CN117611551A - Display screen abnormality detection method and system based on Internet of things - Google Patents
Display screen abnormality detection method and system based on Internet of things Download PDFInfo
- Publication number
- CN117611551A CN117611551A CN202311591921.5A CN202311591921A CN117611551A CN 117611551 A CN117611551 A CN 117611551A CN 202311591921 A CN202311591921 A CN 202311591921A CN 117611551 A CN117611551 A CN 117611551A
- Authority
- CN
- China
- Prior art keywords
- display screen
- value
- image
- pixel
- internet
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 60
- 230000005856 abnormality Effects 0.000 title claims abstract description 37
- 230000007547 defect Effects 0.000 claims abstract description 109
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000001914 filtration Methods 0.000 claims abstract description 26
- 238000012937 correction Methods 0.000 claims abstract description 17
- 230000011218 segmentation Effects 0.000 claims description 35
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 239000010409 thin film Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/30121—CRT, LCD or plasma display
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a display screen abnormality detection method and system based on the Internet of things, comprising the following steps: acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images; carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to obtain original characteristics of image defects; the method and the device can eliminate the periodical repeated texture background to improve the detection accuracy of the display screen by carrying out texture background reconstruction, and can accurately acquire the defect type of the display screen by carrying out the threshold judgment of the defect area of the display screen, thereby being capable of rapidly improving the detection efficiency of the display screen and saving a large amount of labor cost.
Description
Technical Field
The invention relates to the technical field of the Internet of things, in particular to a display screen abnormality detection method and system based on the Internet of things.
Background
The LCD plays an important role in information transmission in the current information society, and has revolutionized in the display industry in the 21 st century, and the LCD panel of the thin film transistor gradually replaces the traditional CRT display, so that the LCD panel becomes a mainstream display terminal, and has the characteristics of thinness, lightness, power saving, high brightness and contrast, and the like, so that the LCD panel is widely applied in different display fields.
In order to meet the market demands, on one hand, the display screen is promoted to develop towards the directions of high resolution, low power consumption, large size and wide viewing angle, and on the other hand, the display screen manufacturers are forced to improve the production process, however, in the production and assembly processes of the display screen, various display screen defects are unavoidable in spite of strict control of each link.
Therefore, the display screen defect detection is an indispensable link in the product appearance detection process. The traditional manual detection method has the advantages of large workload, strong subjectivity, difficult quantification of detection standards and high misjudgment rate, cannot meet the industrial automation requirement of large-scale detection of notebook computers, cannot accurately collect detection data in real time, supports data analysis, and improves the production flow. Therefore, the automatic detection technology for the defects of the display screen is studied deeply, and the method has important significance for improving the quality inspection level of the notebook computer and improving the production flow.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a display screen abnormality detection method based on the Internet of things, which comprises the following steps:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to obtain original characteristics of image defects;
and judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects, and obtaining the abnormal detection result of the display screen according to the threshold value judgment result.
Preferably, the acquiring pixels of the target display screen based on the internet of things, acquiring a pixel image, and reconstructing a texture background of the pixel image, and acquiring a reconstructed texture background image includes:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and performing matrix transformation on the pixel images to obtain a grid texture background image matrix;
singular value decomposition is carried out on the grid-shaped texture background image matrix through a singular value decomposition method, so that a preset number of singular values are obtained;
performing scale screening on the singular values of the preset number to obtain singular values of a preset numerical value;
and reconstructing the image texture background according to the singular value of the preset numerical value, and obtaining a reconstructed texture background image.
Preferably, the calculation formula of the reconstructed texture background image is as follows:
wherein B represents a reconstructed texture background image; lambda (lambda) i Representing the ith singular value; u (u) i An ith filtering scale in the data representing the pixel image; v i Representing the i-th reconstruction parameter value; k represents a preset number.
Preferably, the performing global brightness non-uniformity correction on the reconstructed texture background image by using a mean filtering image difference method to obtain original features of image defects includes:
filtering the grid texture image after the background texture is restrained by a mean filter with a preset window specification to obtain a mean filtering image;
and carrying out differential correction on the mean value filtered image and the grid texture image after the background texture is restrained, and obtaining original characteristics of the image defects.
