CN106204478A - The magneto optic images based on background noise feature space strengthens algorithm - Google Patents
The magneto optic images based on background noise feature space strengthens algorithm Download PDFInfo
- Publication number
- CN106204478A CN106204478A CN201610532328.7A CN201610532328A CN106204478A CN 106204478 A CN106204478 A CN 106204478A CN 201610532328 A CN201610532328 A CN 201610532328A CN 106204478 A CN106204478 A CN 106204478A
- Authority
- CN
- China
- Prior art keywords
- magneto
- background noise
- optic
- column vector
- level image
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 23
- 238000005086 pumping Methods 0.000 claims abstract description 49
- 239000011159 matrix material Substances 0.000 claims abstract description 30
- 230000009466 transformation Effects 0.000 claims abstract description 25
- 238000006243 chemical reaction Methods 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 19
- 230000009467 reduction Effects 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 47
- 238000001514 detection method Methods 0.000 abstract description 11
- 238000003384 imaging method Methods 0.000 abstract description 10
- 230000000694 effects Effects 0.000 abstract description 9
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 7
- 230000002708 enhancing effect Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 239000010409 thin film Substances 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 239000010410 layer Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000012880 independent component analysis Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 238000010147 laser engraving Methods 0.000 description 1
- 239000000696 magnetic material Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000000711 polarimetry Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20182—Noise reduction or smoothing in the temporal domain; Spatio-temporal filtering
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of the magneto optic images based on background noise feature space and strengthen algorithm, it is gray scale column vector without the magneto-optic gray level image under pumping signal according to row preferential conversion by part N width to be measured, merge and obtain background noise gray matrix, try to achieve the K L transformation matrix as background noise feature space matrix of background noise gray matrix, then it is gray scale column vector by the magneto-optic gray level image having under pumping signal according to row preferential conversion, use this gray scale column vector of K L transformation calculations at the projection vector of background noise feature space, K L inverse transformation is used to obtain column vector projection vector the most again, this column vector is reduced into image and obtains background noise gray level image, enhanced magneto-optic gray level image is obtained by there being the magneto-optic gray level image subtracting background noise gray level image under pumping signal.The present invention can effectively reject the interference of background noise, prominent defect information, improves the reinforced effects of the magneto optic images, is favorably improved the magneto-optic imaging system detection efficiency to defect.
Description
Technical field
The invention belongs to magneto-optic imaging detection technology field, more specifically, relate to a kind of based on background noise feature
The magneto optic images in space strengthens algorithm.
Background technology
In the defects detection of metallicl magnetic material, in conjunction with the magneto-optic image checking (Magneto-optic of Magnetic Flux Leakage Inspecting
Imaging, MOI) technology is rapidly developed because of its visualization, rapidity, accuracy, and this technology is imitated based on Faraday magneto-optical
Should, the stray field reflecting defect information is changed into image information, the structure that can be realized as material by analyzing image is good for
Health state-detection.
Magneto-optic imaging detection technology obtains defect information indeed through analyzing light intensity signal, and optical signal is easily
Being interfered, therefore magneto-optic imaging system should possess higher delicate nature.But, even if in actual detection it can be seen that
Being the precision instrument using higher standard, magneto-optic imaging system still can be affected by various interference sources.These interference sources or
It is owing to instrument and equipment degree of accuracy is inadequate, or due to the interference diffraction of light self, it is also possible to the behaviour that operator cause
Make error, all cannot avoid completely.Interference source directly will introduce the background noise of different characteristic in the magneto optic images, with the most scarce
Fall into information weave in, directly affects the detection of defect.The existence of interference significantly reduces defect recognition rate, the most then reduce
Defect information amount, heavy then cause wrong identification, thus become and limit the key point that magneto-optic imaging detection technology is pushed ahead.
