CN108711151A - A kind of weld defects detection method, apparatus, equipment, storage medium and system - Google Patents
A kind of weld defects detection method, apparatus, equipment, storage medium and system Download PDFInfo
- Publication number
- CN108711151A CN108711151A CN201810495731.6A CN201810495731A CN108711151A CN 108711151 A CN108711151 A CN 108711151A CN 201810495731 A CN201810495731 A CN 201810495731A CN 108711151 A CN108711151 A CN 108711151A
- Authority
- CN
- China
- Prior art keywords
- sparse
- magneto
- vector
- matrix
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 106
- 238000001514 detection method Methods 0.000 title claims abstract description 72
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 111
- 238000000034 method Methods 0.000 claims abstract description 44
- 238000011084 recovery Methods 0.000 claims abstract description 40
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 238000005457 optimization Methods 0.000 claims abstract description 25
- 238000003466 welding Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 4
- 230000006698 induction Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000009467 reduction Effects 0.000 claims description 2
- 238000005070 sampling Methods 0.000 abstract description 22
- 230000009466 transformation Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 15
- 208000037656 Respiratory Sounds Diseases 0.000 description 12
- 230000008569 process Effects 0.000 description 10
- 230000005284 excitation Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000003384 imaging method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000009659 non-destructive testing Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 108700020469 14-3-3 Proteins 0.000 description 1
- 102000004899 14-3-3 Proteins Human genes 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 230000008719 thickening Effects 0.000 description 1
- 239000010409 thin film Substances 0.000 description 1
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
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/006—Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
-
- G06T3/06—
-
- 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
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2211/00—Image generation
- G06T2211/40—Computed tomography
- G06T2211/416—Exact reconstruction
Abstract
The embodiment of the invention discloses a kind of weld defects detection method, apparatus, equipment, computer readable storage medium and systems.Wherein, method includes obtaining sparse magneto-optical signal to the magneto-optic figure sparse signal representation of the weldment to be measured of sparsity with canonical orthogonal basic matrix and sparse coefficient；Sparse coefficient is observed using with canonical orthogonal basic matrix incoherent observing matrix, the sparse magneto-optical signal of higher-dimension of transformation gained, which is projected in lower dimensional space, obtains projection value；Compressed sensing restructing algorithm based on Lp norms solves Sparse Optimization, restores the recovery coefficicnt vector of sparse coefficient from projection value；The original high dimensional information that weldment to be measured is rebuild according to recovery coefficicnt vector sum orthonormal basis, using the original image signal as detection weld defect.The technical solution of the application, which realizes, rebuilds the magneto optic images signal based on the weldment to be detected acquired under lack sampling phenomenon, obtains complete, clear, accurate weld defect magneto-optic figure, is conducive to the accuracy of detection for improving weldment defect.
Description
Technical field
The present embodiments relate to laser welding detection technique field, more particularly to a kind of weld defects detection method,
Device, equipment, computer readable storage medium and system.
Background technology
With the development of the heavy industry such as China's Aeronautics and Astronautics, the energy and ship, welding technique has become must in manufacturing industry
One of indispensable material molding and processing technology.Due to welding procedure be vulnerable in process from external environment etc. because
The influence of element, such as speed of welding, bonding power, shield gas flow rate, welding surroundings and workpiece surface situation, weldment can not
It will produce the weld defects such as crackle, lack of penetration, incomplete fusion, stomata, pit, slag inclusion with avoiding, and these defects are likely to result in
Catastrophic failure.In order to ensure the product quality of weldment, it is necessary to timely and effectively detect the defect of postwelding workpiece surface and inside.
In the actual production process, other than range estimation face of weld defect and forming defect, 0.1mm weld defects are less than for some,
Generally can not but effectively it be identified by range estimation.
Magneto-optic imaging is used as lossless detection method, is widely used in the weld seam detection technology of weldment.It is imitated based on magneto-optic
The magneto-optic imaging method schematic diagram answered please refers to Fig.1 shown.Based on Faraday magnetooptical effect, light source is changed into after the polarizer
A branch of polarised light passes through magnet-optical medium, and through magneto-optic thin film reflective surface, due to the presence of externally-applied magnetic field, plane of polarization occurs inclined
Turn, the polarised light for deflecting certain angle is acquired through analyzer by CMOS, changes of magnetic field information be changed into light intensity variation it is real-time at
Picture.
Fig. 2 is weld seam magneto-optic image forming job schematic diagram, is based on faraday's magnetic rotation principle, and AC power gives electromagnetism Tie Tong
Alternating current, and generate alternately changing magnetic field.Light source led generates linearly polarized photon after polarizer, and magneto-optic is popped one's head in by magneto-optic
Medium and reflecting layer composition, incident ray polarized light reflect via reflecting layer and are situated between again by magneto-optic after magnet-optical medium
Matter, analyzer detect corresponding optical signal and by CMOS camera imagings, convert the information of changes of magnetic field to the real-time of light intensity
Imaging.
Traditional magneto-optic imaging nondestructive inspection method acquires time-domain signal by magneto-optical sensor, then is become by Fourier
It changes and rebuilds spatial domain signal, the internal image of weldment is obtained, since Fourier transform is linear transformation, it is therefore desirable to the domains k of acquisition
Signal number is necessarily equal to the pixel number of image area.And according to Shannon-how Gui this special sampling thheorem, the maximum sampling of magneto-optical sensor
When frequency is greater than twice of the frequency of AC excitation signal, friendship has just been fully retained in the magneto optic images information sequence after acquiring
The weld defect information of material in varying signal.And in practical application, the frequency of the Alternating Current Excitation signal applied is 50Hz, and magnetic
The sample frequency range of optical sensor is 0~75Hz, it is seen that the maximum sample frequency of magneto-optical sensor is simultaneously unsatisfactory for sampling thheorem
Condition, it may occur that lack sampling phenomenon, the spectrum overlapping of the magneto optic images information, that is, be higher than sample frequency half frequency content will
It is reconstructed into the signal less than sample frequency half and then aliasing when causing to rebuild, it is difficult to collect and completely clearly weld
Seam defect the magneto optic images information is also just unable to judge accurately welding and whether there is defect.
