-δ2) quotient inverse Fourier transform Fsame(x, y) is as a result saved in the matrix form:
Solve the probability density function f that the adjacent two o'clock of speckle is same speckle primitivesame(x, y):
7. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step Error solving precision unit in rapid 6 is pixel.
8. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step Speckle error amount in rapid 6 and the solution procedure for being respectively worth corresponding probability are as follows:
It delimit in 3 regions σ as error range that may be present, utilizes two deflection error solving precision ε of settingi、εjTo the region It is divided, obtains each sample point coordinates:
I=-Ni,-Ni+1…0…Ni-1,Ni,
J=-Nj,-Nj+1…0…Nj-1,Nj
Wherein,
Confirm the coherent value R between two o'clock gray scalei,j:
Wherein, I~N (1,2 σ2)；
Solve the probability that each sample point obtains maximum value
Wherein, erf (x) is error function, f (Ri,j) it is Ri,jThe probability density function of satisfaction, expression formula are as follows:
Wherein,
The above maximum value asks method using numerical solution.
9. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that step Error mean and variance in rapid 6 include the single error δ of horizontal, vertical both directionU、δV, the composition error comprising two directions δUV, calculation method is as follows:
Seek the error delta δ of each pixelU,ΔδV,ΔδUVMean and variance is denoted as E (Δ δ respectivelyU)、D(ΔδU)、E(ΔδV)、 D(ΔδV)、E(ΔδUV)、D(ΔδUV), calculation method is as follows:
Influence of the sub-district size for error is introduced, sub-district global error is calculated, mean value is denoted as respectively: E (δU)、E(δV)、E (δUV), variance is denoted as D (δ respectivelyU)、D(δV)、D(δUV), as a result are as follows:
10. the binaryzation speckle quality evaluating method according to claim 1 based on power spectrumanalysis, which is characterized in that Overall error mean value and variance, mean value in step 7 are denoted as respectively: EAlways(δU)、EAlways(δV)、EAlways(δUV), variance is denoted as respectively: DAlways (δU)、DAlways(δV)、DAlways(δUV), calculation is as follows:
CN201910427975.5A 2019-05-22 2019-05-22 Binarization speckle quality evaluation method based on power spectrum analysis Active CN110288564B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910427975.5A CN110288564B (en) 2019-05-22 2019-05-22 Binarization speckle quality evaluation method based on power spectrum analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910427975.5A CN110288564B (en) 2019-05-22 2019-05-22 Binarization speckle quality evaluation method based on power spectrum analysis

Publications (2)

Publication Number Publication Date
CN110288564A true CN110288564A (en) 2019-09-27
CN110288564B CN110288564B (en) 2022-08-23

Family

ID=68002478

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910427975.5A Active CN110288564B (en) 2019-05-22 2019-05-22 Binarization speckle quality evaluation method based on power spectrum analysis

Country Status (1)

Country Link
CN (1) CN110288564B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112629828A (en) * 2020-11-27 2021-04-09 奥比中光科技集团股份有限公司 Optical information detection method, device and equipment
CN113077429A (en) * 2021-03-30 2021-07-06 四川大学 Speckle quality evaluation method based on adjacent sub-area correlation coefficient
CN113888614A (en) * 2021-09-23 2022-01-04 北京的卢深视科技有限公司 Depth recovery method, electronic device, and computer-readable storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104833483A (en) * 2015-04-29 2015-08-12 山东大学 Speckle measuring and evaluating method and application in laser projection display
US20170064272A1 (en) * 2015-08-28 2017-03-02 Yanning Zhao Device and method for characterization of subjective speckle formation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104833483A (en) * 2015-04-29 2015-08-12 山东大学 Speckle measuring and evaluating method and application in laser projection display
US20170064272A1 (en) * 2015-08-28 2017-03-02 Yanning Zhao Device and method for characterization of subjective speckle formation

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112629828A (en) * 2020-11-27 2021-04-09 奥比中光科技集团股份有限公司 Optical information detection method, device and equipment
CN113077429A (en) * 2021-03-30 2021-07-06 四川大学 Speckle quality evaluation method based on adjacent sub-area correlation coefficient
CN113888614A (en) * 2021-09-23 2022-01-04 北京的卢深视科技有限公司 Depth recovery method, electronic device, and computer-readable storage medium
CN113888614B (en) * 2021-09-23 2022-05-31 合肥的卢深视科技有限公司 Depth recovery method, electronic device, and computer-readable storage medium

Also Published As

Publication number Publication date
CN110288564B (en) 2022-08-23

Similar Documents

Publication Publication Date Title
CN110288564A (en) Binaryzation speckle quality evaluating method based on power spectrumanalysis
Grediac et al. 50th anniversary article: effect of sensor noise on the resolution and spatial resolution of displacement and strain maps estimated with the grid method
CN107084794B (en) Flame three-dimensional temperature field measuring system and method based on light field layered imaging technology
Zappa et al. Evaluation and improvement of digital image correlation uncertainty in dynamic conditions
CN102840829B (en) Manual mark-basedsystem and manual mark-based method for measuring displacement field inside high temperature object area
WO2015008404A1 (en) Method and device for measuring displacement distribution using regular pattern, and program for such method and device
CN109374580A (en) A kind of Terahertz lamination image probe positional error correction method
CN109186759B (en) Grating spectrometer image quality measuring method and device
CN104024793A (en) Shape inspection method and device
CN103592108A (en) CCD chip modulation transfer function test device and method
CN108051183B (en) Focus type light-field camera parameter calibration method based on first-order theory
Zhang et al. Accuracy improvement in laser stripe extraction for large-scale triangulation scanning measurement system
CN109141273B (en) DMD-based high-speed moving target deformation measurement system and method
Aparna et al. Fatigue testing of continuous GFRP composites using digital image correlation (DIC) technique a review
CN116560080B (en) Single photon single pixel imaging device and method through thick scattering medium
Chen et al. High-speed 3D surface profilometry employing trapezoidal phase-shifting method with multi-band calibration for colour surface reconstruction
Zappa et al. Digital image correlation technique in dynamic applications on deformable targets
CN203390388U (en) Accurate location device for laser focus
Antoš et al. Real-time optical measurement of displacements using subpixel image registration
Sun et al. Sampling light field for photometric stereo
JP2913021B2 (en) Shape measuring method and device
Liao et al. Portable high‐resolution automated 3D imaging for footwear and tire impression capture
CN103383353A (en) Optical vortices-based dynamic speckle test method
Zhao et al. Uncertainty quantification for 3D digital image correlation displacement measurements using Monte Carlo method
CN113483879B (en) Small satellite flutter high-speed video measurement method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information

Inventor after: Liu Cong

Inventor after: Wu Wenjie

Inventor before: Wu Wenjie

Inventor before: Liu Cong

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant