CN110084768B - Defect detection method of LCD light guide plate based on background filtering - Google Patents
Defect detection method of LCD light guide plate based on background filtering Download PDFInfo
- Publication number
- CN110084768B CN110084768B CN201910389155.1A CN201910389155A CN110084768B CN 110084768 B CN110084768 B CN 110084768B CN 201910389155 A CN201910389155 A CN 201910389155A CN 110084768 B CN110084768 B CN 110084768B
- Authority
- CN
- China
- Prior art keywords
- image
- filtering
- light guide
- guide plate
- lcd light
- 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
- 238000001914 filtration Methods 0.000 title claims abstract description 77
- 238000001514 detection method Methods 0.000 title claims abstract description 36
- 230000007547 defect Effects 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 22
- 230000000694 effects Effects 0.000 claims abstract description 9
- 238000012216 screening Methods 0.000 claims description 9
- 238000001228 spectrum Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 238000012804 iterative process Methods 0.000 claims description 2
- 238000012417 linear regression Methods 0.000 claims description 2
- 230000002265 prevention Effects 0.000 claims description 2
- 238000012805 post-processing Methods 0.000 claims 1
- 230000002950 deficient Effects 0.000 abstract description 7
- 239000004973 liquid crystal related substance Substances 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/149—Segmentation; Edge detection involving deformable models, e.g. active contour models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/168—Segmentation; Edge detection involving transform domain methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
Abstract
The invention provides a defect detection method of an LCD light guide plate based on background filtering, which comprises the following steps: s110, collecting an image of an LCD light guide plate; s120, performing iterative guided filtering on the LCD light guide plate image; s130, performing frequency domain filtering on the image subjected to the guide filtering of the LCD light guide plate by adopting a fast Fourier transform method; and S140, dividing the defective area by using a maximum inter-class variance method on the image after the frequency domain filtering of the LCD light guide plate to obtain a detection result. The defect detection method of the LCD light guide plate based on background filtering has good background texture filtering effect and high defect detection rate.
Description
Technical Field
The invention relates to the technical field of detection algorithms, in particular to a defect detection method of an LCD light guide plate based on background filtering.
Background
The liquid crystal display-light guide plate (Liquid Crystal Display-Light Guide Plate LCD-LGP) is an important component in the production of backlight modules of liquid crystal displays, and if a circular lattice structure on the surface of the light guide plate is defective, the brightness and uniformity of the display of the liquid crystal display will be affected, so detecting the surface defect of the light guide plate is a necessary link in the production of LCDs. In recent years, the rapid development of the display manufacturing industry and the trend of high automation and intellectualization of the production line in China promote the development of the automatic optical detection (Auto Optical Inspection AOI) technology of the surface defects based on machine vision, so that the yield and the yield of the LCD are improved, and the cost brought by the manual vision detection is reduced.
However, the existing algorithm cannot well filter complex background textures, and the light guide particles on the surface of the LCD light guide plate have the characteristics of large distribution density, lack of periodic rules of a distribution structure, dispersion attenuation of a spectrogram and the like, so that certain difficulties are brought to a detection method based on frequency domain analysis, and therefore, the existing detection method based on the frequency domain analysis needs to be improved.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a defect detection method of an LCD light guide plate based on background filtering, which has higher detection rate compared with the traditional detection method based on frequency domain analysis. The technical scheme adopted by the invention is as follows:
a surface defect detection method of LCD light guide plate includes:
s110, collecting an image of an LCD light guide plate;
s120, performing iterative guided filtering on the LCD light guide plate image;
s130, performing frequency domain filtering on the image subjected to the guide filtering of the LCD light guide plate by adopting a fast Fourier transform method;
and S140, dividing the defective area of the image after the frequency domain filtering of the LCD light guide plate by using a dynamic threshold method to obtain a detection result.
The invention has the advantages that: the invention builds a guide image, adopts an iterative guide filtering processing mode, introduces an iterative stop criterion according to the structural similarity, carries out preliminary filtering on the image of the LCD light guide plate, and maintains the outline of a defect area to a certain extent while filtering the background information of the image of the LCD light guide plate; the filtered image is subjected to two-dimensional discrete Fourier transform, so that high-frequency noise in the guided filtered image is further eliminated, and finally the target defect is segmented through a dynamic threshold value. Experiments prove that the method has a better filtering effect on the background information of the complicated texture of the LCD light guide plate image and no periodicity rule, and has higher detection rate compared with the traditional detection method based on frequency domain analysis.