In order to solve above-mentioned noise jamming problem, improving detection efficiency, related researcher proposes some image enhaucament
Algorithm.Calendar year 2001, the Udpa of the U.S. utilizes wavelet transformation and morphology technology to realize the process to single frames the magneto optic images, prominent
Defect information [A algorithm].This team has been then made as improving, and proposes Dynamic Filtering, first passes through magneto-optic thin film and examination
The relative movement of part obtains continuous print image sequence, is then based on frame difference method and realizes defect information (dynamic part) and interference information
The separation [B, C algorithm] of (static part).2009, Matteo used singular value decomposition method identification based on Karhunen-Loeve transformation principle
The integrity [D-algorithm] of airplane riveting.2010, Matteo realized the denoising of the magneto optic images based on independent component analysis, and with
The denoising performance of self adaptation dynamic filter and dynamic filter is analyzed [E algorithm].In recent years, Patterson propose based on
Many image processing methods of polarimetry, change the analyzer angle in magneto-optic path and obtain multiple the magneto optic images, and according to
Dependency between image eliminates impact [the F calculation of the factors such as total light intensity change, camera exposure time, light path be non-linear
Method].In above method, the Dynamic Filtering used in B, C algorithm achieves good effect, but both algorithms are
Sampling based on many test points, real-time is the highest.Method in A, D, E, F algorithm, based on single test point sampling, solves real-time
Problem, but when suppressing relatively very noisy to disturb, effect is unsatisfactory.
Summary of the invention
It is an object of the invention to overcome the deficiencies in the prior art, it is provided that a kind of magneto-optic based on background noise feature space
Algorithm for image enhancement, uses Karhunen-Loeve transformation to obtain background noise feature space according to without the magneto optic images under pumping signal, it is achieved to have
Defect and the separation of background noise in the magneto optic images under pumping signal, improve the reinforced effects of the magneto optic images.
For achieving the above object, present invention the magneto optic images based on background noise feature space strengthen algorithm include with
Lower step:
S1: part to be measured is obtained N width the magneto-optic figure under pumping signal without the magneto optic images under pumping signal and 1 width
Picture, wherein N >=1, be converted to gray level image by this N+1 width the magneto optic images；
S2: be gray scale column vector without the magneto-optic gray level image under pumping signal according to row preferential conversion by every width, by N number of row
Vector merging obtains background noise gray matrix W, tries to achieve the Karhunen-Loeve transformation matrix A of background noise gray matrix W, by this conversion square
Battle array as background noise feature space matrix；
S3: the magneto-optic gray level image I under pumping signal will be had to be gray scale column vector B according to row preferential conversion, uses K-L to become
Change the projection vector Y=A calculating gray scale column vector B at background noise feature spaceT(B-mX), wherein mXFor background noise gray scale
The meansigma methods column vector of matrix W；Then projection vector Y is carried out K-L inverse transformation and obtain column vector γ=A*Y+mX, by column vector
γ obtains gray level image I according to the preferential reduction of row1；
S4: use image to make difference method and calculate enhanced magneto-optic gray level image I2=I-I1。
Present invention the magneto optic images based on background noise feature space strengthens algorithm, by part N width to be measured without under pumping signal
Magneto-optic gray level image be gray scale column vector according to row preferential conversion, merge obtain background noise gray matrix, try to achieve background and make an uproar
The Karhunen-Loeve transformation matrix as background noise feature space matrix of sound gray matrix, then will have the magneto-optic gray scale under pumping signal
Image is gray scale column vector according to row preferential conversion, uses Karhunen-Loeve transformation to calculate this gray scale column vector at background noise feature space
Projection vector, the most again to projection vector use K-L inverse transformation obtain column vector, this column vector is reduced into image and is carried on the back
Scape noise gray level image, obtains enhanced magnetic by there being the magneto-optic gray level image subtracting background noise gray level image under pumping signal
Light gray level image.Through it is demonstrated experimentally that the present invention can effectively reject the interference of background noise, highlight defect information, improve magnetic
The reinforced effects of light image, is favorably improved the magneto-optic imaging system detection efficiency to defect.
Accompanying drawing explanation
Fig. 1 is the detailed description of the invention flow process that present invention the magneto optic images based on background noise feature space strengthens algorithm
Figure；
Fig. 2 is part photo to be measured in the present embodiment；
Fig. 3 is the interference Examples of information figure of the magneto optic images in the present embodiment；
Fig. 4 is without under pumping signal and have the Z-shaped defect the magneto optic images exemplary plot under pumping signal；
Fig. 5 is without the Z-shaped defect the magneto optic images collection under pumping signal in the present embodiment；
Fig. 6 is to have the Z-shaped defect the magneto optic images under pumping signal in the present embodiment；
Fig. 7 is the gray level image of background noise component corresponding to Z-shaped defect the magneto optic images shown in Fig. 6；
Fig. 8 is Z-shaped defect the magneto optic images enhanced magneto-optic gray level image shown in Fig. 6；
Fig. 9 is to have under pumping signal and directly make the image that difference obtains without the magneto optic images under pumping signal.
Detailed description of the invention
Below in conjunction with the accompanying drawings the detailed description of the invention of the present invention is described, in order to those skilled in the art is preferably
Understand the present invention.Requiring particular attention is that, in the following description, when known function and design detailed description perhaps
When can desalinate the main contents of the present invention, these are described in and will be left in the basket here.
Embodiment
Fig. 1 is the detailed description of the invention flow process that present invention the magneto optic images based on background noise feature space strengthens algorithm
Figure.Comprise the following steps as it is shown in figure 1, present invention the magneto optic images based on background noise feature space strengthens algorithm:
S101: acquisition the magneto optic images:
Part to be measured is obtained N width and has the magneto optic images under pumping signal without the magneto optic images under pumping signal and 1 width, its
Middle N >=1.The magneto optic images of the present invention strengthens the enhancing that algorithm is all based in Pixel-level, needs to use each pixel
Half-tone information, the most first is required for first changing into magneto-optic gray level image by this N+1 width the magneto optic images.
In order to be able to characterize more abundant noise information, obtain N width without pumping signal under the magneto optic images time, every width figure
The overall light intensity of picture is otherwise varied, i.e. the overall light intensity of each image is different.In general can be become by fine setting magneto-optic
As in system, analyzer angle realizes.
S102: tectonic setting noise characteristic space:
In the present invention, background noise feature space is to be constructed by the magneto optic images collection without pumping signal, can be often
Width regards a sample of background noise as without the magneto optic images under pumping signal.By every width without the magneto-optic gray scale under pumping signal
Image is gray scale column vector according to row preferential conversion, N number of column vector is merged and obtains background noise gray matrix W, it is clear that background
The size of noise gray matrix W is the pixel quantity that Q × N, Q represent every width magneto-optic gray level image.Try to achieve background noise gray scale
The Karhunen-Loeve transformation matrix A of matrix W.Karhunen-Loeve transformation matrix A is exactly background noise feature space matrix, and each of which column vector represents the back of the body
One coordinate axes in scape noise characteristic space.Karhunen-Loeve transformation is a kind of common technology means of area of pattern recognition, and it specifically calculates
Process does not repeats them here.
S103: extraction background noise component:
The magneto-optic gray level image I under pumping signal will be had to be gray scale column vector B according to row preferential conversion, use Karhunen-Loeve transformation
Calculate the gray scale column vector B projection vector Y=A at background noise feature spaceT(B-mX), wherein mXFor background noise Gray Moment
The meansigma methods column vector of battle array W, the most N number of without the magneto-optic gray level image corresponding grey scale column vector under pumping signal averagely arrange to
Amount.Y is i.e. the coordinate system of the point having the magneto-optic gray level image under pumping signal to project in background noise feature space.The most right
Projection vector Y carries out K-L inverse transformation and obtains column vector γ=A*Y+mX, column vector γ is obtained gray-scale map according to the preferential reduction of row
As I1, image I1It it is exactly the background noise component in image I.
The poor enhancing the magneto optic images that obtains of S104: work:
The present embodiment use work difference method come the defect information in the magneto optic images under splitting driving signal and noise information,
Directly with image I subtracted image I1Just obtain enhanced magneto-optic gray level image I2, i.e. I2=I-I1.Because image I2Without the back of the body
Scape noise component(s), contains only defect component, it is achieved thereby that the enhancing of the magneto optic images.
In order to the technique effect of the present invention is described, use instantiation that the present invention is carried out experimental verification.Fig. 2 is this enforcement
Part photo to be measured in example.As in figure 2 it is shown, have selected two kinds of parts to be measured in the present embodiment, one is a size of 199 × 100 × 6
(mm) industrial steel plate, another kind is the stalloy of a size of 160 × 30 × 0.5 (mm).Then use laser engraving machine at this
Steel plate machined a Z-shaped defect and " three " font defect, stalloy machined a straight defect.
Fig. 3 is the interference Examples of information figure of the magneto optic images in the present embodiment.Fig. 3 has intercepted Z-shaped defect in part one to be measured
Near the magneto optic images and part to be measured two in the magneto optic images near straight defect.As it is shown on figure 3, background noise substantially can be divided into
Two classes, a class is plaque-like interference, and its distribution is irregular, and another kind of is striated interference, and its distribution is the most uniform.It is true that
If the most only disconnecting pumping signal, when the most not producing defect and magnetic leakage field signal, the magneto optic images of acquisition is still wrapped
Contain this type of interference, and be distributed constant.
If only observing from image, can only know and image introduces background noise and some of two dimension spy really
Levy, but be not aware that noise source is in which link and its mechanism of production.By reviewing layer by layer, find that plaque-like interference is main next
Come from the LASER Light Source in magneto-optic imaging system, because the laser that LASER Light Source produces is not absolute uniform, the most irregular
Part produces the change of light and shade in the picture, forms the background noise of not excited target effect of signals.Secondly beam expanding lens, the polarizer and
Impurity on analyzer and the dust in air also can produce a small amount of plaque-like interference.Striated interference is due to interference of light
Cause, result from the interlayer between magneto-optic thin film and rear reflector.Due to the existence in this space, just through the laser of magneto-optic thin film
Meet with the laser being reflected, interfere phenomenon, form light and dark striped, and be finally reflected in the picture.
Fig. 4 is without under pumping signal and have the Z-shaped defect the magneto optic images exemplary plot under pumping signal.As shown in Figure 4, when only
When observing image when having pumping signal, feature is more mixed and disorderly, it is impossible to distinguish defect information and interference information.But when knowing without swashing
When encouraging the magneto optic images of signal, it is believed that the most all of feature is all background noise, this is because theoretically, wherein
Should not comprise any defect characteristic.And the most accurately extract without the background noise information in the magneto optic images under pumping signal,
Thus to there being the magneto optic images under pumping signal to strengthen, prominent defect information, is key issue.Therefore the present invention proposes
Use the method that background noise feature space is extracted in Karhunen-Loeve transformation.
Fig. 5 is without the Z-shaped defect the magneto optic images collection under pumping signal in the present embodiment.As it is shown in figure 5, the present embodiment obtains
12 width are taken without the Z-shaped defect the magneto optic images under pumping signal, the overall light intensity difference of each image.Each image size is
320 × 240, after being converted into gray level image, it is into the column vector of 76800 × 1 dimensions by row preferential conversion, is merged into 76800 × 12
The background noise gray matrix W of dimension, is then calculated Karhunen-Loeve transformation matrix A, and this Karhunen-Loeve transformation matrix A is exactly background noise feature
Space matrix.
Fig. 6 is to have the Z-shaped defect the magneto optic images under pumping signal in the present embodiment.By the Z-shaped defect magneto-optic figure shown in Fig. 6
After changing into column vector, Karhunen-Loeve transformation and K-L inverse transformation is used to obtain background noise and divide according to background noise feature space matrix
The gray level image I of amount1, then subtract each other and obtain enhanced magneto-optic gray level image I2.Fig. 7 is Z-shaped defect the magneto optic images shown in Fig. 6
The gray level image of corresponding background noise component.Fig. 8 is Z-shaped defect the magneto optic images enhanced magneto-optic gray level image shown in Fig. 6.
Although comparison diagram 6 and Fig. 8 it is found that in enhanced magneto-optic gray level image the intensity of defect be weakened a part, but greatly
Amount background noise is successfully suppressed, and highlights defect information, such that it is able to make the identification of defect be more prone to.
Theoretically, under same experimental conditions without pumping signal with have the magneto optic images of pumping signal only on defect characteristic
There is difference, it is thus possible to directly it is made difference and realize defect enhancing.Select the Z-shaped defect magnetic under the be shown with pumping signal of Fig. 6
The gray level image of light image is the poorest without the 6th image of the Z-shaped defect the magneto optic images concentration under pumping signal with shown in Fig. 5
Obtain gray level image.Fig. 9 is to have under pumping signal and directly make the image that difference obtains without the magneto optic images under pumping signal.Comparison diagram
8 and Fig. 9 can not well suppress noise it is found that make difference method, this is because after applying pumping signal, not only change
The gray value of defect area in image, the gray value in non-defective region is also affected and uneven.And the present invention based on the back of the body
The magneto optic images in scape noise characteristic space strengthens algorithm and avoids this impact, uses without the background noise image under pumping signal
Construct feature space, thus from there being extraction noise component(s) pumping signal and the magneto optic images, remain the inhomogeneities of noise,
Improve the reinforced effects to the magneto optic images.
In order to preferably compare the reinforced effects of this algorithm, the concept introducing picture contrast characterizes dividing of the magneto optic images
Resolution, when contrast is the highest, the defect characteristic in explanatory diagram picture is the most clear.Being computed, the contrast of image shown in Fig. 8 is
The contrast of image shown in 14.6551, Fig. 9 is 10.9814.Visible, the contrast of the magneto optic images after the enhancing that the present invention obtains
It is much better than the contrast of image obtained by directly work differs from method.Understand through lot of experimental data contrast, use the increasing that the present invention obtains
After strong the contrast of the magneto optic images with do not strengthen before original the magneto optic images compared with, its contrast reaches the two of original the magneto optic images
Times.Visible, the magneto optic images can effectively be realized strengthening by the present invention, prominent defect information, thus improves magneto-optic imaging system pair
The detection efficiency of defect.
Although detailed description of the invention illustrative to the present invention is described above, in order to the technology of the art
Personnel understand the present invention, the common skill it should be apparent that the invention is not restricted to the scope of detailed description of the invention, to the art
From the point of view of art personnel, as long as various change limits and in the spirit and scope of the present invention that determine in appended claim, these
Change is apparent from, and all utilize the innovation and creation of present inventive concept all at the row of protection.
Claims (2)
1. a magneto optic images based on background noise feature space strengthens algorithm, it is characterised in that comprise the following steps:
S1: part to be measured is obtained N width has the magneto optic images under pumping signal without the magneto optic images under pumping signal and 1 width, its
Middle N >=1, is converted to gray level image by this N+1 width the magneto optic images；
S2: be gray scale column vector without the magneto-optic gray level image under pumping signal according to row preferential conversion by every width, by N number of column vector
Merge and obtain background noise gray matrix W, try to achieve the Karhunen-Loeve transformation matrix A of background noise gray matrix W, this transformation matrix is made
For background noise feature space matrix；
S3: the magneto-optic gray level image I under pumping signal will be had to be gray scale column vector B according to row preferential conversion, uses Karhunen-Loeve transformation meter
Calculate the gray scale column vector B projection vector Y=A in background noise background characteristics spaceT(B-mX), wherein mXFor background noise gray scale
The meansigma methods column vector of matrix W；Then projection vector Y is carried out K-L inverse transformation and obtain column vector γ=A*Y+mX, by column vector
γ obtains gray level image I according to the preferential reduction of row1；
S4: use image to make difference method and calculate enhanced magneto-optic gray level image I2=I-I1。
The magneto optic images the most according to claim 1 strengthens algorithm, it is characterised in that in described step S1, and N width is without excitation letter
In the magneto optic images under number, the overall light intensity of each image is different.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610532328.7A CN106204478B (en) | 2016-07-06 | 2016-07-06 | The magneto optic images Enhancement Method based on ambient noise feature space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610532328.7A CN106204478B (en) | 2016-07-06 | 2016-07-06 | The magneto optic images Enhancement Method based on ambient noise feature space |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106204478A true CN106204478A (en) | 2016-12-07 |
| CN106204478B CN106204478B (en) | 2018-09-07 |
Family
ID=57473735
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610532328.7A Expired - Fee Related CN106204478B (en) | 2016-07-06 | 2016-07-06 | The magneto optic images Enhancement Method based on ambient noise feature space |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106204478B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180428A (en) * | 2017-06-09 | 2017-09-19 | 电子科技大学 | A kind of the magneto optic images Enhancement Method based on optical flow algorithm |
| CN107993225A (en) * | 2017-11-28 | 2018-05-04 | 电子科技大学 | A kind of recognition methods of the defects of magneto-optic vortex imaging detection |
| CN112183325A (en) * | 2020-09-27 | 2021-01-05 | 哈尔滨市科佳通用机电股份有限公司 | Road vehicle detection method based on image comparison |
| CN113066072A (en) * | 2021-04-08 | 2021-07-02 | 南昌航空大学 | Method and system for detecting microcrack defects of guide blades of aero-engine |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196983A (en) * | 2006-12-08 | 2008-06-11 | 北京皎佼科技有限公司 | Image recognition method |
| CN101339767A (en) * | 2008-03-21 | 2009-01-07 | 华为技术有限公司 | Background noise excitation signal generating method and apparatus |
| US20140212015A1 (en) * | 2013-01-31 | 2014-07-31 | The Ohio State University | De-noising of Real-time Dynamic Magnetic Resonance Images by the Combined Application of Karhunen-Loeve Transform (KLT) and Wavelet Filtering |
-
2016
- 2016-07-06 CN CN201610532328.7A patent/CN106204478B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196983A (en) * | 2006-12-08 | 2008-06-11 | 北京皎佼科技有限公司 | Image recognition method |
| CN101339767A (en) * | 2008-03-21 | 2009-01-07 | 华为技术有限公司 | Background noise excitation signal generating method and apparatus |
| US20140212015A1 (en) * | 2013-01-31 | 2014-07-31 | The Ohio State University | De-noising of Real-time Dynamic Magnetic Resonance Images by the Combined Application of Karhunen-Loeve Transform (KLT) and Wavelet Filtering |
Non-Patent Citations (1)
| Title |
|---|
| 莫玲 等: "微间隙焊缝磁光图像增强方法", 《焊接技术》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180428A (en) * | 2017-06-09 | 2017-09-19 | 电子科技大学 | A kind of the magneto optic images Enhancement Method based on optical flow algorithm |
| CN107180428B (en) * | 2017-06-09 | 2019-08-20 | 电子科技大学 | A kind of the magneto optic images Enhancement Method based on optical flow algorithm |
| CN107993225A (en) * | 2017-11-28 | 2018-05-04 | 电子科技大学 | A kind of recognition methods of the defects of magneto-optic vortex imaging detection |
| CN107993225B (en) * | 2017-11-28 | 2020-06-23 | 电子科技大学 | Defect identification method for magneto-optical eddy current imaging detection |
| CN112183325A (en) * | 2020-09-27 | 2021-01-05 | 哈尔滨市科佳通用机电股份有限公司 | Road vehicle detection method based on image comparison |
| CN112183325B (en) * | 2020-09-27 | 2021-04-06 | 哈尔滨市科佳通用机电股份有限公司 | Road vehicle detection method based on image comparison |
| CN113066072A (en) * | 2021-04-08 | 2021-07-02 | 南昌航空大学 | Method and system for detecting microcrack defects of guide blades of aero-engine |
| CN113066072B (en) * | 2021-04-08 | 2023-06-06 | 南昌航空大学 | Method and system for detecting microcrack defects of guide blade of aero-engine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106204478B (en) | 2018-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106204478A (en) | The magneto optic images based on background noise feature space strengthens algorithm | |
| TW581865B (en) | Adaptive mask technique for defect inspection | |
| KR100954703B1 (en) | Method and system for detecting defects | |
| Pini et al. | Moving across scales: a quantitative assessment of X-ray CT to measure the porosity of rocks | |
| JP2005315596A (en) | Image inspection method and device | |
| Georgiou et al. | GAMA+ KiDS: Alignment of galaxies in galaxy groups and its dependence on galaxy scale | |
| CN105913401B (en) | Industrial camera photography measurement image luminance compensation method | |
| Li et al. | Divide and conquer: real-time maximum likelihood fitting of multiple emitters for super-resolution localization microscopy | |
| CN113330481B (en) | System and method for using tensor decomposition and singular value decomposition verification | |
| Heintzmann et al. | Calibrating photon counts from a single image | |
| CN105510364A (en) | Nondestructive testing system for industrial part flaws based on X rays and detection method thereof | |
| CN116802688A (en) | Apparatus and method for correspondence analysis within an image | |
| CN111602177A (en) | Method and apparatus for generating a 3D reconstruction of an object | |
| CN114764189A (en) | Microscope system and method for evaluating image processing results | |
| Zhao et al. | Quantitatively mapping local quality of super-resolution microscopy by rolling Fourier ring correlation | |
| EP1071042A2 (en) | Method for determining the components of image noise patterns of an imaging device and use of this method in an imaging device | |
| Chen et al. | Measure and model a 3-D space-variant PSF for fluorescence microscopy image deblurring | |
| Arenhart et al. | Experimental investigation of the influencing factors on the structural resolution for dimensional measurements with CT systems | |
| KR102538268B1 (en) | Detection of defects in the logic area on the wafer | |
| Kölln et al. | Label2label: training a neural network to selectively restore cellular structures in fluorescence microscopy | |
| CN1766607A (en) | Thin layer chromatography quantitative analysis method based on image processing technology | |
| Paul et al. | Automatic noise quantification for confocal fluorescence microscopy images | |
| CN115516297B (en) | Image alignment for noisy images | |
| Sahasrabudhe et al. | Structured spatial domain image and data comparison metrics | |
| Zhao et al. | Neutron image denoising and deblurring based on generative adversarial networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Cheng Yuhua Inventor after: Xia Yongzhao Inventor after: Bai Libing Inventor after: Huang Yiyun Inventor after: Yin Chun Inventor after: Tian Lulu Inventor after: Wang Chao Inventor before: Cheng Yuhua Inventor before: Xia Yongzhao Inventor before: Bai Libing Inventor before: Huang Yiyun Inventor before: Yin Chun Inventor before: Tian Lulu |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180907 Termination date: 20210706 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |