CN108711151B - Welding defect detection method, device, equipment, storage medium and system - Google Patents
Welding defect detection method, device, equipment, storage medium and system Download PDFInfo
- Publication number
- CN108711151B CN108711151B CN201810495731.6A CN201810495731A CN108711151B CN 108711151 B CN108711151 B CN 108711151B CN 201810495731 A CN201810495731 A CN 201810495731A CN 108711151 B CN108711151 B CN 108711151B
- Authority
- CN
- China
- Prior art keywords
- magneto
- sparse
- optical image
- optical
- coefficient
- 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.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 117
- 238000003466 welding Methods 0.000 title claims abstract description 88
- 238000001514 detection method Methods 0.000 title claims abstract description 72
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims abstract description 116
- 238000000034 method Methods 0.000 claims abstract description 54
- 239000013598 vector Substances 0.000 claims abstract description 54
- 230000009467 reduction Effects 0.000 claims abstract description 36
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000005457 optimization Methods 0.000 claims abstract description 27
- 230000008859 change Effects 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 7
- 230000006698 induction Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 abstract description 30
- 230000009466 transformation Effects 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 description 11
- 238000012634 optical imaging Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000005284 excitation Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 239000000523 sample Substances 0.000 description 3
- 230000002238 attenuated effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000011248 coating agent Substances 0.000 description 2
- 238000000576 coating method Methods 0.000 description 2
- 230000000295 complement effect 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
- 238000003384 imaging method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 229910044991 metal oxide Inorganic materials 0.000 description 2
- 150000004706 metal oxides Chemical class 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000001143 conditioned effect 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
- 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
- 238000009659 non-destructive testing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000012788 optical film Substances 0.000 description 1
- -1 pits Substances 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 230000008719 thickening Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/06—Topological mapping of higher dimensional structures onto lower dimensional surfaces
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biochemistry (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Analytical Chemistry (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
Abstract
The embodiment of the invention discloses a welding defect detection method, a welding defect detection device, welding defect detection equipment, a computer readable storage medium and a computer readable storage system. The method comprises the steps of sparsely representing a magneto-optic image signal of a sparse weldment to be detected by using a normalized orthogonal basis matrix and a sparse coefficient to obtain a sparse magneto-optic signal; observing the sparse coefficient by utilizing an observation matrix irrelevant to the orthonormal basis matrix, and projecting the high-dimensional sparse magneto-optical signal obtained by transformation in a low-dimensional space to obtain a projection value; solving a sparse optimization problem by a Lp norm-based compressed sensing reconstruction algorithm, and recovering a reduction coefficient vector of a sparse coefficient from a projection value; and reconstructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normative orthogonal basis to serve as an original image signal for detecting the weld defects. According to the technical scheme, the magneto-optical image signal of the weldment to be detected, which is acquired based on the under-sampling phenomenon, is reconstructed, a complete, clear and accurate magneto-optical image of the weld defects is obtained, and the detection precision of the weldment defects is favorably improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of laser welding detection, in particular to a welding defect detection method, a device, equipment, a computer readable storage medium and a system.
Background
With the development of the aviation, aerospace, energy and ship industries in China, the welding technology has become one of the indispensable material forming and processing technologies in the manufacturing industry. Since the welding process is susceptible to factors from the external environment during the processing, such as welding speed, welding power, shielding gas flow, welding environment, workpiece surface condition, etc., welding defects such as cracks, incomplete penetration, incomplete fusion, pores, pits, slag inclusions, etc., are inevitably generated on the weldment, and the defects may cause catastrophic accidents. In order to ensure the product quality of the weldment, the defects on the surface and inside of the welded workpiece must be timely and effectively detected. In the actual production process, besides visual inspection of welding surface defects and molding defects, some weld defects smaller than 0.1mm can not be effectively identified by visual inspection.
Magneto-optical imaging is widely applied to the welding seam detection technology of weldments as a nondestructive detection method. A schematic diagram of a magneto-optical imaging method based on the magneto-optical effect is shown in fig. 1. Based on Faraday magneto-optical effect, a light source is converted into a beam of polarized light through a polarizer, the polarized light passes through a magneto-optical medium and is reflected by a magneto-optical film reflecting surface, the polarizing surface deflects due to the existence of an external magnetic field, the polarized light deflected by a certain angle is collected by a CMOS (complementary metal oxide semiconductor) through an analyzer, and the information of the change of the magnetic field is converted into real-time imaging of the change of light intensity.
FIG. 2 is a magneto-optical imaging working principle diagram of a weld joint, based on the Faraday magneto-optical rotation principle, an alternating current power supply supplies alternating current to an electromagnet and generates an alternating magnetic field. The light source LED generates linearly polarized light after passing through the polarizer, the magneto-optical probe consists of a magneto-optical medium and a mirror coating, incident linearly polarized light is reflected by the mirror coating after passing through the magneto-optical medium and then passes through the magneto-optical medium again, the analyzer detects a corresponding light signal and images the light signal by the CMOS camera, and information of magnetic field change is converted into real-time imaging of light intensity.
According to the traditional magneto-optical imaging nondestructive detection method, a magneto-optical sensor is used for acquiring a time domain signal, then a Fourier transform is used for reconstructing a space domain signal to obtain an internal image of a weldment, and the Fourier transform is linear transformation, so that the number of k domain signals required to be acquired must be equal to the number of pixels of an image domain. According to Shannon-Nay stettes sampling theorem, when the maximum sampling frequency of the magneto-optical sensor is more than twice the frequency of the alternating current excitation signal, the acquired magneto-optical image information sequence completely keeps the weld defect information of the material in the alternating signal. In practical application, the frequency of an applied alternating excitation signal is 50Hz, the sampling frequency range of the magneto-optical sensor is 0-75 Hz, it can be seen that the maximum sampling frequency of the magneto-optical sensor does not meet the condition of the sampling theorem, an undersampling phenomenon can occur, the frequency spectrums of magneto-optical image information are overlapped, namely, a frequency component higher than half of the sampling frequency is reconstructed into a signal lower than half of the sampling frequency, so that an aliasing phenomenon during reconstruction is caused, complete and clear magneto-optical image information of weld defects is difficult to acquire, and whether defects exist in welding or not can not be accurately judged.
The phenomenon of limited detection of the weld defects is avoided, and a metal defect nondestructive detection technology adopted in the prior art is a ray detection method, and particularly relates to a method for detecting the internal defects of a weldment according to the intensity of rays (such as x rays, gamma rays and the like) attenuated by passing through various parts of the weldment to be detected, wherein the rays have a certain attenuation rule in the process of passing through the weldment to be detected. The attenuation level varies from object to object, and is determined by the thickness of the object, the type of material of the object, and the type of radiation.
The ray detection method is mainly used for detecting the internal volume type defects of the workpiece, the thickness of the workpiece is not easy to exceed 80mm, and corresponding thickening or thinning can be performed according to the attenuation coefficient of the material. The method has the advantages of high detection cost, large detection equipment, great damage to a human body by the generated ray radiation and low detection sensitivity to the microcrack defect.
In view of this, how to implement nondestructive testing of welding defects based on magneto-optical image information of an undersampling phenomenon is an urgent problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the invention aims to provide a welding defect detection method, a welding defect detection device, a welding defect detection equipment, a computer readable storage medium and a welding defect detection system, which are used for reconstructing an original magneto-optical image signal of a weldment to be detected, which is acquired based on an under-sampling phenomenon, obtaining a complete, clear and accurate magneto-optical image of a welding seam defect and being beneficial to improving the detection precision of the welding seam defect.
In order to solve the above technical problems, embodiments of the present invention provide the following technical solutions:
one aspect of the embodiments of the present invention provides a welding defect detection method, including:
obtaining an original magneto-optical image signal of a weldment to be detected, which is acquired in a discretization mode, and performing sparse change expression on the magneto-optical image signal with sparsity by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal;
projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the orthonormal basis matrix, and obtaining a projection value; the product of the observation matrix and the sparse coefficient meets a finite equidistant condition;
solving a sparse optimization problem by a Lp norm-based compressed sensing reconstruction algorithm, and recovering a reduction coefficient vector of the sparse coefficient from the projection value;
and establishing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normalized orthogonal basis weight to serve as a magneto-optical image signal for detecting the defects of the welding seam.
Optionally, the performing sparse change representation on the magneto-optical image signal with sparsity by using the orthonormal basis matrix and the sparse coefficient includes:
selecting a target element meeting a preset amplitude condition from the magneto-optical image signal, and setting a non-target element in the magneto-optical image signal to be 0 to obtain a target signal vector;
and performing sparse change representation on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is the target signal vector; f is the orthonormal basis matrix and s is the sparse coefficient.
Optionally, the observation matrix is a deterministic random matrix constructed based on a deterministic random sequence.
Optionally, the interpolating method based on compressed sensing solves a sparse optimization problem, and recovering a reduction coefficient vector of the sparse coefficient from the projection value includes:
recovering a reduction coefficient vector of the sparse coefficient from the projection value by using a convex optimization basis tracking algorithm, and satisfying the following formula:
wherein S is the reduction coefficient vector, S is the sparse coefficient, F is the orthonormal basis matrix, α is the observation matrix, and W is the projection value.
Another aspect of an embodiment of the present invention provides a welding defect detection apparatus, including:
the sparse change module is used for acquiring an original magneto-optical image signal of a weldment to be detected in a discretization mode, and performing sparse change representation on the magneto-optical image signal with sparsity by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal;
the low-dimensional projection module is used for projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by utilizing an observation matrix irrelevant to the orthonormal basis matrix and obtaining a projection value; the product of the observation matrix and the sparse coefficient meets a finite equidistant condition;
the sparse coefficient reduction module is used for solving a sparse optimization problem based on a Lp norm compressed sensing reconstruction algorithm and recovering a reduction coefficient vector of the sparse coefficient from the projection value;
and the magneto-optical image signal reconstruction module is used for constructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the standard orthogonal basis so as to serve as a magneto-optical image signal for detecting the weld defects.
Optionally, the sparse change module includes:
the target selection unit is used for selecting a target element meeting a preset amplitude value condition from the magneto-optical image signal and setting a non-target element in the magneto-optical image signal to be 0 to obtain a target signal vector;
a sparse representation unit, configured to perform sparse change representation on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is the target signal vector; f is the orthonormal basis matrix and s is the sparse coefficient.
Optionally, the sparse coefficient reduction module is a module that recovers a reduction coefficient vector of the sparse coefficient from the projection value by using a convex optimization basis pursuit algorithm and satisfies the following formula:
wherein S is the reduction coefficient vector, S is the sparse coefficient, F is the orthonormal basis matrix, α is the observation matrix, and W is the projection value.
An embodiment of the present invention further provides a welding defect detection apparatus, which includes a processor, and the processor is configured to implement the steps of the welding defect detection method according to any one of the preceding items when executing the computer program stored in the memory.
The embodiment of the invention also provides a computer readable storage medium, wherein a welding defect detection program is stored on the computer readable storage medium, and when being executed by a processor, the welding defect detection program realizes the steps of the welding defect detection method in any one of the previous items.
The embodiment of the present invention finally provides a welding defect detection system, including:
the device comprises a magnetic field generator, an alternating magneto-optical sensor, a power supply and the welding defect detection equipment;
the magnetic field generator is connected with the power supply and is used for applying an external magnetic field with preset magnetic induction intensity to the weldment to be detected;
the alternating magneto-optical sensor is connected with the welding defect detection equipment and used for acquiring magneto-optical images of the weldment to be detected and sending the magneto-optical images to the processor;
the processor is configured to implement the steps of the welding defect detection method according to any one of the preceding 4 when executing the computer program stored in the memory.
The embodiment of the invention provides a welding defect detection method, which comprises the steps of carrying out sparse change representation on a high-dimensional original magneto-optical image signal with sparsity of a weldment to be detected by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal; projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the normative orthogonal basis matrix, and obtaining a projection value; the product of the observation matrix and the sparse coefficient meets the finite equidistant condition; solving a sparse optimization problem by a Lp norm-based compressed sensing reconstruction algorithm, and recovering a reduction coefficient vector of a sparse coefficient from a projection value; and reconstructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normative orthogonal basis to serve as a magneto-optical image signal for detecting the weld defects.
The technical scheme provided by the application has the advantages that based on the fact that the magnetic-optical image information of the under-sampled weld defect is sparse (or compressible) in a space domain, a time domain or a certain specific domain, the magnetic-optical signals of a high-dimensional space can be projected to a low-dimensional space by using an observation matrix, the projection values of the small-amount space magnetic-optical signals contain enough information for reconstructing the weld defect signals, and the weld defect image is accurately reconstructed at a high probability by solving an optimization problem. The magneto-optical image reconstruction method can be used for completely visualizing the detection of the surface and the internal defects of the welding piece to be detected under the condition that the sampling frequency of the magneto-optical sensor is not more than the frequency of the double alternating current excitation signal, and the welding piece is not damaged, so that the reconstruction of the original magneto-optical image signal of the welding piece to be detected, which is acquired under the undersampling phenomenon, is realized, a complete, clear and accurate magneto-optical image of the welding piece defect is obtained, the magneto-optical image is used as the magneto-optical information for subsequently detecting the defect type or whether the defect exists in the welding piece to be detected, the nondestructive detection of the defect of the welding piece is realized, the detection precision of the defect of the welding piece is favorably improved, and the magneto-optical image reconstruction method has very important significance for ensuring the performance and the service life of the structure of the welding piece.
In addition, the embodiment of the invention also provides a corresponding implementation device, equipment and system for the welding defect detection method, so that the method has higher practicability, and the device, the equipment and the system have corresponding advantages.
Drawings
In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
Fig. 1 is a schematic diagram illustrating a principle of a magneto-optical imaging method based on a magneto-optical effect according to an embodiment of the present invention;
FIG. 2 is a magneto-optical imaging operational diagram of a weld according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of a welding defect detection method according to an embodiment of the present invention;
fig. 4 is a reduction diagram of a single-frame dynamic magneto-optical grayscale map of a weld surface crack at a sampling frequency according to the technical solution of the present application provided in the embodiment of the present invention;
FIG. 5 is a single-frame dynamic magneto-optical grayscale map of weld surface cracks at a sampling frequency according to an embodiment of the present invention;
FIG. 6 is a restored image of a single-frame dynamic magneto-optical grayscale image of a weld surface crack at another sampling frequency according to the technical solution of the present application;
FIG. 7 is a single frame dynamic magneto-optical grayscale map of weld surface cracks at another sampling frequency provided by an embodiment of the present invention;
fig. 8 is a structural diagram of a specific embodiment of a welding defect detection apparatus according to an embodiment of the present invention;
fig. 9 is a structural diagram of a specific embodiment of a welding defect detection system according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The terms "including" and "having," and any variations thereof, in the description and claims of this application and the drawings described above, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may include other steps or elements not expressly listed.
Having described the technical solutions of the embodiments of the present invention, various non-limiting embodiments of the present application are described in detail below.
Referring to fig. 3, fig. 3 is a schematic flow chart of a welding defect detection method according to an embodiment of the present invention, where the embodiment of the present invention includes the following:
s301: the method comprises the steps of obtaining an original magneto-optical image signal of a weldment to be detected, wherein the original magneto-optical image signal is acquired in a discretization mode, and performing sparse change representation on the magneto-optical image signal with sparsity by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal.
The magneto-optical image signal is a magneto-optical image acquired by a magneto-optical sensor on the surface and inside of the welding seam of the weldment, the more discontinuous the acquired magneto-optical signal is, the better the signal is, the more diffuse the signal is, and the low relevance of the data is ensured. The magneto-optical sensor can collect information at any sampling frequency, which is not limited in this application.
The original magneto-optical image signal is a magneto-optical image acquired by a magneto-optical sensor of the weldment to be measured.
The precondition of the compressed sensing algorithm is that the original information must be sparse and can be sparsely represented in a space domain, a frequency domain or other domains. The original magneto-optical image signal acquired by the magneto-optical sensor can be sparsely represented in a space domain, a frequency domain, a wavelet domain or other domains, the original magneto-optical image signal can be represented by Y, Y is a finite-length one-dimensional discrete time signal, and the magneto-optical image signal can be sparsely represented by using a normalized orthogonal basis matrix and a sparse coefficient, namely:
Y=Fs;
wherein Y is an original magneto-optical image signal; f is a normalized orthogonal basis matrix and s is a sparse coefficient.
In the above formula, the orthonormal basis matrix F is a sparse basis matrix, s is an equivalent representation of Y in the F domain, if the non-zero number of the sparse matrix is k, s is k-sparse, in general, the expansion coefficient s of Y in the orthonormal basis matrix F is exponentially attenuated by a certain magnitude, the sparse matrix s is not strictly sparse, has very few large coefficients and many small coefficients, the small coefficients are close to zero, and the small coefficients can be approximated to zero, and compression is achieved by using a transformation that a large-amplitude (for example, an amplitude larger than 6) target element is retained from a magneto-optical image signal containing global information of an original signal, and a non-target element (a small-amplitude element, for example, an amplitude close to 0) in the magneto-optical image signal is set to 0, each target element contains information of a full image to some extent, and in this case, taking off a part of acquired data does not result in permanent loss of an information portion (they are still lost by a part of the image (they are set by a small-amplitude element) in the magneto-optical image signal Contained in other data) to obtain a target signal vector;
and performing sparse change expression on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is a target signal vector; f is a normalized orthogonal basis matrix and s is a sparse coefficient.
S302: and projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the normal orthogonal basis matrix, and obtaining a projection value.
Based on the basic idea of the compressed sensing theory, if the signal is sparse (or compressible) in the time-space domain or some transform domain, the transformed high-dimensional signal can be projected onto a low-dimensional space by using an observation matrix unrelated to the transform basis, and then the original signal can be reconstructed from the small number of projections with high probability by solving an optimization problem.
The observation matrix is used for sampling an original magneto-optical image signal, the sensing matrix A is a matrix actually acting on a coefficient vector with sparsity, the sensing matrix is a true observation matrix from the perspective of compressed sensing, and in order to ensure that the sensing matrix maps two different sparse signals into the same set (a one-to-one mapping relation from an original space to a low-dimensional space is ensured), the product of the observation matrix and the sparse coefficient needs to meet a finite equidistant condition.
The observation matrix may be a deterministic random matrix constructed based on a deterministic random sequence, such as a random gaussian measurement matrix, and a different matrix unrelated to the orthonormal basis matrix F may be obtained as the observation matrix by changing the sampling frequency of the magneto-optical sensor and the alternating frequency of the alternating current excitation device.
In order to obtain different information with as much original information as possible when the sparse matrix is reduced from a high dimension to a low dimension by using the observation matrix, it is necessary that the correlation of each column vector of the observation matrix is small. When a deterministic random observation matrix is constructed by utilizing a deterministic random sequence, random measurement is an optimal strategy for a sparse matrix, the original information of a sparse coefficient s can be recovered by almost the fewest measurements, and a constant required in analysis is very small. The random gaussian measurement matrix is the most commonly used measurement matrix in compressed sensing research, and elements in the matrix are subject to normal distribution when the mean value is zero and the variance is 1/M, and are independent of each other, that is:
α ij m is the number of rows constituting the observation matrix.
Whether the observation matrix can be used for compressed sensing is conditioned on the fact that a limited isometry must be satisfied. The Gaussian random measurement matrix has the advantages that the Gaussian random measurement matrix is almost irrelevant to any sparse matrix, the limited equidistant condition is met, all column subsets of the observation matrix are required to be approximately orthogonal under the limited equidistant condition, the property ensures that the sparse coefficient cannot exist in the space of the observation matrix after the observation matrix projects the sparse coefficient (namely after observation), and the projection cannot cause confusion when two different sparse coefficients are projected to a unified low-dimensional measurement value. At the same time, the number of measurement values required by the method is small, for example, for the original data with the length of N and the sparsity of K, only the original data is requiredThe individual measurements allow a high probability of recovering the reconstructed original data, where c is a very small constant.
Given the following function:
wherein p (T) is a periodic function, T is the period of the periodic function, r and Z are real numbers respectively, p (rT) is an initial value, and M and N are the number of rows and columns of the observation matrix; formula (1) is known to produce sequences of arbitrary length: x is the number of 1 ,x 2 ,……x n The next value of the observation sequence is x n+1 (ii) a Consider a compound of formula (2):
x n+1 =f(af -1 (x n )); (2a)
y n =f(bf -1 (x n ))); (2b)
generation of the sequence y from formula (2) n Initial value x 0 Randomly taking the sequence in the range of [ -1, 1), constructing the sequence into a matrix according to a row or column priority principle, and randomly taking the sequence y n Constructing an observation matrix alpha according to a priority principle 1 :
C 1 For observing the matrix alpha 1 Sum of squares of all elements in (1), coefficientHas a normalizing function to enable z n =y n 0.5, constructing the sequence into a matrix according to the principle of row or column priority, and constructing the random sequence z n Constructing an observation matrix alpha according to a priority principle:
coefficient of performancePlays a role of normalization, and the observation matrix must satisfy finite equidistanceThe property (RIP) that all column subsets of the observation matrix are approximately orthogonal is required by the finite equidistant condition, and the property ensures that the sparse coefficient does not exist in the observation matrix space after the observation matrix projects the sparse coefficient (namely after observation), and the projection does not cause confusion when two different projections are projected on a uniform low-dimensional measurement value. For any k sparse coefficient, if there is a constantSo thatIt is true that the first and second sensors,is an arbitrary constant in the range of [0,1 ], x is a sparse signal (i.e. a sparse coefficient S signal in the application), and α is an observation matrix; the observation matrix is considered to satisfy the finite equidistant property and can be used for measuring the sparse matrix in the compressed sensing.
Projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the normative orthogonal basis matrix, and obtaining a projection value, namely W is alpha Y; f is a normal orthogonal basis matrix, alpha is an observation matrix, and W is a projection value.
S303: and solving a sparse optimization problem by using a Lp norm-based compressed sensing reconstruction algorithm, and recovering a reduction coefficient vector of a sparse coefficient from the projection value.
The Lp norm-based compressive sensing reconstruction algorithm can include a greedy algorithm, an orthogonal matching algorithm, a convex optimization algorithm and the like, and specifically which algorithm is adopted is not limited in any way in the application.
And recovering a reduction coefficient vector of the sparse coefficient from the projection value by using a convex optimization basis tracking algorithm, and satisfying the following formula:
in the formula, S is a reduction coefficient vector, S is a sparse coefficient, F is a normalized orthogonal basis matrix, alpha is an observation matrix, and W is a projection value.
In a specific embodiment, the above process may be:
s ═ s for sparse coefficients 1 ,s 2 ,…,s n The p-norm of the } is:
the reconstruction of the compressed sensing is a process of recovering an original signal from a sparse coefficient, because the observation quantity is far less than the signal length, the reconstruction is to solve an underdetermined equation problem, when the original magneto-optical image signal is sparse or compressible, and the 2K-order equidistant constant of a sensing matrix A is less than 1, the reconstruction of the compressed sensing signal can be converted into a non-convex optimization problem solution restricting 0-norm minimization, and the problem of solving the underdetermined equation can be converted into a minimum 0-norm problem. Due to Y k If Fs, then the sparse coefficient s is F T Y k The unified expression is: (s.t. is in the sense of "subject to" -, i.e. "abbreviation for subject to")
min||F T Y k || p s.t.W=αFY k =AY k ,p=0,1;
Wherein s is a sparse coefficient and is a transposed matrix of the sparse coefficient, F is an orthonormal basis matrix, alpha is an observation matrix, W is a projection value, Y is a zero-crossing coefficient k Is the original signal with sparseness.
When p is 0, a 0-norm results, which effectively represents the number of non-zero terms in the sparse coefficient. Then, solving the equation set W As, a is a sensing matrix, i.e. the problem of finding the projection value (observation set) of the original magneto-optical image signal in the low-dimensional space is transformed into the minimum 0-norm problem:
min||s|| 0 s.t.W=As;
the signals are accurately reconstructed by solving the 0-norm minimization problem, and the vectors conforming to the equation W As known from the statistical theory and the combination optimization theory are many, and the sparsest one is the original magneto-optical image signal to be found. Due to the sparse coefficients (length N, sparsity ofK) In a linear combination of all non-zero term positionsAnd (3) the optimal solution can be obtained only by listing one by one, which is very difficult. When the equal distance of the sensing matrix A meets the RIP condition, the non-convex 0-norm minimization and the relaxed 1-norm minimization are equivalent, the non-convex 0-norm minimization problem is relaxed into a convex optimization problem, and the convex optimization is a mathematical optimization problem for researching the convex function minimization problem. The local minimum value in the convex optimization model is a global minimum value, and the set of the global minimum values is a convex set, so that the reconstruction problem of compressed sensing can be solved by using the property.
Solving a convex optimization problem, wherein the most representative method is a 1-norm minimization method-BP algorithm (also called basis pursuit), and a mathematical model of the method is as follows:
min||s|| 1 s.t.W=As;
by solving for p 1 The projection value data can be obtained accurately by minimizing the medium 1-norm, and the result approximation effect is better. The projection value is length M as a measured value, so that the problem of compressed sensing is to solve an underdetermined equation W-alpha Y to obtain an original magneto-optical image signal Y on the basis of a known measured value and an observation matrix, and then the final equation becomes a problem of 1-norm minimization of W-alpha Fs, W, alpha and F are known, and s is solved.
S304: and reconstructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normalized orthogonal basis to serve as a magneto-optical image signal for detecting the weld defects.
After obtaining the reduction coefficient vector, since in S301, the original magneto-optical signal is sparsely changed, that is, Y is Fs, when reconstruction is performed, after obtaining the reduction coefficient obtained by recovering the sparse coefficient, the reconstructed original magneto-optical signal, that is, the original magneto-optical signal that is reduced may be Y 'Fs, S is the reduction coefficient vector, and Y' is the reconstructed original magneto-optical signal.
The information of spectrum signal aliasing artifacts of the original magneto-optical signal in the under-sampling phenomenon is restored based on a compressive sensing algorithm, and a complete and clear magneto-optical image of the weldment to be detected is obtained.
And the restored or reconstructed magneto-optical image information is used as a sample image for detecting the welding seam of the weldment, and the sample image is identified by adopting an image identification algorithm (such as a fuzzy clustering algorithm) to determine whether the weldment to be detected has defects or not, and further identify the defects as cracks, pits, incomplete fusion or incomplete penetration and the like when the defects exist.
In the technical scheme provided by the embodiment of the invention, based on that the under-sampled magneto-optical image information of the weld defect is sparse (or compressible) in a space domain, a time domain or a certain specific domain, a magneto-optical signal in a high-dimensional space can be projected into a low-dimensional space by using an observation matrix, the projection values of the magneto-optical signals in a small amount of space contain enough information for reconstructing the weld defect signal, and the weld defect image is accurately reconstructed at a high probability by solving an optimization problem. The magneto-optical image reconstruction method can be used for completely visualizing the detection of the surface and the internal defects of the welding piece to be detected under the condition that the sampling frequency of the magneto-optical sensor is not more than the frequency of the double alternating current excitation signal, and the welding piece is not damaged, so that the reconstruction of the original magneto-optical image signal of the welding piece to be detected, which is acquired under the undersampling phenomenon, is realized, a complete, clear and accurate magneto-optical image of the welding piece defect is obtained, the magneto-optical image is used as the magneto-optical information for subsequently detecting the defect type or whether the defect exists in the welding piece to be detected, the nondestructive detection of the defect of the welding piece is realized, the detection precision of the defect of the welding piece is favorably improved, and the magneto-optical image reconstruction method has very important significance for ensuring the performance and the service life of the structure of the welding piece.
In addition, in order to further prove that the technical solution of the present application has obvious effects, the present application also provides specific examples, please refer to fig. 4-7, which are single-frame dynamic magneto-optical grayscale graphs of weld surface cracks at different sampling frequencies, which specifically may include:
in order to verify the effectiveness of the technical scheme, the information acquisition is carried out on the surface crack of the welded workpiece by adopting the working principle of magneto-optical imaging of the welded workpiece as shown in figure 2, the resolution of a CMOS (complementary metal oxide semiconductor) of a magneto-optical sensor is 400pixel x 400pixel, the applied alternating frequency is 50Hz, the excitation voltage is 220V, different sampling frequencies of the magneto-optical sensor are changed, and the magneto-optical image of the surface crack of the welded workpiece reconstructed by the technical scheme and the magneto-optical image of the surface crack of the welded workpiece undersampled by the magneto-optical imaging are obtained. Fig. 4 and 5 are gray scale graphs of cracks on the surface of a weld joint when the sampling frequency is 75Hz, wherein fig. 4 is a restoration graph recovered by adopting the technical scheme of the application, fig. 5 is a magneto-optical gray scale graph under undersampling, fig. 6 and 7 are gray scale graphs of cracks on the surface of the weld joint at the same position when the sampling frequency is 55Hz, wherein fig. 6 is the restoration graph recovered by adopting the technical scheme of the application, and fig. 7 is the magneto-optical gray scale graph under undersampling.
As can be seen, the clarity and completeness of fig. 4 and 6 are respectively stronger than those of fig. 5 and 7, which proves the effectiveness of the solution of the present application in restoring the original magneto-optical image information.
According to the method, an induction magnetic field is generated at the welding seam by adopting an alternating-current excitation field with changeable frequency, a magneto-optical imaging method is used for collecting magneto-optical signals of the welding seam defects, based on a compressive sensing theory, the sparse compressibility of the signals is utilized to represent the signals sparsely in a space domain, the whole original signals are reconstructed from data with less sampling than the traditional magneto-optical signals, more accurate visual information of the welding seam defects can be obtained within the same sampling time, and the nondestructive detection of the welding seam defects of the weldment is realized. The method and the device overcome the problem of undersampling of signals acquired by a magneto-optical imaging method in laser welding, can shorten the data scanning time of the magneto-optical sensor, perform sparse representation on the original magneto-optical signals, effectively reduce the generation of aliasing artifacts, and improve the image reconstruction quality. In addition, the acquisition and compression of the magneto-optical image of the weld defects of the weldment are carried out at a low speed, and the signal recovery process is an optimized calculation process, so that the sampling and calculation cost of the magneto-optical sensor is greatly reduced. Because the frequency of the alternating current excitation device is adjustable, magneto-optical information of the weld defects in different magnetic field intensity directions can be obtained, and the detection efficiency of the weld defects is improved.
The embodiment of the invention also provides a corresponding implementation device for the welding defect detection method, so that the method has higher practicability. In the following, the welding defect detecting apparatus provided by the embodiment of the present invention is introduced, and the welding defect detecting apparatus described below and the welding defect detecting method described above may be referred to correspondingly.
Referring to fig. 8, fig. 8 is a structural diagram of a welding defect detection apparatus according to an embodiment of the present invention, in which the apparatus may include:
and the sparse change module 801 is used for acquiring the discretely acquired original magneto-optical image signals of the weldment to be detected, and performing sparse change representation on the magneto-optical image signals with sparsity by using the canonical orthogonal basis matrix and the sparse coefficient to obtain high-dimensional sparse magneto-optical signals.
The low-dimensional projection module 802 is configured to project the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the canonical orthogonal basis matrix, and obtain a projection value; the product of the observation matrix and the sparse coefficient satisfies a finite equidistant condition.
And the sparse coefficient reduction module 803 is configured to solve a sparse optimization problem based on a Lp norm compressed sensing reconstruction algorithm, and recover a reduction coefficient vector of a sparse coefficient from the projection value.
And the magneto-optical image signal reconstruction module 804 is used for reconstructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normative orthogonal basis to serve as a magneto-optical image signal for detecting the weld defects.
Optionally, in some implementations of this embodiment, the sparse change module 801 may include:
the target selection unit is used for selecting a target element meeting a preset amplitude value condition from the magneto-optical image signal and setting a non-target element in the magneto-optical image signal to be 0 to obtain a target signal vector;
a sparse representation unit, configured to perform sparse change representation on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is a target signal vector; f is a normalized orthogonal basis matrix and s is a change domain representation of the target signal vector.
Optionally, in another implementation manner of this embodiment, the sparse coefficient restoring module 803 is a module that recovers a restoration coefficient vector of a sparse coefficient from a projection value by using a convex optimization basis pursuit algorithm, and satisfies the following formula:
in the formula, S is a reduction coefficient vector, S is a sparse coefficient, F is a normalized orthogonal basis matrix, alpha is an observation matrix, and W is a projection value.
The functions of the functional modules of the welding defect detection apparatus in the embodiment of the present invention may be specifically implemented according to the method in the above method embodiment, and the specific implementation process may refer to the related description of the above method embodiment, which is not described herein again.
Therefore, the method and the device have the advantages that the original magneto-optical image signals of the weldment to be detected, which are acquired based on the under-sampling phenomenon, are reconstructed, the complete, clear and accurate magneto-optical image of the weld defects is obtained, and the detection precision of the weldment defects is improved.
An embodiment of the present invention further provides a welding defect detection apparatus, which may include:
a memory for storing a computer program;
a processor for executing a computer program to implement the steps of the welding defect detection method according to any of the above embodiments.
The functions of the functional modules of the welding defect detection device according to the embodiments of the present invention may be specifically implemented according to the method in the above method embodiments, and the specific implementation process may refer to the related description of the above method embodiments, which is not described herein again.
Therefore, the method and the device have the advantages that the original magneto-optical image signals of the weldment to be detected, which are acquired based on the under-sampling phenomenon, are reconstructed, the complete, clear and accurate magneto-optical image of the weld defects is obtained, and the detection precision of the weldment defects is improved.
Finally, an embodiment of the present invention provides a computer-readable storage medium, in which a welding defect detection program is stored, and the steps of the welding defect detection method according to any one of the above embodiments are performed by a processor in the face recognition program.
The functions of the functional modules of the computer-readable storage medium according to the embodiment of the present invention may be specifically implemented according to the method in the foregoing method embodiment, and the specific implementation process may refer to the related description of the foregoing method embodiment, which is not described herein again.
Therefore, the method and the device have the advantages that the original magneto-optical image signals of the weldment to be detected, which are acquired based on the under-sampling phenomenon, are reconstructed, the complete, clear and accurate magneto-optical image of the weld defects is obtained, and the detection precision of the weldment defects is improved.
An embodiment of the present invention further provides a welding defect detecting system, please refer to fig. 9, which may include:
a power supply 901, a magnetic field generator 902, a weldment to be tested 903, an alternating magneto-optical sensor 904 and a processor 905;
the magnetic field generator 902 is connected with the power supply 901 and is used for applying an external magnetic field with preset magnetic induction intensity to the weldment 903 to be tested.
The alternating magneto-optical sensor 904 is connected with the processor 905, and is used for acquiring a magneto-optical image of the weldment 903 to be tested and sending the magneto-optical image to the processor 905.
The processor 905 executes a computer program to implement the steps of the weld defect detection method according to any one of the above embodiments.
The power supply 901 may be an ac power supply or a dc power supply, which does not affect the implementation of the present application.
The functions of the functional modules of the welding defect detection system according to the embodiment of the present invention may be specifically implemented according to the method in the above method embodiment, and the specific implementation process may refer to the related description of the above method embodiment, which is not described herein again.
Therefore, the method and the device have the advantages that the original magneto-optical image signals of the weldment to be detected, which are acquired based on the under-sampling phenomenon, are reconstructed, the complete, clear and accurate magneto-optical image of the weld defects is obtained, and the detection precision of the weldment defects is improved.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed in the embodiment corresponds to the method disclosed in the embodiment, so that the description is simple, and the relevant points can be referred to the description of the method part.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The invention provides a welding defect detection method, a welding defect detection device and welding defect detection equipment. Computer-readable storage media and systems are described in detail. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (8)
1. A welding defect detection method is characterized by comprising the following steps:
the method comprises the steps of obtaining an original magneto-optical image signal of a weldment to be detected, wherein the original magneto-optical image signal is acquired in a discretization mode, and sparse change representation is carried out on the magneto-optical image signal with sparsity by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal;
projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by using an observation matrix irrelevant to the orthonormal basis matrix, and obtaining a projection value; the product of the observation matrix and the sparse coefficient satisfies a finite equidistant condition;
solving a sparse optimization problem by a Lp norm-based compressed sensing reconstruction algorithm, and recovering a reduction coefficient vector of the sparse coefficient from the projection value;
establishing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the standard orthogonal basis weight to serve as a magneto-optical image signal for detecting the weld defects;
wherein the Lp norm-based compressive sensing reconstruction algorithm solves a sparse optimization problem, and recovering a reduction coefficient vector of the sparse coefficient from the projection value comprises:
recovering a reduction coefficient vector of the sparse coefficient from the projection value by using a convex optimization basis tracking algorithm, and satisfying the following formula:
wherein S is the reduction coefficient vector, S is the sparse coefficient, F is the orthonormal basis matrix, α is the observation matrix, and W is the projection value.
2. The weld defect detection method of claim 1, wherein the sparsely varying representation of the sparsely populated magneto-optical image signals using orthonormal basis matrices and sparsity coefficients comprises:
selecting a target element meeting a preset amplitude condition from the magneto-optical image signal, and setting a non-target element in the magneto-optical image signal to be 0 to obtain a target signal vector;
performing sparse change representation on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is the target signal vector; f is the orthonormal basis matrix and s is the sparse coefficient.
3. The weld defect detection method of claim 2, wherein the observation matrix is a deterministic random matrix constructed based on a deterministic random sequence.
4. A welding defect detection device, comprising:
the sparse change module is used for acquiring an original magneto-optical image signal of a weldment to be detected in a discretization mode, and performing sparse change representation on the magneto-optical image signal with sparsity by utilizing a normalized orthogonal basis matrix and a sparse coefficient to obtain a high-dimensional sparse magneto-optical signal;
the low-dimensional projection module is used for projecting the high-dimensional sparse magneto-optical signal in a low-dimensional space by utilizing an observation matrix irrelevant to the orthonormal basis matrix and obtaining a projection value; the product of the observation matrix and the sparse coefficient meets a finite equidistant condition;
the sparse coefficient reduction module is used for solving a sparse optimization problem based on a Lp norm compressed sensing reconstruction algorithm and recovering a reduction coefficient vector of the sparse coefficient from the projection value;
the magneto-optical image signal reconstruction module is used for constructing original high-dimensional information of the weldment to be detected according to the reduction coefficient vector and the normalized orthogonal basis weight to serve as a magneto-optical image signal for detecting the weld defects;
the sparse coefficient restoration module is a module which recovers a restoration coefficient vector of the sparse coefficient from the projection value by using a convex optimization basis tracking algorithm and satisfies the following formula:
wherein S is the reduction coefficient vector, S is the sparse coefficient, F is the orthonormal basis matrix, α is the observation matrix, and W is the projection value.
5. The welding defect detection device of claim 4, wherein the sparse variation module comprises:
the target selection unit is used for selecting a target element meeting a preset amplitude value condition from the magneto-optical image signal and setting a non-target element in the magneto-optical image signal to be 0 to obtain a target signal vector;
a sparse representation unit, configured to perform sparse change representation on the target signal vector according to the following formula:
Y k =Fs;
in the formula, Y k Is the target signal vector; f is the orthonormal basis matrix and s is the sparse coefficient.
6. A welding defect detection apparatus comprising a processor for implementing the steps of the welding defect detection method of any one of claims 1 to 3 when executing a computer program stored in a memory.
7. A computer-readable storage medium, having a welding defect detection program stored thereon, which when executed by a processor implements the steps of the welding defect detection method of any one of claims 1 to 3.
8. A welding defect detection system, comprising:
the device comprises a magnetic field generator, an alternating magneto-optical sensor, a power supply and a processor;
the magnetic field generator is connected with the power supply and is used for applying an external magnetic field with preset magnetic induction intensity to the weldment to be detected;
the alternating magneto-optical sensor is connected with the processor and used for acquiring magneto-optical images of the weldment to be detected and sending the magneto-optical images to the processor;
the processor is adapted to carry out the steps of the method of weld defect detection according to any one of claims 1 to 3 when executing a computer program stored in the memory.
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 CN108711151A (en) | 2018-10-26 |
CN108711151B true 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) |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6101817B2 (en) * | 2013-11-01 | 2017-03-22 | Eizo株式会社 | Image quality improvement device, image display device, image quality improvement method, and computer program |
CN104063873B (en) * | 2014-07-08 | 2017-11-14 | 华东交通大学 | A kind of Model For The Bush-axle Type Parts surface defect online test method based on compressed sensing |
CN105181122B (en) * | 2015-08-06 | 2018-06-15 | 兰州理工大学 | Mechanical oscillation signal data compression acquisition method |
CN105741336B (en) * | 2016-03-07 | 2019-04-30 | 中国科学院上海高等研究院 | A kind of splits' positions perception imaging method and system based on stochastical sampling |
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
Non-Patent Citations (1)
Title |
---|
Single-Pixel Imaging via Compressive Sampling;Marco F. Duarte等;《IEEE SIGNAL PROCESSING MAGAZINE》;20080331;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN108711151A (en) | 2018-10-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Borwankar et al. | An optical surface inspection and automatic classification technique using the rotated wavelet transform | |
Haueisen et al. | Evaluation of inverse algorithms in the analysis of magnetic flux leakage data | |
Starck et al. | Weak lensing mass reconstruction using wavelets | |
Faridani et al. | Introduction to local tomography | |
CN110231394A (en) | The irregular crackle imaging method of nonferromugnetic material based on ac magnetic field | |
Kathirmani et al. | Online data compression of MFL signals for pipeline inspection | |
Haddar et al. | Near-field linear sampling method for axisymmetric eddy current tomography | |
CN108711151B (en) | Welding defect detection method, device, equipment, storage medium and system | |
Zhang et al. | Extraction of pipeline defect feature based on variational mode and optimal singular value decomposition | |
Li et al. | A novel gas/liquid two-phase flow imaging method through electrical resistance tomography with DDELM-AE sparse dictionary | |
Soni et al. | Non-destructive fusion method for image enhancement of Eddy current sub-surface defect images | |
Huang et al. | A localization method for subsea pipeline based on active magnetization | |
Betta et al. | Fast 2D crack profile reconstruction by image processing for Eddy-Current Testing | |
Elshafiey et al. | Image fusion based enhancement of nondestructive evaluation systems | |
Daida et al. | Measuring topography of small-scale water surface waves | |
Zhang et al. | An extended-DORT method and its application in a cavity configuration | |
Huang et al. | Magnetic Flux Leakage: Theories and Imaging Technologies | |
Berkels et al. | Joint denoising and distortion correction of atomic scale scanning transmission electron microscopy images | |
Prachetaa et al. | Image processing for NDT images | |
Mukherjee et al. | Dynamic defect detection in fast, robust nde methods by transfer learning based optimally binned hypothesis tests | |
Safdarnejad et al. | Analysis of EC-GMR data for detection of cracks under fasteners (CUFs) | |
Watson | Non-destructive testing and evaluation applications of quantum well Hall effect sensors | |
Cai | Iterative blind deconvolution and its application in characterization of eddy current NDE signals | |
Nitzsche et al. | Compensating model imperfections during image reconstruction via Resesop | |
Cheng et al. | Phase Retrieval Algorithm Based on Transport of Intensity Equation under the Fusion of Regularization and Grating Modulation |
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 |