Drawings
Fig. 1 is a flowchart of a defect detection method of an LCD light guide plate according to the present invention.
Fig. 2 is an effect image of the LCD light guide plate image after iterative guided filtering in step S120 of the present invention.
Fig. 3 is an effect image of the LCD light guide plate image after frequency domain filtering in step S130 of the present invention.
Fig. 4 is a reconstructed image obtained by performing two-dimensional inverse discrete fourier transform in step S130 according to the present invention, from which the background is removed.
Fig. 5 is a view showing a defective area image obtained by dividing the defective area of the frequency domain filtered image of the LCD light guide plate by using a dynamic thresholding method in step S140 of the present invention.
Detailed Description
The following describes specific embodiments of the present invention in detail with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the invention, are not intended to limit the invention.
Note that, the LCD-LGP: liquid Crystal Display-Light Guide Plate, a liquid crystal screen light guide plate; SSIM: structural similarity index measure, structural similarity index.
As one aspect of the present invention, there is provided a surface defect detection method of an LCD light guide plate, wherein the surface defect detection method of the LCD light guide plate, as shown in fig. 1, includes:
s110, collecting an image of an LCD light guide plate;
s120, performing iterative guided filtering on the LCD light guide plate image;
s130, performing frequency domain filtering on the image subjected to the guide filtering of the LCD light guide plate by adopting a fast Fourier transform method;
and S140, dividing the defective area of the image after the frequency domain filtering of the LCD light guide plate by using a dynamic threshold method to obtain a detection result.
The defect detection algorithm of the LCD light guide plate based on background filtering provided by the invention is characterized in that an iterative guide filtering processing mode is adopted by constructing a guide image, an iterative stop criterion is introduced according to structural similarity, the image of the LCD light guide plate is subjected to preliminary filtering, and the outline of a defect area is maintained to a certain extent while the background information of the image of the LCD light guide plate is filtered; the filtered image is subjected to two-dimensional discrete Fourier transform, so that high-frequency noise in the guided filtered image is further eliminated, and finally the target defect is segmented through a dynamic threshold value. Experiments prove that the method has a good filtering effect on the background information of the LCD light guide plate, which is complex in image and has no periodicity rule, and has a higher detection rate compared with the traditional detection method based on frequency domain analysis.
Specifically, S120, performing iterative guided filtering on the LCD light guide plate image, including:
the collected LCD light guide plate image is subjected to block processing; for example into hundreds of sub-images;
constructing a corresponding initial guide image according to the length and the width of each sub-image;
setting the gray initial value of each pixel point in all initial guide images to 0;
wherein I is a guide image, I i Is a guideTo the pixel value of the image, i and k each represent the position of the pixel point, ω k Represents a two-dimensional window of radius r centered on pixel k, p represents the input image, q represents the output image, E is the value q that causes the output image to be at pixel i i And the value p of the input image at pixel point i i Cost function with minimum gap, a k And b k The coefficients are obtained by linear regression, at a k In the expression of (2), ω is a two-dimensional window ω k The total number of pixel points in (a),is p in a two-dimensional window omega k Average value of μ k Is the guiding image I in a two-dimensional window omega k Average value of>Is the guiding image I in a two-dimensional window omega k In (a) is the variance of epsilon is the prevention of a k Oversized regularization parameters due to the two-dimensional window ω k The pixel position i is covered by the pixel points with the number of |omega| so that a is utilized k And b k Mean value of>And->The effect of removing noise is achieved;
taking the image after the first guide filtering as the guide image of the next guide filtering, namely continuously updating the guide image in the iterative process, wherein the iterative guide filtering formula is as follows:
in the formula, G represents guide filtering,expressed in q in the t-th iteration t-1 The guiding filtering process is that the guiding image and p is the input image, and the output image is q t R represents the radius of the two-dimensional window, s is a smoothing factor, and the inverse of the regularization parameter is taken and rounded down;
the structural similarity index SSIM (structural similarity index measure) is used for constructing an iteration stopping criterion of the guided filtering, and the structural similarity and the iteration stopping criterion are expressed as follows:
|SSIM (n+1) -SSIM (n) |<ζ,n=1,2,3,...,m
wherein mu is t Sum mu t+1 Respectively representing the output image q after guiding and filtering at the time t and the time t+1 t And q t+1 Mean, sigma of t Sum sigma t+1 Respectively representing the output image q after guiding and filtering at the time t and the time t+1 t And q t+1 Standard deviation of sigma of t(t+1) Representing the output image q after the guide filtering at the time t and the time t+1 t And q t+1 Covariance of C 1 And C 2 For a constant, let ζ be the iterative guided filter stop factor, ζ be an adjustable positive number much smaller than 1, SSIM (n) Structural similarity value representing nth order guided filtered output image and n-1 order guided filtered output image, SSIM (n+1) Structural similarity value representing the n+1th order guided filtered output image and the n order guided filtered output image, when SSIM (n+1) With SSIM (n) When the absolute value of the difference is much smaller than ζ, it can be considered that the guided filtering effect after the iteration hardly changes.