The phenomenon for avoiding above-mentioned weld defect detection limited, the metal defect non-destructive testing technology that the prior art mostly uses are
Ray detection method specially utilizes ray (such as x-ray, gamma-rays etc.) to have centainly during across weldment to be measured
Attenuation law, then according to by across weldment to be measured the decaying of each position after transmitted intensity lacked inside the weldment to detect
A kind of sunken method.The attenuation degrees of different objects is different, the degree of decaying by the thickness of object, the material category of object and
The type of ray and determine.
Ray detection method is mainly for detection of inside workpiece volume flaw, and the thickness of workpiece is not easy more than 80mm,
Corresponding thickening can be done according to the attenuation coefficient of material or is thinned.This method testing cost is high, and detection device is larger, generation
Ray radiation is very big to human injury, relatively low to the detection sensitivity of micro-crack defect.
In consideration of it, how the magneto optic images information based on lack sampling phenomenon, realize the non-destructive testing of weld defect, be ability
Field technique personnel's urgent problem to be solved.
Invention content
The purpose of the embodiment of the present invention is to provide a kind of weld defects detection method, apparatus, equipment, computer-readable storage
Medium and system realize the original the magneto optic images signal of the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling, obtain
To complete, clear, accurate weld defect the magneto optic images, be conducive to the accuracy of detection for improving weldment defect.
In order to solve the above technical problems, the embodiment of the present invention provides following technical scheme：
On the one hand the embodiment of the present invention provides a kind of weld defects detection method, including：
The original the magneto optic images signal for obtaining the weldment to be measured of discretization acquisition, utilizes canonical orthogonal basic matrix and sparse system
It is several that sparse variation expression is carried out to the magneto optic images signal with sparsity, obtain the sparse magneto-optical signal of higher-dimension；
The sparse magneto-optical signal of the higher-dimension is projected in using with the canonical orthogonal basic matrix incoherent observing matrix
Lower dimensional space, and obtain projection value；The observing matrix and the product of the sparse coefficient meet limited equidistant condition；
Compressed sensing restructing algorithm based on Lp norms solves Sparse Optimization, from the projection value described in recovery
The recovery coefficicnt vector of sparse coefficient；
The original high dimensional information that the weldment to be measured is rebuild according to orthonormal basis described in the recovery coefficicnt vector sum, with
The magneto optic images signal as detection weld defect.
Optionally, described that the magneto optic images signal with sparsity is carried out using canonical orthogonal basic matrix and sparse coefficient
Sparse variation indicates：
The object element for meeting and presetting amplitude perturbations is chosen from the magneto optic images signal, and the magneto optic images are believed
Non-targeted element is set to 0 in number, obtains echo signal vector；
According to following formula, sparse variation is carried out to the echo signal vector and is indicated：
Yk=Fs；
In formula, YkFor echo signal vector；F is the canonical orthogonal basic matrix, and s is the sparse coefficient.
Optionally, the observing matrix is the certainty random matrix constructed based on certainty random sequence.
Optionally, the interpolation method based on compressed sensing solves Sparse Optimization, extensive from the projection value
The recovery coefficicnt vector of the sparse coefficient includes again：
The recovery coefficicnt vector for being restored the sparse coefficient from the projection value using convex optimization base tracing algorithm, is met
Following formula：
In formula, S is recovery coefficicnt vector, and s is the sparse coefficient, and F is the canonical orthogonal basic matrix, and α is institute
Observing matrix is stated, W is the projection value.
On the other hand the embodiment of the present invention provides a kind of weld defects detection device, including：
Sparse variation module acquires the original the magneto optic images signal of weldment to be measured for obtaining discretization, just using specification
It hands over basic matrix and sparse coefficient to carry out sparse variation to the magneto optic images signal with sparsity to indicate, it is sparse to obtain higher-dimension
Magneto-optical signal；
Low dimension projective module, it is with the canonical orthogonal basic matrix incoherent observing matrix that the higher-dimension is dilute for utilizing
Thin magneto-optical signal is projected in lower dimensional space, and obtains projection value；The observing matrix and the product satisfaction of the sparse coefficient have
Limit equidistant condition；
Sparse coefficient recovery module solves Sparse Optimization for the compressed sensing restructing algorithm based on Lp norms,
Restore the recovery coefficicnt vector of the sparse coefficient from the projection value；
The magneto optic images signal reconstruction module, described in being rebuild according to orthonormal basis described in the recovery coefficicnt vector sum
The original high dimensional information of weldment to be measured, using the magneto optic images signal as detection weld defect.
Optionally, the sparse variation module includes：
Object selection unit meets the object element for presetting amplitude perturbations for being chosen from the magneto optic images signal,
And set to 0 non-targeted element in the magneto optic images signal, obtain echo signal vector；
Rarefaction representation unit, for according to following formula, carrying out sparse variation to the echo signal vector and indicating：
Yk=Fs；
In formula, YkFor echo signal vector；F is the canonical orthogonal basic matrix, and s is the sparse coefficient.
Optionally, the sparse coefficient recovery module is restores institute using convex optimization base tracing algorithm from the projection value
The recovery coefficicnt vector of sparse coefficient is stated, and meets the module of following formula：
In formula, S is recovery coefficicnt vector, and s is the sparse coefficient, and F is the canonical orthogonal basic matrix, and α is institute
Observing matrix is stated, W is the projection value.
The embodiment of the present invention additionally provides a kind of weld defects detection equipment, including processor, and the processor is for holding
It is realized when the computer program stored in line storage as described in preceding any one the step of weld defects detection method.
The embodiment of the present invention additionally provides a kind of computer readable storage medium, is deposited on the computer readable storage medium
Weld defects detection program is contained, the welding is scarce any one of before being realized when the weld defects detection program is executed by processor
The step of falling into detection method.
The embodiment of the present invention finally additionally provides a kind of weld defects detection system, including：
Magnetic field generator, alternation magneto-optical sensor, power supply and weld defects detection equipment as previously described；
The magnetic field generator is connected with the power supply, for applying the additional magnetic for presetting magnetic induction intensity for weldment to be measured
；
The alternation magneto-optical sensor is connected with the weld defects detection equipment, the magnetic for acquiring the weldment to be measured
Light image, and it is sent to the processor；
It realizes to weld as described in preceding 4 any one when the processor is for executing the computer program stored in memory and lack
The step of falling into detection method.