Preferably, the step of judging the threshold value of the defect area of the display screen for the original feature of the image defect, and obtaining the abnormal detection result of the display screen according to the threshold value judgment result includes:
performing image binarization processing on the original characteristics of the image defects to obtain target foreground pixels;
based on the target foreground pixel, calculating an eight-neighborhood connected domain of the target foreground pixel;
marking the defect area of the eight neighborhood connected domain obtained by calculation, and determining a display screen defect candidate area;
traversing the display screen candidate region, calculating the contrast of the traversed display screen candidate region, and obtaining the contrast of the display screen candidate region;
performing defect segmentation on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region to obtain a segmentation result;
performing defect threshold judgment according to the segmentation result to obtain a threshold judgment result;
and carrying out defect marking according to the threshold judgment result, obtaining marking information, and classifying display screen defects according to the marking information to obtain display screen abnormality detection results.
Preferably, the display screen abnormality detection result at least includes one or more of the following: punctiform defects, blocky defects and Mura defects.
Preferably, the calculation formula of the contrast ratio of the candidate area of the display screen is as follows:
wherein,
wherein,
wherein g (x, y) represents a gray value of a pixel having coordinates (x, y); r is R defeat Representing a candidate region of the display screen; r is R bg A background region representing a candidate region of the display screen; a, a c The abscissa of the center pixel which represents the minimum circumscribed positive rectangle in the center of the spatial position of the background area; b c The ordinate representing the center pixel of the minimum circumscribed positive rectangle in the space position center of the background area; weight (x, y) represents the weight value of a pixel with coordinates (x, y); l represents the sum of weights of the (x, y) pixel in coordinates; w represents a contrast parameter; m represents the total number of pixels in the candidate region of the display screen; q represents the width value of a rectangular pixel block with the background area position center as the rectangular center; h represents the height value of a rectangular pixel block centered on the background area position center.
Preferably, the defect segmentation is performed on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region, and a segmentation result is obtained, which comprises the following specific steps:
step S1: setting initialization parameters according to the contrast ratio of the candidate areas of the display screen;
step S2: calculating a first cluster center value and a first membership value of the candidate region of the display screen based on the initialization parameters;
step S3: judging whether the Euclidean distance from the first clustering center value to a preset clustering center value is smaller than or equal to an iteration stop threshold value, and obtaining a first judging result;
step S4: stopping iteration when the first judgment result is yes, and obtaining a segmentation result;
step S5: when the first judgment result is negative, performing two rounds of iteration according to the first clustering central value and the first membership value, and calculating a second clustering central value and a second membership value …;
step S6: judging whether the Euclidean distance from the n-th clustering central value to the n-1 clustering central value is smaller than or equal to a preset iteration stop threshold value, and obtaining an n-th judging result;
step S7: returning to the step S5 to carry out the n+1th iteration when the n judgment result is negative;
step S8: and stopping iteration when the nth judging result is yes, and obtaining a segmentation result.
Preferably, the iterative calculation formula of the cluster center value and the membership value is as follows:
wherein,
wherein mu is ij Representing an ith class membership value corresponding to a jth sample point; θ i A cluster center value representing the i-th class; σ represents the number of categories of data in the target foreground pixel, where k=1, 2, …, σ; delta represents a membership factor; x is x j Represents the j-th sample point, where j=1, 2, …, n; II x j -θ i II represents the Euclidean distance from the jth sample point to the cluster center value of the ith class;a representation; the membership factor delta corresponds to the class i membership value corresponding to the jth sample point.
Based on the same inventive concept, the invention also provides a display screen abnormality detection system based on the Internet of things, comprising:
the image reconstruction module is used for acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
the feature acquisition module is used for carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to acquire original features of image defects;
and the abnormality detection module is used for judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects and obtaining the abnormality detection result of the display screen according to the threshold value judgment result.
Preferably, the image reconstruction module is specifically configured to:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and performing matrix transformation on the pixel images to obtain a grid texture background image matrix;
singular value decomposition is carried out on the grid-shaped texture background image matrix through a singular value decomposition method, so that a preset number of singular values are obtained;
performing scale screening on the singular values of the preset number to obtain singular values of a preset numerical value;
and reconstructing the image texture background according to the singular value of the preset numerical value, and obtaining a reconstructed texture background image.
Preferably, the calculation formula of the reconstructed texture background image in the image reconstruction module is as follows:
wherein B represents a reconstructed texture background image; lambda (lambda) i Representing the ith singular value; u (u) i An ith filtering scale in the data representing the pixel image; v i Representing the i-th reconstruction parameter value; k represents a preset number.
Preferably, the feature acquisition module is specifically configured to:
filtering the grid texture image after the background texture is restrained by a mean filter with a preset window specification to obtain a mean filtering image;
and carrying out differential correction on the mean value filtered image and the grid texture image after the background texture is restrained, and obtaining original characteristics of the image defects.
Preferably, the abnormality detection module is specifically configured to:
performing image binarization processing on the original characteristics of the image defects to obtain target foreground pixels;
based on the target foreground pixel, calculating an eight-neighborhood connected domain of the target foreground pixel;
marking the defect area of the eight neighborhood connected domain obtained by calculation, and determining a display screen defect candidate area;
traversing the display screen candidate region, calculating the contrast of the traversed display screen candidate region, and obtaining the contrast of the display screen candidate region;
performing defect segmentation on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region to obtain a segmentation result;
performing defect threshold judgment according to the segmentation result to obtain a threshold judgment result;
and carrying out defect marking according to the threshold judgment result, obtaining marking information, and classifying display screen defects according to the marking information to obtain display screen abnormality detection results.
Preferably, the display screen abnormality detection result in the abnormality detection module at least includes one or more of the following: punctiform defects, blocky defects and Mura defects.
Preferably, the calculation formula of the contrast ratio of the candidate region of the display screen in the abnormality detection module is as follows:
wherein,
wherein,
wherein g (x, y) represents a gray value of a pixel having coordinates (x, y); r is R defeat Representing a candidate region of the display screen; r is R bg A background region representing a candidate region of the display screen; a, a c The abscissa of the center pixel which represents the minimum circumscribed positive rectangle in the center of the spatial position of the background area; b c The ordinate representing the center pixel of the minimum circumscribed positive rectangle in the space position center of the background area; weight (x, y) represents the weight value of a pixel with coordinates (x, y); l represents the sum of weights of the (x, y) pixel in coordinates; w represents a contrast parameter; m represents the total number of pixels in the candidate region of the display screen; q is expressed in the backgroundThe center of the area position is the width value of a rectangular pixel block with a rectangular center; h represents the height value of a rectangular pixel block centered on the background area position center.
Preferably, in the abnormality detection module, according to the contrast of the candidate region of the display screen, the candidate region of the display screen is subjected to defect segmentation by a fuzzy C-means clustering algorithm, and a segmentation result is obtained, which specifically comprises the following steps:
step S1: setting initialization parameters according to the contrast ratio of the candidate areas of the display screen;
step S2: calculating a first cluster center value and a first membership value of the candidate region of the display screen based on the initialization parameters;
step S3: judging whether the Euclidean distance from the first clustering center value to a preset clustering center value is smaller than or equal to an iteration stop threshold value, and obtaining a first judging result;
step S4: stopping iteration when the first judgment result is yes, and obtaining a segmentation result;
step S5: when the first judgment result is negative, performing two rounds of iteration according to the first clustering central value and the first membership value, and calculating a second clustering central value and a second membership value …;
step S6: judging whether the Euclidean distance from the n-th clustering central value to the n-1 clustering central value is smaller than or equal to a preset iteration stop threshold value, and obtaining an n-th judging result;
step S7: returning to the step S5 to carry out the n+1th iteration when the n judgment result is negative;
step S8: and stopping iteration when the nth judging result is yes, and obtaining a segmentation result.
Preferably, the iterative calculation formula of the cluster center value and the membership value in the anomaly detection module is as follows:
wherein,
wherein mu is ij Representing an ith class membership value corresponding to a jth sample point; θ i A cluster center value representing the i-th class; σ represents the number of categories of data in the target foreground pixel, where k=1, 2, …, σ; delta represents a membership factor; x is x j Represents the j-th sample point, where j=1, 2, …, n; II x j -θ i II represents the Euclidean distance from the jth sample point to the cluster center value of the ith class;a representation; the membership factor delta corresponds to the class i membership value corresponding to the jth sample point.