Specifically, S130, performing frequency domain filtering on the image after guide filtering of the LCD light guide plate by using a fast fourier transform method, including:
the image after the guide filtering is marked as F (x, y), the image is subjected to two-dimensional discrete Fourier transform to obtain a corresponding frequency domain image F (u, v), and the F (u, v) is subjected to modulo calculation, and the formula is as follows:
wherein M and N represent the size of the image, the frequency domain of the image is represented by F (u, v), u and v are frequency variables, u=0, 1,2,..m-1; v=0, 1, 2..n-1, x and y are space variables, R (u, v) and I (u, v) represent the real and imaginary parts of F (u, v), respectively;
calculating a power spectrum image, and carrying out Gaussian high-pass filtering on the obtained power spectrum image, wherein the formula is as follows:
P(u,v)=|F(u,v)| 2
P G (u,v)=G(u,v)P(u,v)
wherein P (u, v) is a power spectrum image, P G (u, v) is a power spectrum image after Gaussian high pass, and G (u, v) is a Gaussian filter;
using global threshold pairs P G The highlight region in (u, v) is segmented as follows:
T MinGrayValue <g<T MaxGrayValue
wherein T is MinGrayValue Representing a pixel gray value lower threshold value, T, adopted for segmentation MaxGrayValue Representing an upper limit threshold value of gray values of pixel points adopted by segmentation; g is the gray value of the pixel point meeting the condition;
screening the region subjected to global threshold segmentation according to the area size, wherein the area value occupied by each pixel point is recorded as 1, and the condition of region screening is as follows:
T MinAera <r<T MaxAera
wherein T is MinAera Represents the area lower threshold value, T, adopted by screening MaxAera Representing an upper threshold of area employed for screening;
the respective centroid is calculated for all the sub-regions screened out, and the formula is as follows:
wherein u is m And v m Is the coordinate of the centroid, N r The number of pixels in the connected domain (i.e. each sub-region is screened out);
taking the mass center of each sub-region as a circle center, taking R as a radius as a circle, and setting the gray value of the pixel point in the circular region to be zero;
the frequency domain image from which the high-frequency information is removed is subjected to two-dimensional inverse discrete Fourier transform to obtain a reconstructed image from which the background is removed; the formula is as follows:
specifically, S140, dividing the defective area of the image after the frequency domain filtering of the LCD light guide plate by using a dynamic threshold method, to obtain a detection result, including:
let the reconstructed image after the frequency domain filtering be g (s, t), carry out the mean filtering to the reconstructed image g (s, t), the formula is as follows:
wherein C is xy Indicating that the center point of the mean filter is at (x, y), and the window size of the mean filter is m and n;
let beta be the offset term, filter the mean value to the image f 2 (x, y) pixel-by-pixel comparison with the reconstructed image after frequency domain filtering, and fluctuation vision of gray values of pixel points within the range of the offset term betaAnd taking the fluctuation outside the range of the offset term beta as the gray value of the pixel point of the target area as the acceptable gray value, namely finishing the detection of the target defect.
The following describes in detail the implementation procedure of the surface defect detection method for the LCD light guide plate provided by the present invention with reference to fig. 2 to 5;
1. acquiring an image of a product to be detected by adopting a line scanning camera, as shown in fig. 2;
2. the image is segmented, in this embodiment: the tile image size is 512pixels x 512pixels;
3. constructing a guide image with the size 512pixels being 512pixels, and enabling the pixel value of each pixel point to be 0;
4. performing iterative guided filtering on the sub-images of the blocks; the formula is as follows:
5. according to the iteration stop criterion, the iteration times of the iterative guided filtering process are adaptively selected, and the formula is as follows:
|SSIM (n+1) -SSIM (n) |<ζ,n=1,2,3,...,m
6. the image after the iterative guided filtering is subjected to two-dimensional discrete Fourier transform, the formula is as follows, and the result is shown in fig. 3:
7. the power spectrum image is calculated as follows:
P(u,v)=|F(u,v)| 2
8. the power spectral image is gaussian filtered as follows:
P G (u,v)=G(u,v)P(u,v)
9. using global threshold pairs P G The highlight region in (u, v) is segmented as follows:
T MinGrayValue <g<T MaxGrayValue
10. the regions after global threshold segmentation are screened according to the area size, and the conditions of region screening are as follows:
T MinAera <r<T MaxAera
11. the respective centroid is calculated for all the sub-regions screened out, and the formula is as follows:
12. taking the mass center of each sub-region as a circle center, taking R as a radius as a circle, and setting the gray value of the pixel point in the circular region to be zero;
13. the frequency domain image from which the high frequency information is removed is subjected to inverse fourier transform to obtain a reconstructed image from which the background is removed, and the formula is as follows, and the result is shown in fig. 4:
14. let the reconstructed image after the frequency domain filtering be g (s, t), and perform the mean filtering on the image g (s, t), the formula is as follows:
15. let beta be the offset term, filter the mean value to the image f 2 (x, y) and the reconstructed image after the frequency domain filtering are compared pixel by pixel, the fluctuation of the gray value of the pixel point in the range of the offset term beta is regarded as an acceptable gray value, the fluctuation outside the range of the offset term beta is regarded as the gray value of the pixel point of the target area, namely the detection of the target defect is completed, the offset term can be set as an externally adjustable parameter, and the detection result is shown in figure 5.
Finally, it should be noted that the above-mentioned embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same, and although the present invention has been described in detail with reference to examples, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention, and all such modifications and equivalents are intended to be encompassed in the scope of the claims of the present invention.
Claims (3)
1. The defect detection method of the LCD light guide plate based on background filtering is characterized by comprising the following steps of:
s110, collecting an image of an LCD light guide plate;
s120, performing iterative guided filtering on the LCD light guide plate image;
s130, performing frequency domain filtering on the image subjected to the guide filtering of the LCD light guide plate by adopting a fast Fourier transform method;
s140, dividing a defect area of the image after the frequency domain filtering of the LCD light guide plate by using a dynamic threshold method to obtain a detection result;
s120, performing iterative guided filtering on the LCD light guide plate image, specifically including:
the collected LCD light guide plate image is subjected to block processing;
constructing a corresponding initial guide image according to the length and the width of each sub-image;
setting the gray initial value of each pixel point in all initial guide images to 0;
wherein I is a guide image, I i For guiding the pixel value of the image, i and k each represent the position of the pixel point, ω k Represents a two-dimensional window of radius r centered on pixel k, p represents the input image, q represents the output image, E is the value q that causes the output image to be at pixel i i And the value p of the input image at pixel point i i Cost function with minimum gap, a k And b k The coefficients are obtained by linear regression, at a k In the expression of (2), ω is a two-dimensional window ω k The total number of pixel points in (a),is p in a two-dimensional window omega k Average value of μ k Is the guiding image I in a two-dimensional window omega k Average value of>Is the guiding image I in a two-dimensional window omega k In (a) is the variance of epsilon is the prevention of a k Oversized regularization parameters due to the two-dimensional window ω k The pixel position i is covered by the pixel points with the number of |omega| so that a is utilized k And b k Mean value of>And->The effect of removing noise is achieved;
taking the image after the first guide filtering as the guide image of the next guide filtering, namely continuously updating the guide image in the iterative process, wherein the iterative guide filtering formula is as follows:
in the formula, G represents guide filtering,expressed in q in the t-th iteration t-1 The guiding filtering process is that the guiding image and p is the input image, and the output image is q t R represents the radius of the two-dimensional window, s is a smoothing factor, and the inverse of the regularization parameter is taken and rounded down;
constructing an iteration stopping criterion of guided filtering by using the structural similarity index SSIM, wherein the structural similarity and the iteration stopping criterion are expressed as follows:
|SSIM (n+1) -SSIM (n) |<ζ,n=1,2,3,...,m
wherein mu is t Sum mu t+1 Respectively representing the output image q after guiding and filtering at the time t and the time t+1 t And q t+1 Mean, sigma of t Sum sigma t+1 Respectively representing the output image q after guiding and filtering at the time t and the time t+1 t And q t+1 Standard deviation of sigma of t(t+1) At tGuide filter for moment of moment t+1 post-processing output image q t And q t+1 Covariance of C 1 And C 2 For a constant, let ζ be the iterative guided filter stop factor, ζ be an adjustable positive number much smaller than 1, SSIM (n) Structural similarity value representing nth order guided filtered output image and n-1 order guided filtered output image, SSIM (n+1) Structural similarity value representing the n+1th order guided filtered output image and the n order guided filtered output image, when SSIM (n+1) With SSIM (n) When the absolute value of the difference is much smaller than ζ, it can be considered that the guided filtering effect after the iteration hardly changes.
2. The method for detecting defects of a background-filter-based LCD light guide plate as claimed in claim 1, wherein,
s130, performing frequency domain filtering on the image subjected to the guide filtering of the LCD light guide plate by adopting a fast Fourier transform method, wherein the method specifically comprises the following steps:
the image after the guide filtering is marked as F (x, y), the image is subjected to two-dimensional discrete Fourier transform to obtain a corresponding frequency domain image F (u, v), and the F (u, v) is subjected to modulo calculation, and the formula is as follows:
wherein M and N represent the size of the image, the frequency domain of the image is represented by F (u, v), u and v are frequency variables, u=0, 1,2,..m-1; v=0, 1, 2..n-1, x and y are space variables, R (u, v) and I (u, v) represent the real and imaginary parts of F (u, v), respectively;
calculating a power spectrum image, and carrying out Gaussian high-pass filtering on the obtained power spectrum image, wherein the formula is as follows:
P(u,v)=|F(u,v)| 2
P G (u,v)=G(u,v)P(u,v)
wherein P (u, v) is a power spectrum image, P G (u, v) is a power spectrum image after Gaussian high pass, and G (u, v) is a Gaussian filter;
using global threshold pairs P G The highlight region in (u, v) is segmented as follows:
T MinGrayValue <g<T MaxGrayValue
wherein T is MinGrayValue Representing a pixel gray value lower threshold value, T, adopted for segmentation MaxGrayValue Representing an upper limit threshold value of gray values of pixel points adopted by segmentation; g is the gray value of the pixel point meeting the condition;
screening the region subjected to global threshold segmentation according to the area size, wherein the area value occupied by each pixel point is recorded as 1, and the condition of region screening is as follows:
T MinAera <r<T MaxAera
wherein T is MinAera Represents the area lower threshold value, T, adopted by screening MaxAera Representing an upper threshold of area employed for screening;
the respective centroid is calculated for all the sub-regions screened out, and the formula is as follows:
wherein u is m And v m Is the coordinate of the centroid, N r The number of pixels in each sub-area selected as the connected area;
taking the mass center of each sub-region as a circle center, taking R as a radius as a circle, and setting the gray value of the pixel point in the circular region to be zero;
the frequency domain image from which the high-frequency information is removed is subjected to two-dimensional inverse discrete Fourier transform to obtain a reconstructed image from which the background is removed; the formula is as follows:
3. the method for detecting defects of a background-filter-based LCD light guide plate as claimed in claim 1, wherein,
and S140, dividing a defect area of the image after the frequency domain filtering of the LCD light guide plate by using a dynamic threshold method to obtain a detection result, wherein the detection result comprises the following steps of:
let the reconstructed image after the frequency domain filtering be g (s, t), carry out the mean filtering to the reconstructed image g (s, t), the formula is as follows:
wherein C is xy Indicating that the center point of the mean filter is at (x, y), and the window size of the mean filter is m and n;
let beta be the offset term, filter the mean value to the image f 2 And (x, y) comparing the obtained image with the reconstructed image after frequency domain filtering pixel by pixel, wherein the fluctuation of the gray value of the pixel point in the range of the offset term beta is regarded as an acceptable gray value, and the fluctuation outside the range of the offset term beta is regarded as the gray value of the pixel point of the target area, namely, the detection of the target defect is completed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910389155.1A CN110084768B (en) | 2019-05-10 | 2019-05-10 | Defect detection method of LCD light guide plate based on background filtering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910389155.1A CN110084768B (en) | 2019-05-10 | 2019-05-10 | Defect detection method of LCD light guide plate based on background filtering |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110084768A CN110084768A (en) | 2019-08-02 |
CN110084768B true CN110084768B (en) | 2023-06-20 |
Family
ID=67419603
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910389155.1A Active CN110084768B (en) | 2019-05-10 | 2019-05-10 | Defect detection method of LCD light guide plate based on background filtering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110084768B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110530883B (en) * | 2019-09-30 | 2022-08-02 | 凌云光技术股份有限公司 | Defect detection method |
CN113658154A (en) * | 2021-08-24 | 2021-11-16 | 凌云光技术股份有限公司 | Image detection method and device based on frequency domain periodic texture removal |
US20240020819A1 (en) * | 2022-07-12 | 2024-01-18 | Onto Innovation Inc. | Periodic-pattern background removal |
CN114998354A (en) * | 2022-08-08 | 2022-09-02 | 江苏多孚多智能装备科技有限公司 | Steel belt roll mark detection method based on graph filtering |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108416740A (en) * | 2018-01-22 | 2018-08-17 | 大连大学 | A kind of iteration self-adapting median filtering algorithm for eliminating salt-pepper noise |
CN108876768A (en) * | 2018-05-30 | 2018-11-23 | 杭州舜浩科技有限公司 | Light guide plate shadow defect inspection method |
CN109102497A (en) * | 2018-07-13 | 2018-12-28 | 杭州舜浩科技有限公司 | High-resolution light guide plate image defect detection method |
-
2019
- 2019-05-10 CN CN201910389155.1A patent/CN110084768B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108416740A (en) * | 2018-01-22 | 2018-08-17 | 大连大学 | A kind of iteration self-adapting median filtering algorithm for eliminating salt-pepper noise |
CN108876768A (en) * | 2018-05-30 | 2018-11-23 | 杭州舜浩科技有限公司 | Light guide plate shadow defect inspection method |
CN109102497A (en) * | 2018-07-13 | 2018-12-28 | 杭州舜浩科技有限公司 | High-resolution light guide plate image defect detection method |
Also Published As
Publication number | Publication date |
---|---|
CN110084768A (en) | 2019-08-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110084768B (en) | Defect detection method of LCD light guide plate based on background filtering | |
CN111383209B (en) | Unsupervised flaw detection method based on full convolution self-encoder network | |
CN106556940B (en) | A kind of background suppression method in TFT-LCD screen automatic optics inspection | |
CN110286124B (en) | Machine vision-based refractory brick measuring system | |
CN107657606B (en) | Method and device for detecting brightness defect of display device | |
KR101635461B1 (en) | System and Method for Automatically Detecting a Mura Defect using Mask Filtering in Wavelet Transform | |
CN115330784B (en) | Cloth surface defect detection method | |
CN109632808B (en) | Edge defect detection method and device, electronic equipment and storage medium | |
Li et al. | Automatic saw-mark detection in multicrystalline solar wafer images | |
CN109102497B (en) | High-resolution light guide plate image defect detection method | |
CN113646801A (en) | Defect detection method and device for defect image and computer readable storage medium | |
CN113160088B (en) | Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy | |
CN114881915A (en) | Symmetry-based mobile phone glass cover plate window area defect detection method | |
CN109064439B (en) | Partition-based single-side light-entering type light guide plate shadow defect extraction method | |
CN116030055B (en) | Quality detection method for elevator-related accessories | |
CN107895371B (en) | Textile flaw detection method based on peak coverage value and Gabor characteristics | |
CN108830851B (en) | LCD rough spot defect detection method | |
CN115131354A (en) | Laboratory plastic film defect detection method based on optical means | |
Ma et al. | An automatic detection method of Mura defects for liquid crystal display | |
CN114332081B (en) | Textile surface abnormity determination method based on image processing | |
Choi et al. | Real-time defects detection algorithm for high-speed steel bar in coil | |
CN115797473A (en) | Concrete forming evaluation method for civil engineering | |
CN110400319A (en) | A kind of spinning cake greasy dirt partitioning algorithm based on domain division method | |
CN113129265B (en) | Method and device for detecting surface defects of ceramic tiles and storage medium | |
CN116958127B (en) | Silicon wafer quality assessment method based on image features |
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 |