An embodiment of the present invention provides a kind of weld defects detection methods, utilize canonical orthogonal basic matrix and sparse coefficient pair
There is weldment to be measured the original the magneto optic images signal of the higher-dimension of sparsity to carry out sparse variation expression, obtain the sparse magneto-optic letter of higher-dimension
Number；The sparse magneto-optical signal of higher-dimension is projected in lower dimensional space using with canonical orthogonal basic matrix incoherent observing matrix, and is obtained
To projection value；The product of observing matrix and sparse coefficient meets limited equidistant condition；Compressed sensing reconstruct based on Lp norms is calculated
Method solves Sparse Optimization, restores the recovery coefficicnt vector of sparse coefficient from projection value；According to recovery coefficicnt vector sum
Orthonormal basis rebuilds the original high dimensional information of weldment to be measured, using the magneto optic images signal as detection weld defect.
The advantages of technical solution provided by the present application, is, based on the weld defect the magneto optic images information after lack sampling in sky
Domain, time domain or some special domain are sparse (or compressible), can be with an observing matrix by the magneto-optical signal of higher dimensional space
It projects in lower dimensional space, these a small amount of space magneto-optical signal projection values include the enough information of weld defect signal reconstruction, lead to
It crosses and solves the Exact Reconstruction that an optimization problem carries out weld defect image high probability.The application can adopting in magneto-optical sensor
Sample frequency no more than twice of AC excitation signal frequency, and in the case of not destroying weldment, weldment weld seam to be detected
Surface and Inner Defect Testing are graphical completely visual, realize the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Original the magneto optic images signal, complete, clear, accurate weld defect the magneto optic images are obtained, as subsequent detection weldment to be measured
Defect classification or whether the Magneto-Optical information of existing defects, realize weldment defect non-destructive testing, be conducive to improve weldment defect
Accuracy of detection, to ensure weldment structure performance and used life have a very important significance.
In addition, the embodiment of the present invention provides corresponding realization device, equipment also directed to weld defects detection method and is
System, further such that the method has more practicability, described device, equipment and system have the advantages that corresponding.
Description of the drawings
It, below will be to embodiment or existing for the clearer technical solution for illustrating the embodiment of the present invention or the prior art
Attached drawing is briefly described needed in technology description, it should be apparent that, the accompanying drawings in the following description is only this hair
Some bright embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the magneto-optic imaging method principle schematic provided in an embodiment of the present invention based on magneto-optic effect；
Fig. 2 is weld seam magneto-optic image forming job schematic diagram provided in an embodiment of the present invention；
Fig. 3 is a kind of flow diagram of weld defects detection method provided in an embodiment of the present invention；
Fig. 4 is list of the technical scheme provided in an embodiment of the present invention to face of weld crackle under a kind of sample frequency
The also artwork of frame dynamic magneto-optic gray-scale map；
Fig. 5 is the single frames dynamic magneto-optic gray scale of face of weld crackle under a kind of sample frequency provided in an embodiment of the present invention
Figure；
Fig. 6 is technical scheme provided in an embodiment of the present invention to face of weld crackle under another sample frequency
The also artwork of single frames dynamic magneto-optic gray-scale map；
Fig. 7 is the single frames dynamic magneto-optic gray scale of face of weld crackle under another sample frequency provided in an embodiment of the present invention
Figure；
Fig. 8 is a kind of specific implementation mode structure chart of weld defects detection device provided in an embodiment of the present invention；
Fig. 9 is a kind of specific implementation mode structure chart of weld defects detection system provided in an embodiment of the present invention.
Specific implementation mode
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, shall fall within the protection scope of the present invention.
Term " comprising " and " having " in the description and claims of this application and above-mentioned attached drawing and they appoint
What is deformed, it is intended that is covered and non-exclusive is included.Such as contain the process of series of steps or unit, method, system, production
The step of product or equipment are not limited to list or unit, but the step of may include not listing or unit.
After describing the technical solution of the embodiment of the present invention, the various non-limiting realities of detailed description below the application
Apply mode.
Referring first to Fig. 3, Fig. 3 is a kind of flow diagram of weld defects detection method provided in an embodiment of the present invention,
The embodiment of the present invention may include the following contents：
S301：The original the magneto optic images signal for obtaining the weldment to be measured of discretization acquisition, using canonical orthogonal basic matrix and
Sparse coefficient carries out sparse variation to the magneto optic images signal with sparsity and indicates, obtains the sparse magneto-optical signal of higher-dimension.
The magneto optic images signal is the face of weld that weldment is acquired by magneto-optical sensor and the magneto optic images of inside, is acquired
The more discontinuous magneto-optical signal the better, the more spreads the more good, ensures that the correlation of these data is very low.Magneto-optical sensor can adopted arbitrarily
Information is acquired under sample frequency, the application is not limited in any way this.
Original the magneto optic images signal is the magneto optic images that magneto-optical sensor acquires weldment to be measured.
The precondition of compressed sensing algorithm is that raw information is necessary for sparse, the energy in spatial domain, frequency domain or other domains
Rarefaction representation.The collected original the magneto optic images signal of magneto-optical sensor can be dilute in spatial domain, frequency domain, wavelet field or other domains
It dredges and indicates, original the magneto optic images signal can be indicated with Y, and Y is to have limit for length's one-dimensional discrete time signal, using canonical orthogonal group moment
Battle array and sparse coefficient carry out sparse variation to the magneto optic images signal and indicate, i.e.,：
Y=Fs；
In formula, Y is original the magneto optic images signal；F is canonical orthogonal basic matrix, and s is sparse coefficient.
In above-mentioned formula, orthonormal basis matrix F is sparse basis array, and s is the equivalent representation of Y in the domains F, if sparse matrix
Non-zero number be k, then s be referred to as k- sparse, and expansion coefficient ss of the Y under orthogonal basis matrix F is by certain magnitude under normal conditions
Exponentially decay, sparse matrix s is not stringent sparse, there is considerably less big coefficient and many small coefficients, small coefficient approximation
In zero, small coefficient approximation can be set to zero, be realized and compressed with this transformation, i.e., from the global information for containing original signal
Retain the object element of higher magnitude (such as amplitude is more than 6) in the magneto optic images signal, and will be non-targeted in the magneto optic images signal
Element (compared with the element of small magnitude, such as amplitude is close to 0) is set to 0, each object element contains entirely to a certain extent
The information of image, taking down a part of collected data in this case can't cause a part of image information permanently to be lost
(they are remained contained in other data) obtains echo signal vector；
According to following formula, sparse variation is carried out to echo signal vector and is indicated：
Yk=Fs；
In formula, YkFor echo signal vector；F is canonical orthogonal basic matrix, and s is sparse coefficient.
S302：The sparse magneto-optical signal of higher-dimension is projected in low-dimensional using with canonical orthogonal basic matrix incoherent observing matrix
Space, and obtain projection value.
Basic thought based on compressive sensing theory, if signal time-space domain or some transform domain be it is sparse (or can
Compression), then one can be projected to the high dimensional signal for converting gained with the transformation incoherent observing matrix of base with one
On a lower dimensional space, then original can be reconstructed with high probability from these a small amount of projections by solving an optimization problem
Beginning signal.
Observing matrix is the observing matrix for being sampled to original the magneto optic images signal, and perceiving matrix A=α F is
Practical function is in the matrix on the coefficient vector with sparsity, and from the viewpoint of compressed sensing, perception matrix is only really
Observing matrix in meaning, and two different sparse signals are mapped in the same set (guarantor in order to ensure to perceive matrix
One by one mapping relations of the former space of card to lower dimensional space), the product of observing matrix and sparse coefficient needs to meet limited equidistant item
Part.
Observing matrix may be based on the certainty random matrix of certainty random sequence construction, such as random Gaussian measures square
Battle array can be obtained and canonical orthogonal group moment by the alternative frequency of the sample frequency and AC excitation device of change magneto-optical sensor
The incoherent different matrixes of battle array F, as observing matrix.
It is as much as possible in order to which when dropping to low-dimensional from higher-dimension to sparse matrix using observing matrix raw information can be obtained
Different information, this needs each column vector correlation of observing matrix small.Certainty is being constructed using certainty random sequence
Random observation matrix, random measurement are optimal strategy for sparse matrix, it is only necessary to which almost minimum measurement can be extensive
Required constant very little when restoring the raw information of sparse coefficient s again, and analyzing.Random Gaussian calculation matrix is compressed sensing
Most common calculation matrix in research, the element in the matrix obey mean value be zero, normal distribution when variance is 1/M, and
Between element independently of each other, i.e.,：
αijFor the element that the i-th row jth in observing matrix arranges, M is the line number for constructing observing matrix.
Can observing matrix be used for compressed sensing, and condition is to must satisfy limited isometry.Gaussian random calculation matrix
Advantage is that it is almost uncorrelated with arbitrary sparse matrix, meets the equidistant condition that is limited, and limited equidistant condition requires observation square
All row subset nearly orthogonals of battle array, which ensure that after observing matrix projects sparse coefficient (after observing), sparse
Coefficient is not present in observing matrix space, and two different projecting to just will not be unified low-dimensional measured value by such projection
On cause to obscure.The number of measurements that it is needed simultaneously is fewer, for example, being N for length, degree of rarefication is the initial data of K,
Only needA measured value can high probability recovery and rebuilding go out initial data, wherein c be one very
Small constant.
To such as minor function：
Wherein, p (t) is periodic function, and T is its period, and r and Z are respectively real number, and p (rT) is initial value, and M and N are observation
The line number and columns of matrix；Known formula (1) generates the sequence of random length：x1, x2... ... xn, next value of the observation sequence
For xn+1；Consider shaped like formula (2)：
xn+1=f (af-1(xn))； (2a)
yn=f (bf-1(xn)))； (2b)
Sequences y is generated by formula (2)n, initial value x0[- 1,1) it is arbitrarily taken in range, it will according to row or column priority principle
Sequence structure is at matrix, by random sequence ynObserving matrix α is constructed according to priority principle1：
C1For observing matrix α1The quadratic sum of middle all elements, coefficientPlay normalization, enables zn=yn- 0.5, it presses
According to row or column preferential principle by sequence structure at matrix, by random sequence znObserving matrix is constructed according to preferential principle
α：
CoefficientPlay normalization, while above-mentioned observing matrix must satisfy limited equidistant property (RIP), it is limited
Equidistant condition requires all row subset nearly orthogonals of observing matrix, the property to ensure that observing matrix projects sparse coefficient
Afterwards (after observing), sparse coefficient is not present in observing matrix space, and such projection will not be two different projections
It causes to obscure on to unified low-dimensional measured value.To arbitrary k sparse coefficients, if there are constantsSo thatIt sets up,Wei [0,1) arbitrary constant in range, x are sparse letter
Number (namely sparse coefficient S signals in the application), α is observing matrix；Then think that observing matrix meets limited equidistant property, energy
It is enough in the measurement to sparse matrix in compressed sensing.
The sparse magneto-optical signal of higher-dimension is projected in lower dimensional space using with canonical orthogonal basic matrix incoherent observing matrix,
And obtain projection value, i.e. W=α Y；F is canonical orthogonal basic matrix, and α is observing matrix, and W is projection value.
S303：Compressed sensing restructing algorithm based on Lp norms solves Sparse Optimization, restores dilute from projection value
The recovery coefficicnt vector of sparse coefficient.
Compressed sensing restructing algorithm based on Lp norms can packet greedy algorithm, orthogonal matching algorithm, convex optimized algorithm etc., tool
Body uses which kind of algorithm, the application not to be limited in any way this.
The recovery coefficicnt vector for restoring sparse coefficient from projection value using convex optimization base tracing algorithm, meets following public affairs
Formula：
In formula, S is recovery coefficicnt vector, and s is sparse coefficient, and F is canonical orthogonal basic matrix, and α is observing matrix, and W is to throw
Shadow value.
In a kind of specific embodiment, the above process can be：
S={ the s of sparse coefficient1, s2..., snP- norms be：
N is the columns of sparse coefficient s；
The reconstruct of compressed sensing is exactly to restore the process of original signal from sparse coefficient, since observation quantity is far smaller than believed
Number length when original the magneto optic images signal is sparse or compressible, and is felt so reconstruct is to solve for a underdetermined equation problem
Know that the equidistant constant of 2K ranks of matrix A is less than 1, then the signal reconstruction of compressed sensing can be converted into constraint 0- norm minimums
Non-convex duty Optimization, minimum 0 norm problem can be converted by solving underdetermined equation problem.Due to Yk=Fs, then sparse system
Number s=FTYk, uniform expression is：(s.t. is the meaning of " being limited to ", i.e. the abbreviation of " subject to ")
min||FTYk||pS.t.W=α FYk=AYk, p=0,1；
In formula, s is sparse coefficient, is the transposed matrix of sparse coefficient, and F is canonical orthogonal basic matrix, and α is observing matrix, W
For projection value, YkFor the original signal with sparsity.
As p=0,0- norms are obtained, it actually indicates the number of nonzero term in sparse coefficient.Then, equation is solved
Group W=As, A are perception matrix, that is, original the magneto optic images signal is asked to turn in the problem of projection value (observation set) of lower dimensional space
Turn to minimum 0- norms problem：
min||s||0S.t.W=As；
By solving 0- norm minimums problem come Accurate Reconstruction signal, is known by statistical theory and Combinatorial Optimization theory, accorded with
There are many vector for closing equation W=As, and most that sparse seeks to the original the magneto optic images signal looked for.Since sparse coefficient is (long
Degree is N, degree of rarefication K) in the linear combination of all nonzero term positions haveKind, optimal solution can just be obtained one by one by listing, very
It is difficult.Equidistant when perception matrix A meets RIP conditions, and non-convex 0- norm minimums are of equal value with the 1- norm minimums to relax
, by non-convex 0- norm minimum problems relaxation at a convex optimization problem, convex optimization is research convex function minimization problem
Mathematical optimization problem.Local minimum in convex Optimized model is global minimum, and the set of global minimum is one
Convex set can utilize this property to solve the reconstruction of compressed sensing.
Convex optimization problem solving, wherein most representative is that (also known as base chases after 1- norm minimums method-BP algorithm
Track), mathematical model is as follows：
min||s||1S.t.W=As；
By solving p1The comparison that middle 1- norm minimums can obtain projection value data acquisition is accurate, the effect as a result approached
Fruit is more preferable.It is M with for measured value that projection value, which is length, then being exactly in known measurements and observation square the problem of compressed sensing
Battle array on the basis of, solve underdetermined equation W=α Y obtain original the magneto optic images signal Y, then final equation reform into W=α Fs, W,
α, F are it is known that seek the 1- norm minimum problems of s.
S304：The original high dimensional information that weldment to be measured is rebuild according to recovery coefficicnt vector sum orthonormal basis, using as inspection
Survey the magneto optic images signal of weld defect.
After obtaining recovery coefficicnt vector, since in S301, original magneto-optical signal is carried out sparse variation, i.e. Y=
Fs, then when being rebuild, after obtaining the recovery coefficicnt that sparse coefficient is restored, then the original magneto-optic letter rebuild
Number, namely the original magneto-optical signal restored can be Y`=FS, S is recovery coefficicnt vector, and Y` is the original magneto-optic letter of reconstruct
Number.
Original magneto-optical signal in lack sampling phenomenon to that the information of spectrum signal aliasing artefacts occurs, is calculated based on compressed sensing
Method is restored the frequency spectrum of overlapping, has obtained complete, the clearly the magneto optic images of weldment to be measured.
Sample image of the obtained the magneto optic images information as weldment weld seam detection is either rebuild in reduction, is known using image
Other algorithm (such as fuzzy clustering algorithm) determines whether weldment to be measured is defective, has scarce by the way that the sample image is identified
Further identify that defect is crackle, pit, incomplete fusion or lack of penetration etc. when sunken.
In technical solution provided in an embodiment of the present invention, based on the weld defect the magneto optic images information after lack sampling in sky
Domain, time domain or some special domain are sparse (or compressible), can be with an observing matrix by the magneto-optical signal of higher dimensional space
It projects in lower dimensional space, these a small amount of space magneto-optical signal projection values include the enough information of weld defect signal reconstruction, lead to
It crosses and solves the Exact Reconstruction that an optimization problem carries out weld defect image high probability.The application can adopting in magneto-optical sensor
Sample frequency no more than twice of AC excitation signal frequency, and in the case of not destroying weldment, weldment weld seam to be detected
Surface and Inner Defect Testing are graphical completely visual, realize the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Original the magneto optic images signal, complete, clear, accurate weld defect the magneto optic images are obtained, as subsequent detection weldment to be measured
Defect classification or whether the Magneto-Optical information of existing defects, realize weldment defect non-destructive testing, be conducive to improve weldment defect
Accuracy of detection, to ensure weldment structure performance and used life have a very important significance.
In addition, in order to further prove that the technical solution of the application has a significant effect, present invention also provides specific
Example, please refer to Fig. 4-Fig. 7, be different sample frequencys under face of weld crackle single frames dynamic magneto-optic gray-scale map, specifically may be used
Including：
To verify the validity of technical scheme, using weld seam magneto-optic image forming job principle butt welding as shown in Figure 2
The face of weld crackle of workpiece carries out information collection afterwards, and the resolution ratio of magneto-optical sensor CMOS is 400pixel × 400pixel,
The alternative frequency of application is 50Hz, excitation voltage 220V, changes the different sample frequency of magneto-optical sensor, obtains and pass through this Shen
The face of weld crackle the magneto optic images rebuild of technical solution and the face of weld crackle magneto-optic of lower lack sampling please be imaged by magneto-optic
Image.The face of weld crackle gray-scale map that Fig. 4 and Fig. 5 is sample frequency when being 75Hz, wherein Fig. 4 are to use present techniques side
Also artwork, Fig. 5 that case is restored are the magneto-optic gray-scale maps under lack sampling, the same position that Fig. 6 and Fig. 7 are sample frequency when being 55Hz
Face of weld crackle gray-scale map, wherein Fig. 6 be using technical scheme restore also artwork, Fig. 7 be lack sampling under
Magneto-optic gray-scale map.
It can be seen that the clarity and integrality of Fig. 4 and Fig. 6 are better than Fig. 5 and Fig. 7 respectively, it was demonstrated that this application technical side
Case has validity to restoring original the magneto optic images information.
From the foregoing, it will be observed that induced field is generated in commissure using the AC excitation field that frequency can be changed, with magneto-optic imaging side
Method acquires weld defect magneto-optical signal, is based on compressive sensing theory, sparse in spatial domain using the sparse compressibility of the signal
It indicates, is sampled than conventional magneto-optic and rebuild entire original signal in few data, can be obtained within the identical sampling time more accurate
Weld defect visual information realizes the non-destructive testing of weldment weld defect.The application overcome in laser welding with magneto-optic at
Image space method acquires the lack sampling problem of signal, can shorten the magneto-optical sensor data scanning time, is carried out to original magneto-optical signal dilute
It dredges and indicates, effectively reduce the generation of aliasing artefacts, improve image reconstruction quality.In addition, the magneto optic images of weldment weld defect
Acquisition and compression are carried out with low rate simultaneously, and signaling protein14-3-3 process is the process that an optimization calculates, and makes magneto-optical sensor
Sampling and calculate cost substantially reduce.Since the frequency of AC excitation device is adjustable, different magnetic field intensity direction can be obtained
The Magneto-Optical information of lower weld defect, improves the detection efficiency of weld defect.
The embodiment of the present invention provides corresponding realization device also directed to weld defects detection method, further such that method
With more practicability.Weld defects detection device provided in an embodiment of the present invention is introduced below, welding described below
Defect detecting device can correspond reference with above-described weld defects detection method.
Referring to Fig. 8, Fig. 8 is weld defects detection device provided in an embodiment of the present invention under a kind of specific implementation mode
Structure chart, the device may include：
Sparse variation module 801, the original the magneto optic images signal of the weldment to be measured for obtaining discretization acquisition, utilizes rule
The orthogonal basic matrix of model and sparse coefficient carry out sparse variation to the magneto optic images signal with sparsity and indicate, it is sparse to obtain higher-dimension
Magneto-optical signal.
Low dimension projective module 802, for using with the incoherent observing matrix of canonical orthogonal basic matrix by the sparse magnetic of higher-dimension
Optical signal is projected in lower dimensional space, and obtains projection value；The product of observing matrix and sparse coefficient meets limited equidistant condition.
Sparse coefficient recovery module 803 solves sparse optimization for the compressed sensing restructing algorithm based on Lp norms and asks
Topic restores the recovery coefficicnt vector of sparse coefficient from projection value.
The magneto optic images signal reconstruction module 804, for rebuilding weldment to be measured according to recovery coefficicnt vector sum orthonormal basis
Original high dimensional information, using as detection weld defect the magneto optic images signal.
Optionally, in some embodiments of the present embodiment, the sparse variation module 801 may include：
Object selection unit meets the object element for presetting amplitude perturbations for being chosen from the magneto optic images signal, and will
Non-targeted element is set to 0 in the magneto optic images signal, obtains echo signal vector；
Rarefaction representation unit, for according to following formula, carrying out sparse variation to echo signal vector and indicating：
Yk=Fs；
In formula, YkFor echo signal vector；F is canonical orthogonal basic matrix, and s is the variation domain representation of echo signal vector.
Optionally, in other embodiments of the present embodiment, the sparse coefficient recovery module 803 is using convex excellent
Change base tracing algorithm and restore the recovery coefficicnt vector of sparse coefficient from projection value, and meets the module of following formula：
In formula, S is recovery coefficicnt vector, and s is sparse coefficient, and F is canonical orthogonal basic matrix, and α is observing matrix, and W is to throw
Shadow value.
The function of each function module of weld defects detection device of the embodiment of the present invention can be according in above method embodiment
Method specific implementation, specific implementation process is referred to the associated description of above method embodiment, and details are not described herein again.
From the foregoing, it will be observed that the embodiment of the present invention realizes the original of the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Beginning the magneto optic images signal obtains complete, clear, accurate weld defect the magneto optic images, is conducive to the detection for improving weldment defect
Precision.
The embodiment of the present invention additionally provides a kind of weld defects detection equipment, it may include：
Memory, for storing computer program；
Processor, for executing computer program to realize weld defects detection method described in any one embodiment as above
Step.
The function of each function module of weld defects detection equipment described in the embodiment of the present invention can be implemented according to the above method
Method specific implementation in example, specific implementation process are referred to the associated description of above method embodiment, no longer superfluous herein
It states.
Known by upper, the embodiment of the present invention realizes the original of the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Beginning the magneto optic images signal obtains complete, clear, accurate weld defect the magneto optic images, is conducive to the detection for improving weldment defect
Precision.
The embodiment of the present invention finally additionally provides a kind of computer readable storage medium, is stored with weld defects detection journey
Sequence, when the recognition of face program is executed by processor as above described in any one embodiment the step of weld defects detection method.
The function of each function module of computer readable storage medium described in the embodiment of the present invention can be according to above method reality
The method specific implementation in example is applied, specific implementation process is referred to the associated description of above method embodiment, herein no longer
It repeats.
Known by upper, the embodiment of the present invention realizes the original of the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Beginning the magneto optic images signal obtains complete, clear, accurate weld defect the magneto optic images, is conducive to the detection for improving weldment defect
Precision.
The embodiment of the present invention additionally provides a kind of weld defects detection system, refers to Fig. 9, it may include：
Power supply 901, magnetic field generator 902, weldment to be measured 903, alternation magneto-optical sensor 904 and processor 905；
Magnetic field generator 902 is connected with power supply 901, and the additional of magnetic induction intensity is preset for applying for weldment 903 to be measured
Magnetic field.