Compared with the closest prior art, the invention has the following beneficial effects:
display screen abnormality detection method and system based on the Internet of things comprise the following steps: acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images; carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to obtain original characteristics of image defects; the method and the device can eliminate periodical repeated texture background to improve the detection accuracy of the display screen by carrying out texture background reconstruction on the original characteristics of the image defects and carrying out threshold judgment on the defect areas of the display screen, can clearly distinguish differences between the background and the defects by carrying out global brightness non-uniform correction and can strengthen images of the defect areas of the display screen.
Drawings
Fig. 1 is a schematic flow chart of a display screen abnormality detection method based on the internet of things, which is provided by the invention;
FIG. 2 is a flowchart of a fuzzy C-means clustering algorithm in a display screen abnormality detection method based on the Internet of things, which is provided by the invention;
fig. 3 is a schematic diagram of connection between modules of a display screen abnormality detection system based on the internet of things.
Detailed Description
The following describes the embodiments of the present invention in further detail with reference to the drawings.
Example 1:
the flow chart of the display screen abnormality detection method based on the Internet of things provided by the invention is shown in fig. 1, and comprises the following steps:
step 1: acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
step 2: carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to obtain original characteristics of image defects;
step 3: and judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects, and obtaining the abnormal detection result of the display screen according to the threshold value judgment result.
Specifically, the step 1 includes:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and performing matrix transformation on the pixel images to obtain a grid texture background image matrix;
singular value decomposition is carried out on the grid-shaped texture background image matrix through a singular value decomposition method, so that a preset number of singular values are obtained;
performing scale screening on the singular values of the preset number to obtain singular values of a preset numerical value;
reconstructing the image texture background according to the singular value of the preset numerical value, and obtaining a reconstructed texture background image;
the calculation formula of the reconstructed texture background image is as follows:
wherein B represents a reconstructed texture background image; lambda (lambda) i Representing the ith singular value; u (u) i An ith filtering scale in the data representing the pixel image; v i Representing the i-th reconstruction parameter value; k represents a preset number.
The color filter glass substrate is regularly arranged with pixels with different colors because horizontal and vertical electronic elements are distributed on the display screen;
the pixels can be acquired and extracted through the internet of things technology, and the background texture removing operation of the acquired pixels in the step 1 can be performed, so that the accuracy of subsequent display screen abnormality detection can be ensured.
Specifically, the step 2 includes:
filtering the grid texture image after the background texture is restrained by a mean filter with a preset window specification to obtain a mean filtering image;
and carrying out differential correction on the mean value filtered image and the grid texture image after the background texture is restrained, and obtaining original characteristics of the image defects.
The frequency and direction expression of the filter are similar to those of human eyes, and the filter is sensitive to the edges of the image and insensitive to the illumination change of the image, so that the filter is suitable for expression and separation of textures under the condition of illumination change;
the center frequency of the filter corresponds to the size of a time domain window, and when the center frequency changes, the extracted features contain different frequency components and are reflected to the time domain to be local image features and global image features;
the key point of the abnormal detection of the display screen defect is that the phenomenon of uneven global brightness of the display screen regional image is removed under the condition of retaining original characteristics;
therefore, the average filtering image and the grid texture image after the background texture inhibition are subjected to differential correction, and original defect characteristics can be reserved to the greatest extent, so that the accuracy of detection is improved.
Specifically, the step 3 includes:
performing image binarization processing on the original characteristics of the image defects to obtain target foreground pixels;
based on the target foreground pixel, calculating an eight-neighborhood connected domain of the target foreground pixel;
marking the defect area of the eight neighborhood connected domain obtained by calculation, and determining a display screen defect candidate area;
traversing the display screen candidate region, calculating the contrast of the traversed display screen candidate region, and obtaining the contrast of the display screen candidate region;
performing defect segmentation on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region to obtain a segmentation result;
performing defect threshold judgment according to the segmentation result to obtain a threshold judgment result;
performing defect marking according to the threshold judgment result, obtaining marking information, and classifying display screen defects according to the marking information to obtain display screen abnormality detection results;
the display screen abnormality detection result at least comprises one or more of the following: punctiform defects, blocky defects, and Mura defects;
the display screen candidate area contrast is calculated as follows:
wherein,
wherein,
wherein g (x, y) represents a gray value of a pixel having coordinates (x, y); r is R defeat Representing a candidate region of the display screen; r is R bg A background region representing a candidate region of the display screen; a, a c The abscissa of the center pixel which represents the minimum circumscribed positive rectangle in the center of the spatial position of the background area; b c The ordinate representing the center pixel of the minimum circumscribed positive rectangle in the space position center of the background area; weight (x, y) represents the weight value of a pixel with coordinates (x, y); l represents the sum of weights of the (x, y) pixel in coordinates; w represents a contrast parameter; m represents the total number of pixels in the candidate region of the display screen; q represents the width value of a rectangular pixel block with the background area position center as the rectangular center; h represents the height value of a rectangular pixel block with the background area position center as the rectangular center;
as shown in fig. 2, according to the contrast of the candidate region of the display screen, the candidate region of the display screen is subjected to defect segmentation by a fuzzy C-means clustering algorithm, and a segmentation result is obtained, which specifically comprises the following steps:
step S1: setting initialization parameters according to the contrast ratio of the candidate areas of the display screen;
step S2: calculating a first cluster center value and a first membership value of the candidate region of the display screen based on the initialization parameters;
step S3: judging whether the Euclidean distance from the first clustering center value to a preset clustering center value is smaller than or equal to an iteration stop threshold value, and obtaining a first judging result;
step S4: stopping iteration when the first judgment result is yes, and obtaining a segmentation result;
step S5: when the first judgment result is negative, performing two rounds of iteration according to the first clustering central value and the first membership value, and calculating a second clustering central value and a second membership value …;
step S6: judging whether the Euclidean distance from the n-th clustering central value to the n-1 clustering central value is smaller than or equal to a preset iteration stop threshold value, and obtaining an n-th judging result;
step S7: returning to the step S5 to carry out the n+1th iteration when the n judgment result is negative;
step S8: stopping iteration when the nth judgment result is yes, and obtaining a segmentation result;
the iterative calculation formula of the clustering center value and the membership value is as follows:
wherein,
wherein mu is ij Representing an ith class membership value corresponding to a jth sample point; θ i A cluster center value representing the i-th class; σ represents the number of categories of data in the target foreground pixel, where k=1, 2, …, σ; delta represents a membership factor; x is x j Represents the j-th sample point, where j=1, 2, …, n; II x j -θ i II represents the Euclidean distance from the jth sample point to the cluster center value of the ith class;a representation; the membership factor delta corresponds to the class i membership value corresponding to the jth sample point.
The fuzzy C-means clustering algorithm of the invention carries out defect segmentation on the display screen defect candidate region, can accurately segment the defect region of the fuzzy boundary of the display screen, and carries out threshold judgment on the defect region of the display screen according to the segmentation result, thus accurately obtaining the defect type of the display screen, thereby rapidly improving the detection efficiency of the display screen, saving a great amount of labor cost and avoiding the problems of missing detection, false detection and the like.
Example 2:
the invention provides a display screen abnormality detection system module connection structure diagram based on the Internet of things, which is shown in fig. 3, and comprises: :
the image reconstruction module is used for acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
the feature acquisition module is used for carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to acquire original features of image defects;
and the abnormality detection module is used for judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects and obtaining the abnormality detection result of the display screen according to the threshold value judgment result.
Specifically, the image reconstruction module is specifically configured to:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and performing matrix transformation on the pixel images to obtain a grid texture background image matrix;
singular value decomposition is carried out on the grid-shaped texture background image matrix through a singular value decomposition method, so that a preset number of singular values are obtained;
performing scale screening on the singular values of the preset number to obtain singular values of a preset numerical value;
and reconstructing the image texture background according to the singular value of the preset numerical value, and obtaining a reconstructed texture background image.
The calculation formula of the reconstructed texture background image in the image reconstruction module is as follows:
wherein B represents a reconstructed texture background image; lambda (lambda) i Representing the ith singular value; u (u) i An ith filtering scale in the data representing the pixel image; v i Representing the i-th reconstruction parameter value; k represents a preset number.
Specifically, the feature acquisition module is specifically configured to:
filtering the grid texture image after the background texture is restrained by a mean filter with a preset window specification to obtain a mean filtering image;
and carrying out differential correction on the mean value filtered image and the grid texture image after the background texture is restrained, and obtaining original characteristics of the image defects.
Specifically, the abnormality detection module is specifically configured to:
performing image binarization processing on the original characteristics of the image defects to obtain target foreground pixels;
based on the target foreground pixel, calculating an eight-neighborhood connected domain of the target foreground pixel;
marking the defect area of the eight neighborhood connected domain obtained by calculation, and determining a display screen defect candidate area;
traversing the display screen candidate region, calculating the contrast of the traversed display screen candidate region, and obtaining the contrast of the display screen candidate region;
performing defect segmentation on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region to obtain a segmentation result;
performing defect threshold judgment according to the segmentation result to obtain a threshold judgment result;
and carrying out defect marking according to the threshold judgment result, obtaining marking information, and classifying display screen defects according to the marking information to obtain display screen abnormality detection results.
The display screen abnormality detection result in the abnormality detection module at least comprises one or more of the following: punctiform defects, blocky defects and Mura defects.
The contrast ratio calculation formula of the candidate area of the display screen in the abnormality detection module is as follows:
wherein,
wherein,
wherein g (x, y) represents a gray value of a pixel having coordinates (x, y); r is R defeat Representing a candidate region of the display screen; r is R bg A background region representing a candidate region of the display screen; a, a c The abscissa of the center pixel which represents the minimum circumscribed positive rectangle in the center of the spatial position of the background area; b c The ordinate representing the center pixel of the minimum circumscribed positive rectangle in the space position center of the background area; weight (x, y) represents the weight value of a pixel with coordinates (x, y); l represents the sum of weights of the (x, y) pixel in coordinates; w represents a contrast parameter; m represents the total number of pixels in the candidate region of the display screen; q represents the width value of a rectangular pixel block with the background area position center as the rectangular center; h represents the height value of a rectangular pixel block centered on the background area position center.
The method comprises the following specific steps of:
step S1: setting initialization parameters according to the contrast ratio of the candidate areas of the display screen;
step S2: calculating a first cluster center value and a first membership value of the candidate region of the display screen based on the initialization parameters;
step S3: judging whether the Euclidean distance from the first clustering center value to a preset clustering center value is smaller than or equal to an iteration stop threshold value, and obtaining a first judging result;
step S4: stopping iteration when the first judgment result is yes, and obtaining a segmentation result;
step S5: when the first judgment result is negative, performing two rounds of iteration according to the first clustering central value and the first membership value, and calculating a second clustering central value and a second membership value …;
step S6: judging whether the Euclidean distance from the n-th clustering central value to the n-1 clustering central value is smaller than or equal to a preset iteration stop threshold value, and obtaining an n-th judging result;
step S7: returning to the step S5 to carry out the n+1th iteration when the n judgment result is negative;
step S8: and stopping iteration when the nth judging result is yes, and obtaining a segmentation result.
The iterative calculation formula of the clustering center value and the membership value in the anomaly detection module is as follows:
wherein,
wherein mu is ij Representing an ith class membership value corresponding to a jth sample point; θ i A cluster center value representing the i-th class; σ represents the number of categories of data in the target foreground pixel, where k=1, 2, …, σ; delta represents a membership factor; x is x j Represents the j-th sample point, where j=1, 2, …, n; II x j -θ i II represents the Euclidean distance from the jth sample point to the cluster center value of the ith class;a representation; the membership factor delta corresponds to the class i membership value corresponding to the jth sample point.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that the foregoing embodiments are merely for illustrating the technical solution of the present invention and not for limiting the scope of protection thereof, and although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that various changes, modifications or equivalents may be made to the specific embodiments of the application after reading the present invention, and these changes, modifications or equivalents are within the scope of protection of the claims appended hereto.
Claims (10)
1. The display screen abnormality detection method based on the Internet of things is characterized by comprising the following steps:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to obtain original characteristics of image defects;
and judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects, and obtaining the abnormal detection result of the display screen according to the threshold value judgment result.
2. The method for detecting abnormal display screen based on internet of things according to claim 1, wherein the acquiring pixels of the target display screen based on internet of things, acquiring pixel images, reconstructing texture background of the pixel images, and acquiring reconstructed texture background images comprises:
acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and performing matrix transformation on the pixel images to obtain a grid texture background image matrix;
singular value decomposition is carried out on the grid-shaped texture background image matrix through a singular value decomposition method, so that a preset number of singular values are obtained;
performing scale screening on the singular values of the preset number to obtain singular values of a preset numerical value;
and reconstructing the image texture background according to the singular value of the preset numerical value, and obtaining a reconstructed texture background image.
3. The method for detecting abnormal display screen based on internet of things according to claim 2, wherein the calculation formula of the reconstructed texture background image is as follows:
wherein B represents a reconstructed texture background image; lambda (lambda) i Representing the ith singular value; u (u) i An ith filtering scale in the data representing the pixel image; v i Representation ofAn ith reconstruction parameter value; k represents a preset number.
4. The method for detecting abnormal display screen based on internet of things according to claim 1, wherein the performing global brightness non-uniformity correction on the reconstructed texture background image by means of a mean value filtering image difference method to obtain original image defect features comprises:
filtering the grid texture image after the background texture is restrained by a mean filter with a preset window specification to obtain a mean filtering image;
and carrying out differential correction on the mean value filtered image and the grid texture image after the background texture is restrained, and obtaining original characteristics of the image defects.
5. The method for detecting abnormal display screen based on the internet of things according to claim 1, wherein the performing threshold judgment on the original image defect feature for the display screen defect area, and obtaining the abnormal display screen detection result according to the threshold judgment result comprises:
performing image binarization processing on the original characteristics of the image defects to obtain target foreground pixels;
based on the target foreground pixel, calculating an eight-neighborhood connected domain of the target foreground pixel;
marking the defect area of the eight neighborhood connected domain obtained by calculation, and determining a display screen defect candidate area;
traversing the display screen candidate region, calculating the contrast of the traversed display screen candidate region, and obtaining the contrast of the display screen candidate region;
performing defect segmentation on the display screen defect candidate region through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate region to obtain a segmentation result;
performing defect threshold judgment according to the segmentation result to obtain a threshold judgment result;
and carrying out defect marking according to the threshold judgment result, obtaining marking information, and classifying display screen defects according to the marking information to obtain display screen abnormality detection results.
6. The method for detecting abnormal display screen based on internet of things according to claim 5, wherein the abnormal display screen detection result at least comprises one or more of the following: punctiform defects, blocky defects and Mura defects.
7. The method for detecting abnormal display screen based on internet of things according to claim 5, wherein the contrast of the candidate region of the display screen is calculated as follows:
wherein,
wherein,
wherein g (x, y) represents a gray value of a pixel having coordinates (x, y); r is R defeat Representing a candidate region of the display screen; r is R bg A background region representing a candidate region of the display screen; a, a c The abscissa of the center pixel which represents the minimum circumscribed positive rectangle in the center of the spatial position of the background area; b c The ordinate representing the center pixel of the minimum circumscribed positive rectangle in the space position center of the background area; weight (x, y) represents the weight value of a pixel with coordinates (x, y); l represents the sum of weights of the (x, y) pixel in coordinates; w represents a contrast parameter; m represents the total number of pixels in the candidate region of the display screen; q represents the width value of a rectangular pixel block with the background area position center as the rectangular center; h represents the height value of a rectangular pixel block centered on the background area position center.
8. The method for detecting abnormal display screen based on the internet of things according to claim 5, wherein the defect segmentation is performed on the display screen defect candidate area through a fuzzy C-means clustering algorithm according to the contrast of the display screen candidate area, and a segmentation result is obtained, specifically comprising the following steps:
step S1: setting initialization parameters according to the contrast ratio of the candidate areas of the display screen;
step S2: calculating a first cluster center value and a first membership value of the candidate region of the display screen based on the initialization parameters;
step S3: judging whether the Euclidean distance from the first clustering center value to a preset clustering center value is smaller than or equal to an iteration stop threshold value, and obtaining a first judging result;
step S4: stopping iteration when the first judgment result is yes, and obtaining a segmentation result;
step S5: when the first judgment result is negative, performing two rounds of iteration according to the first clustering central value and the first membership value, and calculating a second clustering central value and a second membership value …;
step S6: judging whether the Euclidean distance from the n-th clustering central value to the n-1 clustering central value is smaller than or equal to an iteration stop threshold value, and obtaining an n-th judging result;
step S7: returning to the step S5 to carry out the n+1th iteration when the n judgment result is negative;
step S8: and stopping iteration when the nth judging result is yes, and obtaining a segmentation result.
9. The method for detecting abnormal display screen based on the internet of things according to claim 8, wherein the iterative calculation formula of the cluster center value and the membership value is as follows:
wherein,
wherein mu is ij Representing an ith class membership value corresponding to a jth sample point; θ i A cluster center value representing the i-th class; σ represents the number of categories of data in the target foreground pixel, where k=1, 2, …, σ; delta represents a membership factor; x is x j Represents the j-th sample point, where j=1, 2, …, n; ||x j -θ i The I represents the Euclidean distance from the j-th sample point to the cluster center value of the i-th class;a representation; the membership factor delta corresponds to the class i membership value corresponding to the jth sample point.
10. The internet of things-based display screen abnormality detection system according to claim 1, comprising:
the image reconstruction module is used for acquiring pixels of a target display screen based on the Internet of things, acquiring pixel images, and reconstructing texture backgrounds of the pixel images to acquire reconstructed texture background images;
the feature acquisition module is used for carrying out global brightness non-uniformity correction on the reconstructed texture background image by a mean value filtering image difference method to acquire original features of image defects;
and the abnormality detection module is used for judging the threshold value of the defect area of the display screen according to the original characteristics of the image defects and obtaining the abnormality detection result of the display screen according to the threshold value judgment result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311591921.5A CN117611551A (en) | 2023-11-24 | 2023-11-24 | Display screen abnormality detection method and system based on Internet of things |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311591921.5A CN117611551A (en) | 2023-11-24 | 2023-11-24 | Display screen abnormality detection method and system based on Internet of things |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117611551A true CN117611551A (en) | 2024-02-27 |
Family
ID=89949294
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311591921.5A Pending CN117611551A (en) | 2023-11-24 | 2023-11-24 | Display screen abnormality detection method and system based on Internet of things |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117611551A (en) |
-
2023
- 2023-11-24 CN CN202311591921.5A patent/CN117611551A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109683360B (en) | Liquid crystal panel defect detection method and device | |
CN107543828B (en) | Workpiece surface defect detection method and system | |
CN107845087A (en) | The detection method and system of the uneven defect of liquid crystal panel lightness | |
CN111861996B (en) | Printed fabric defect detection method | |
JP2003066398A (en) | Method of classifying and processing unevenness of liquid crystal display panel, apparatus for the same and program | |
CN108318491A (en) | A kind of fabric defect detection method based on frequency spectrum curvature analysis | |
CN113034464B (en) | Visual real-time detection method for defects of liquid crystal display under multiple backgrounds | |
CN115775236B (en) | Visual detection method and system for surface micro defects based on multi-scale feature fusion | |
CN112419229A (en) | Display screen linear defect detection method and device and storage medium | |
CN116152242A (en) | Visual detection system of natural leather defect for basketball | |
CN112070762A (en) | Mura defect detection method and device for liquid crystal panel, storage medium and terminal | |
CN116664565A (en) | Hidden crack detection method and system for photovoltaic solar cell | |
CN117392042A (en) | Defect detection method, defect detection apparatus, and storage medium | |
CN113808104B (en) | Metal surface defect detection method and system based on blocking | |
CN109829912B (en) | Defect detection method of TFT array substrate | |
CN116883408A (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
CN116363136B (en) | On-line screening method and system for automatic production of motor vehicle parts | |
CN116152261B (en) | Visual inspection system for quality of printed product | |
Cui et al. | Liquid crystal display defects in multiple backgrounds with visual real‐time detection | |
CN117611551A (en) | Display screen abnormality detection method and system based on Internet of things | |
CN116503340A (en) | Micro oled panel defect detection method, device and equipment | |
CN115330705A (en) | Skin paint surface defect detection method based on adaptive weighting template NCC | |
CN115797314A (en) | Part surface defect detection method, system, equipment and storage medium | |
CN115457323A (en) | Visual inspection system-based method for classifying non-uniform surface defects of mobile phone glass screen | |
CN112419244A (en) | Concrete crack segmentation method and device based on YOLOv4 target detection model and tubular flow field algorithm |
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 |