Alternation magneto-optical sensor 904 is connected with processor 905, the magneto optic images for acquiring weldment 903 to be measured, and sends
To processor 905.
Processor 905 executes computer program to realize the step of weld defects detection method described in any one embodiment as above
Suddenly.
Power supply 901 can be AC power or DC power supply, this does not influence the realization of the application.
The function of each function module of weld defects detection system described in the embodiment of the present invention can be implemented according to the above method
Method specific implementation in example, specific implementation process are referred to the associated description of above method embodiment, no longer superfluous herein
It states.
Known by upper, the embodiment of the present invention realizes the original of the weldment to be detected acquired under the phenomenon that rebuilding based on lack sampling
Beginning the magneto optic images signal obtains complete, clear, accurate weld defect the magneto optic images, is conducive to the detection for improving weldment defect
Precision.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with it is other
The difference of embodiment, just to refer each other for same or similar part between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, depends on the specific application and design constraint of technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Above to a kind of weld defects detection method, apparatus provided by the present invention, equipment.Computer readable storage medium
And system is described in detail.Principle and implementation of the present invention are described for specific case used herein,
The explanation of above example is only intended to facilitate the understanding of the method and its core concept of the invention.It should be pointed out that for this technology
For the those of ordinary skill in field, without departing from the principle of the present invention, several improvement can also be carried out to the present invention
And modification, these improvement and modification are also fallen within the protection scope of the claims of the present invention.
Claims (10)
1. a kind of weld defects detection method, which is characterized in that including：
The original the magneto optic images signal for obtaining the weldment to be measured of discretization acquisition, utilizes canonical orthogonal basic matrix and sparse coefficient pair
The magneto optic images signal with sparsity carries out sparse variation and indicates, obtains the sparse magneto-optical signal of higher-dimension；
The sparse magneto-optical signal of the higher-dimension is projected in low-dimensional using with the canonical orthogonal basic matrix incoherent observing matrix
Space, and obtain projection value；The observing matrix and the product of the sparse coefficient meet limited equidistant condition；
Compressed sensing restructing algorithm based on Lp norms solves Sparse Optimization, restores described sparse from the projection value
The recovery coefficicnt vector of coefficient；
The original high dimensional information that the weldment to be measured is rebuild according to orthonormal basis described in the recovery coefficicnt vector sum, using as
Detect the magneto optic images signal of weld defect.
2. weld defects detection method according to claim 1, which is characterized in that it is described using canonical orthogonal basic matrix and
Sparse coefficient carries out sparse variation expression to the magneto optic images signal with sparsity：
The object element for meeting and presetting amplitude perturbations is chosen from the magneto optic images signal, and will be in the magneto optic images signal
Non-targeted element is set to 0, and obtains echo signal vector；
According to following formula, sparse variation is carried out to the echo signal vector and is indicated：
Yk=Fs；
In formula, YkFor echo signal vector；F is the canonical orthogonal basic matrix, and s is the sparse coefficient.
3. weld defects detection method according to claim 2, which is characterized in that the observing matrix is based on certainty
The certainty random matrix of random sequence construction.
4. the weld defects detection method according to claims 1 to 3 any one, which is characterized in that described to be based on Lp models
Several compressed sensing restructing algorithms solves Sparse Optimization, restores the reduction system of the sparse coefficient from the projection value
Number vector includes：
The recovery coefficicnt vector for being restored the sparse coefficient from the projection value using convex optimization base tracing algorithm, is met following
Formula：
In formula, S is recovery coefficicnt vector, and s is the sparse coefficient, and F is the canonical orthogonal basic matrix, and α is the sight
Matrix is surveyed, W is the projection value.
5. a kind of weld defects detection device, which is characterized in that including：
Sparse variation module acquires the original the magneto optic images signal of weldment to be measured for obtaining discretization, utilizes orthonormal basis
Matrix and sparse coefficient carry out sparse variation to the magneto optic images signal with sparsity and indicate, obtain the sparse magneto-optic of higher-dimension
Signal；
Low dimension projective module, for using with the incoherent observing matrix of canonical orthogonal basic matrix by the sparse magnetic of the higher-dimension
Optical signal is projected in lower dimensional space, and obtains projection value；The product of the observing matrix and the sparse coefficient meets limited etc.
Away from condition；
Sparse coefficient recovery module solves Sparse Optimization, from institute for the compressed sensing restructing algorithm based on Lp norms
State the recovery coefficicnt vector for restoring the sparse coefficient in projection value；
The magneto optic images signal reconstruction module, it is described to be measured for being rebuild according to orthonormal basis described in the recovery coefficicnt vector sum
The original high dimensional information of weldment, using the magneto optic images signal as detection weld defect.
6. weld defects detection device according to claim 5, which is characterized in that the sparse variation module includes：
Object selection unit meets the object element for presetting amplitude perturbations for being chosen from the magneto optic images signal, and will
Non-targeted element is set to 0 in the magneto optic images signal, obtains echo signal vector；
Rarefaction representation unit, for according to following formula, carrying out sparse variation to the echo signal vector and indicating：
Yk=Fs；
In formula, YkFor echo signal vector；F is the canonical orthogonal basic matrix, and s is the sparse coefficient.
7. weld defects detection device according to claim 5 or 6, which is characterized in that the sparse coefficient recovery module
To restore the recovery coefficicnt vector of the sparse coefficient from the projection value using convex optimization base tracing algorithm, and meet following
The module of formula：
In formula, S is recovery coefficicnt vector, and s is the sparse coefficient, and F is the canonical orthogonal basic matrix, and α is the sight
Matrix is surveyed, W is the projection value.
8. a kind of weld defects detection equipment, which is characterized in that including processor, the processor is deposited for executing in memory
It is realized when the computer program of storage as described in any one of Claims 1-4 the step of weld defects detection method.
9. a kind of computer readable storage medium, which is characterized in that be stored with welding on the computer readable storage medium and lack
Detection program is fallen into, realizes when the weld defects detection program is executed by processor and is welded as described in any one of Claims 1-4
The step of defect inspection method.
10. a kind of weld defects detection system, which is characterized in that including：
Magnetic field generator, alternation magneto-optical sensor, power supply and processor；
The magnetic field generator is connected with the power supply, for applying the externally-applied magnetic field for presetting magnetic induction intensity for weldment to be measured；
The alternation magneto-optical sensor is connected with the processor, the magneto optic images for acquiring the weldment to be measured, and sends
To the processor；
It is realized as described in any one of Claims 1-4 when the processor is for executing the computer program stored in memory
The step of weld defects detection method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810495731.6A CN108711151B (en) | 2018-05-22 | 2018-05-22 | Welding defect detection method, device, equipment, storage medium and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810495731.6A CN108711151B (en) | 2018-05-22 | 2018-05-22 | Welding defect detection method, device, equipment, storage medium and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108711151A true CN108711151A (en) | 2018-10-26 |
| CN108711151B CN108711151B (en) | 2022-09-13 |
Family
ID=63868344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810495731.6A Active CN108711151B (en) | 2018-05-22 | 2018-05-22 | Welding defect detection method, device, equipment, storage medium and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108711151B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104063873A (en) * | 2014-07-08 | 2014-09-24 | 华东交通大学 | Shaft sleeve part surface defect on-line detection method based on compressed sensing |
| WO2015064672A1 (en) * | 2013-11-01 | 2015-05-07 | Eizo株式会社 | Image quality improvement device, image display device, image quality improvement method, and computer program |
| CN105181122A (en) * | 2015-08-06 | 2015-12-23 | 兰州理工大学 | Mechanical vibration signal data compression and acquisition method |
| CN105741336A (en) * | 2016-03-07 | 2016-07-06 | 中国科学院上海高等研究院 | Partition compression perception imaging method based on random sampling and system |
| CN107036973A (en) * | 2017-06-07 | 2017-08-11 | 广东工业大学 | A kind of weld seam veining structure nondestructive detection system |
| CN107705342A (en) * | 2017-09-15 | 2018-02-16 | 哈尔滨工程大学 | A kind of infrared image reconstructing method based on self-adapting generalized orthogonal matching pursuit |
-
2018
- 2018-05-22 CN CN201810495731.6A patent/CN108711151B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015064672A1 (en) * | 2013-11-01 | 2015-05-07 | Eizo株式会社 | Image quality improvement device, image display device, image quality improvement method, and computer program |
| CN104063873A (en) * | 2014-07-08 | 2014-09-24 | 华东交通大学 | Shaft sleeve part surface defect on-line detection method based on compressed sensing |
| CN105181122A (en) * | 2015-08-06 | 2015-12-23 | 兰州理工大学 | Mechanical vibration signal data compression and acquisition method |
| CN105741336A (en) * | 2016-03-07 | 2016-07-06 | 中国科学院上海高等研究院 | Partition compression perception imaging method based on random sampling and system |
| CN107036973A (en) * | 2017-06-07 | 2017-08-11 | 广东工业大学 | A kind of weld seam veining structure nondestructive detection system |
| CN107705342A (en) * | 2017-09-15 | 2018-02-16 | 哈尔滨工程大学 | A kind of infrared image reconstructing method based on self-adapting generalized orthogonal matching pursuit |
Non-Patent Citations (3)
| Title |
|---|
| MARCO F. DUARTE等: "Single-Pixel Imaging via Compressive Sampling", 《IEEE SIGNAL PROCESSING MAGAZINE》 * |
| 徐光柱，雷帮军: "《实用性目标检测与跟踪算法原理与应用》", 30 June 2015, 北京：国防出版社 * |
| 李勇，等: "基于压缩感知的X 射线螺旋焊管焊缝缺陷检测", 《焊接质量控制与管理》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108711151B (en) | 2022-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103892856B (en) | A kind of method and system for obtaining human bone mineral density value | |
| Faridani et al. | Introduction to local tomography | |
| CN105510364A (en) | Nondestructive testing system for industrial part flaws based on X rays and detection method thereof | |
| Xie et al. | Artifact removal using GAN network for limited-angle CT reconstruction | |
| CN116503507A (en) | Magnetic particle image reconstruction method based on pre-training model | |
| CN113066143A (en) | Distributed quantum imaging method, device, system and computer readable storage medium | |
| CN105319225A (en) | Scanning method for achieving high-resolution large-view-field CL imaging of plate-shaped samples | |
| Almansouri et al. | Deep neural networks for non-linear model-based ultrasound reconstruction | |
| CN108711151A (en) | A kind of weld defects detection method, apparatus, equipment, storage medium and system | |
| Ferreira et al. | Non-Gaussian spectra in cosmic microwave background temperature anisotropies | |
| CN113298700A (en) | High-resolution image reconstruction method in scattering scene | |
| Chen et al. | Application of edge detection based on hexagonal image structure to delamination detection of carbon fiber reinforced polymer material | |
| Deniţiu et al. | Phase transitions and cosparse tomographic recovery of compound solid bodies from few projections | |
| Wang et al. | Effects of compressed speckle image on digital image correlation for vibration measurement | |
| Yu et al. | Efficient phase retrieval based on dark fringe extraction and phase pattern construction with a good anti-noise capability | |
| Kleman et al. | Full waveform inversion-based ultrasound computed tomography acceleration using two-dimensional convolutional neural networks | |
| CN109102552A (en) | A kind of pixel codomain filtering ultrasonic imaging method for reconstructing of non-uniform shapes constraint | |
| Sun et al. | Anatomic surface reconstruction from sampled point cloud data and prior models | |
| Li et al. | Learning to reconstruct crack profiles for eddy current nondestructive testing | |
| Daida et al. | Measuring topography of small-scale water surface waves | |
| CN113177340A (en) | Three-dimensional contact stress characterization method and device based on stereoscopic vision | |
| Chang et al. | A novel method of radiographic image enhancement based on phase symmetry | |
| CN112150570A (en) | Compressed sensing magnetic resonance imaging method based on iterative p-threshold projection algorithm | |
| Bhaskar et al. | FPGA implementation of non-subsampled Shearlet transform for image fusion | |
| TSAO | Image reconstruction using ramp response signatures |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |