CN115131336A - Dark line defect detection method under four-color background picture of display screen - Google Patents
Dark line defect detection method under four-color background picture of display screen Download PDFInfo
- Publication number
- CN115131336A CN115131336A CN202210867075.4A CN202210867075A CN115131336A CN 115131336 A CN115131336 A CN 115131336A CN 202210867075 A CN202210867075 A CN 202210867075A CN 115131336 A CN115131336 A CN 115131336A
- Authority
- CN
- China
- Prior art keywords
- image
- color
- detected
- display screen
- defect detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 55
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 claims abstract description 27
- 239000013598 vector Substances 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 238000012549 training Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 238000007792 addition Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 abstract description 5
- 230000009466 transformation Effects 0.000 abstract description 3
- 230000000737 periodic effect Effects 0.000 abstract 1
- 238000012706 support-vector machine Methods 0.000 description 10
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004020 luminiscence type Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G06T5/70—
-
- G06T5/90—
-
- 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
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a dark line defect detection method under a four-color background picture of a display screen, which comprises the steps of preprocessing an image to be detected, carrying out low-pass filtering on the image to be detected by adopting a Butterworth filter, and removing periodic textures of the display screen; then, carrying out color cast detection on the image to be detected, converting the image to be detected into column vectors, inputting the column vectors into the trained SVM model for classification, and marking color cast areas to obtain a color cast detection diagram; then, dark line defect detection is carried out on the image to be detected, the image to be detected is transferred to an HSI color space, and self-adaptive variable interval gamma transformation is carried out on the I space component to obtain a color equilibrium image; and finally, processing the color balance image by adopting a Sauvola local threshold segmentation method to obtain a defect detection binary image, and finishing defect detection. The method can detect the dark line display defects under the four-color background picture (including the condition of color cast display) of the display screen, and has the characteristic of real-time display defect detection.
Description
Technical Field
The invention relates to an automatic detection technology of display screen defects, in particular to a dark line defect detection method under a four-color background picture of a display screen.
Background
With the development of science and technology, TFT-LCD has advantages of high resolution, high brightness, small size, etc. and is widely used, which prompts manufacturers to concentrate on improving product quality and production efficiency, and product detection is an important link to hold product quality. In the quality inspection of display screen products, because the display four-color picture comprises RGB basic colors and composite white color, the image and color display characteristics of the TFT-LCD can be well shown, and therefore, the display defect detection of the display luminescence unit is generally required to be carried out under the background mode of displaying the four-color picture (red, green, blue and white).
A defect detection method of a mobile phone glass screen based on a difference image method is provided in the literature 'Jianchuanxia, high-key, mobile phone glass screen surface defect visual detection method research [ J ]. packaging engineering', a registration method of mutual information is adopted to realize registration of a template image and an image to be detected, a difference image is obtained through difference operation, and finally a defect detection binary image is obtained through an Niblack method. In The documents "Yu Cui, Liquid crystal Display devices in multiple background with visual real-time detection [ J ], Journal Of The morphology For Information Display", The grid is used as The detection background, and The Otsu method and The particle-based morphology processing method are used to accurately identify The edge defect, but The defect can only be detected when The TFT-LCD displays a pure color picture. Therefore, these methods cannot perform dark line defect detection while displaying a four-color background picture.
Disclosure of Invention
The invention aims to provide a dark line defect detection method under a four-color background picture of a display screen.
In order to achieve the purpose, the invention is implemented according to the following technical scheme:
the invention comprises the following steps:
s1, preprocessing an image to be detected, converting the image to be detected from a space domain to a frequency domain by utilizing Fourier transform, carrying out low-pass filtering on a spectrogram, and converting the filtered spectrogram back to the space domain to obtain the image to be detected with surface textures removed;
s2, converting the image to be detected into a column vector and inputting the column vector into an SVM (support vector machine) for color cast detection to obtain a color cast detection binary image;
and S3, transferring the image to be detected to an HSI color space, and carrying out self-adaptive change interval Gamma correction on the I component to obtain a color balance image.
S4, using a Sauvula local dynamic threshold segmentation method to rapidly perform threshold segmentation on the color equilibrium image by utilizing the integral image to obtain a defect detection binary image;
and S5, processing the defect detection binary image by a mathematical morphology closed operation method to obtain a defect-free mask image, and carrying out differential operation on the mask image and the defect detection binary image to obtain a final result.
The invention has the beneficial effects that:
the invention relates to a dark line defect detection method under a four-color background picture of a display screen, which is compared with the prior art. And then, carrying out self-adaptive variable interval gamma transformation on the image histogram to obtain a color equilibrium image. And then obtaining a binary image by using a Sauvula local threshold segmentation method, and finally obtaining a defect detection binary image through post-processing to finish defect detection. The method can realize the detection of the defects of the display screen under the multi-color background, can completely segment the defect parts, and has good application prospect.
Drawings
FIG. 1 is a histogram of image I of the present invention;
FIG. 2 is a schematic flow chart of an embodiment of the present invention;
FIG. 3 is a schematic diagram of an algorithm architecture according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating defect detection results according to an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to the drawings and specific embodiments, which are provided herein for the purpose of illustrating the invention and are not to be construed as limiting the invention.
The specific flow of the display screen defect detection method is shown in fig. 2, the algorithm architecture is shown in fig. 3, and the method specifically comprises the following steps:
s1, converting the to-be-detected image from the airspace to the frequency domain by utilizing Fourier transform, carrying out low-pass filtering on the spectrogram, and converting the filtered spectrogram back to the airspace to obtain the to-be-detected image with surface textures removed. The method specifically comprises the following substeps:
s11, carrying out fast Fourier transform on the image to be detected, wherein the two-dimensional Fourier transform is as shown in a formula (1). Transferring the image from the spatial domain to the frequency domain, and moving the zero frequency component to the center of the frequency spectrum to obtain a spectrogram of the image to be detected;
where F (x, y) represents an image matrix of size mxn, x is 0,1,2, …, m-1, y is 0,1,2, …, n-1, F (u, v) represents the fourier transform of F (x, y), the coordinate system in which F (u, v) is located is called the frequency domain, and the mxn matrix defined by u is 0,1,2, …, m-1 and v is 0,1,2, …, n-1 is called the frequency domain matrix.
S12, performing low-pass filtering on the spectrogram by using a Butterworth low-pass filter, wherein the cut-off frequency is 80.
Wherein D (u, v) represents the distance from the center point (u, v) to the center point in the frequency domain, D0 represents the cut-off frequency, and n represents the number of Butterworth filters.
And S13, finally, converting the filtered spectrogram back to a space domain by using inverse Fourier transform to obtain the to-be-detected image without high-frequency textures.
And S2, a Support Vector Machine (SVM) is a rapid and reliable linear classifier, and the performance is very good under the condition of limited data quantity. The support vector machine can give an optimal segmentation hyperplane for two groups of marked vectors, and can divide the two groups of vectors into two types, so that the distance from the vector closest to the hyperplane in the two groups of vectors to the hyperplane is as far as possible. The SVM model is trained by using a training set, and an optimal decision boundary is obtained. Then, the image to be detected with the size of mxnx3 is converted into mxnxn (r, g, b) vectors with the size of 1 × 3, the mxnxn vectors are input into an SVM classifier to carry out color cast detection, and the input vectors can be divided into two types through decision boundaries, namely a defect type and a background type. And setting the gray value of the defect class as 255 and the gray value of the background class as 0 to obtain a color cast detection binary image.
The method specifically comprises the following substeps:
and S21, collecting a plurality of points from the standard graph as training positive samples, collecting a plurality of points of a defect part from the 3 defect graphs as training negative samples, and training the SVM model. The kernel function employed is a polynomial kernel function:
k(p i ,p j )＝(ap i T p j +c) d (3)
p i ,p j the input vector, in which a is the positive coefficient of the polynomial, c is the coefficient of the constant term, and d represents the degree of the polynomial, can be obtained by training.
S22, converting the to-be-inspected image with the size of m multiplied by n multiplied by 3 into m multiplied by n column vectors with the size of 1 multiplied by 3, inputting the to-be-inspected image into a trained SVM classifier for secondary classification, setting the vector value of a positive sample as (0,0,0), and setting the vector value of a negative sample as (255 ).
And S23, converting the m multiplied by n 1 multiplied by 3 column vectors back to the m multiplied by n multiplied by 3 binary image to finish the color cast detection.
S3, transferring the image to be detected to an HSI (Hue-Saturation-Intensity) color space, wherein H, S, I three parameters are used for describing color characteristics in the HSI color space, and H defines the frequency of color and is called Hue; s represents the shade degree of the color, called saturation; i denotes intensity or brightness. The conversion relationship between the HSI color space and the RGB color space is shown in equation (4). For a given RGB image, the H component of each RGB pixel can be obtained by equation (4):
wherein, the first and the second end of the pipe are connected with each other,
the saturation component S is given by:
the component I is represented by the formula (4). And (3) calculating a histogram of the I image, as shown in the histogram in fig. 1, smoothing the histogram, obtaining a gray scale maximum valley point, and performing adaptive variable interval gamma conversion on the image by using the gray scale maximum valley point to obtain a color balance image.
S31, transferring the image to be detected from the RGB color space to the HSI color space, wherein the expression of the I space is as follows:
s32, calculating a histogram of the image I, and performing convolution smoothing processing on the histogram in order to reduce the influence of histogram noise. Then, a set U of n points P on the smoothed histogram curve is set such that I' (x) is 0, I ″ (x) > 0, where a condition max (x) is satisfied i ) If the point P' where i is 1.. n is called the maximum valley point, λ is 255/x.
S33, carrying out self-adaptive variable interval gamma correction on the image I, wherein the self-adaptive variable interval gamma correction can be expressed as:
t (I) denotes the changed pixel value, I, I 0 And I max The pixel value of the input image, the starting point of the variable interval pixel and the maximum value of the pixel are respectively. γ represents a transform coefficient in the transform process, and γ is generally equal to 0.4 through a plurality of experiments. λ (0 < λ ≦ 1) represents the interval transformation range.
And S4, calculating a dynamic threshold t by using an integral Sauvula method, and performing defect background binarization. And taking the current pixel point as a neighborhood center, calculating the mean value and the standard deviation of the gray level in the neighborhood by using an integral image to dynamically calculate the binarization threshold value of the pixel point, and finally binarizing.
S41. the integrated value of a point (x, y) on the input image T may be defined as the sum of all pixel values above and to the left of (x, y). The integral value of the (x, y) position can be written as:
s42. by integrating the image, calculating the local average m (x, y) for any window size can be obtained by two additions and two subtractions, without the need to go through each pixel value in turn. Let the current pixel point be (x, y), the area of the point be M × n, T (x, y) represents the gray value at (x, y), the average value M (x, y) of gray values in the M × n area and the local variance S 2 (x, y) may be expressed as:
s43, the dynamic threshold t (x, y) at the pixel point (x, y) is:
in this embodiment, the method of equation (12) is used to determine the local optimal threshold t of the residual image.
And S5, in order to extract the defects from the binary image, a mathematical morphology method is required to be used for processing. The closed operation means that the image is expanded and then corroded, fine defects in the foreground image can be filled up through the closed operation, and the overall shape and the position of the image are not changed. By performing the close operation on the binary image, a mask image without defects can be obtained. And carrying out differential operation on the mask image and the binary-free image to obtain a defect part.
FIGS. 4(a) and (c) are images to be inspected containing defects. Wherein, fig. 4(a) is an image to be detected containing line defects, wherein the first and third block regions of the image to be detected from top to bottom contain dark lines, and fig. 4(b) is a binary image of the detection result of the dark lines. Fig. 4(c) is a diagram of an image to be detected containing a color cast defect, in which the brightness of the second block area is too low, and a color cast occurs in the fourth block area, and fig. 4(d) is a binary diagram of a color cast detection result. The method of the invention utilizes an SVM image classifier to carry out color cast detection on the image to obtain a color cast detection binary image. And secondly, the Butterworth low-pass filter and the self-adaptive gamma correction are comprehensively used for equalizing the image, and the influence of the background image on defect detection is eliminated. And finally, carrying out binarization on the image by using an integral Sauvula method and mathematical morphology closed operation and optimizing a processing result to obtain a defect detection binary image. The scheme provided by the invention can detect the display defects under the condition of four-color pictures of the TFT-LCD display screen, namely has the technical characteristics of being suitable for detecting the display dark line defects under the condition that the color difference exists in the four-color pictures of the display screen. And color cast detection can be carried out aiming at color cast defects without specific shapes, and dark point and dark line detection can be carried out after the color cast detection is finished.
The technical solution of the present invention is not limited to the limitations of the above specific embodiments, and all technical modifications made according to the technical solution of the present invention fall within the protection scope of the present invention.
Claims (5)
1. A dark line defect detection method under a four-color background picture of a display screen is characterized by comprising the following steps:
s1: preprocessing an image to be detected: converting the to-be-detected image from a space domain to a frequency domain by utilizing Fourier transform, carrying out low-pass filtering on the spectrogram, and converting the filtered spectrogram back to the space domain to obtain the to-be-detected image with surface textures removed;
s2: converting the image to be detected into a column vector and inputting the column vector into a trained SVM model for color cast detection to obtain a color cast detection binary image;
s3: transferring the image to be detected to an HSI color space, and carrying out Gamma correction on the component I in a self-adaptive change interval to obtain a color balance image;
s4: performing threshold segmentation on the color equilibrium image quickly by using a Sauvula local dynamic threshold segmentation method and utilizing an integral image to obtain a defect detection binary image;
s5: and processing the defect detection binary image by a mathematical morphology closed operation method to obtain a defect-free mask image, and performing differential operation on the mask and the defect detection binary image to obtain a final result.
2. The method for detecting the dark line defect of the display screen under the four-color background picture according to claim 1, wherein: the step S1 specifically includes the following sub-steps:
s11: performing fast Fourier transform on the image to be detected, wherein the two-dimensional Fourier transform is as shown in formula (1), namely, converting the image from a space domain to a frequency domain, and moving a zero-frequency component to a frequency spectrum center to obtain a spectrogram of the image to be detected;
wherein F (x, y) represents an image matrix with size of m × n, x is 0,1,2, …, m-1, y is 0,1,2, …, n-1, F (u, v) represents fourier transform of F (x, y), coordinate system of F (u, v) is called frequency domain, and m × n matrix defined by u is 0,1,2, …, m-1 and v is 0,1,2, …, n-1 is called frequency domain matrix;
s12: low-pass filtering the spectrogram by using a Butterworth low-pass filter, wherein the cut-off frequency is 80;
wherein D (u, v) represents the distance from the center point (u, v) to the center point in the frequency domain, D0 represents the cut-off frequency, and n represents the number of Butterworth filters;
s13: and finally, converting the filtered spectrogram back to a space domain by using inverse Fourier change to obtain the to-be-detected image without high-frequency textures.
3. The method for detecting the dark line defect of the display screen under the four-color background picture according to claim 1, wherein: the step S2 specifically includes the following sub-steps:
s21: collecting a plurality of points from the standard graph as training positive samples, collecting a plurality of points of the defect part from the 3 defect graphs as training negative samples, and training the SVM model; the kernel function employed is a polynomial kernel function:
p i ,p j is an input vector, wherein a is a positive coefficient of a polynomial, c is a coefficient of a constant term, and d represents the degree of the polynomial;
s22: converting the to-be-detected image with the size of mxnx3 into mxn 1 × 3 column vectors, inputting the m × n 1 × 3 column vectors into a trained SVM classifier for secondary classification, setting the vector values of the positive samples to be 0,0 and 0, and setting the vector values of the negative samples to be 255,255 and 255;
s23: and (4) converting the m × n 1 × 3 column vectors back to the m × n × 3 binary image to complete color cast detection.
4. The method for detecting the dark line defect of the display screen under the four-color background picture according to claim 1, wherein: the step S3 specifically includes the following sub-steps:
s31: transferring the image to be detected from the RGB color space to the HSI color space, wherein the expression of the I space is as follows:
s32: in order to reduce the histogram noise effect, the histogram of image I needs to be first subjected to convolution smoothing, and then a set U of n points P on the smoothed histogram curve is set such that I' (x) is 0, I ″ (x) > 0, where the condition max (x) is satisfied i ) N, which is called the maximum valley point, λ is 255/x;
s33: carrying out adaptive variable interval gamma correction on the image I, and expressing as follows:
t (I) denotes the changed pixel value, I, I 0 And I max Respectively representing the pixel value of an input image, the starting point of a variable interval pixel and the maximum value of the pixel; γ represents a transform coefficient in the transform process.
5. The method for detecting the dark line defect of the display screen under the four-color background picture according to claim 1, wherein: the step S4 specifically includes the following sub-steps:
s41: the integrated value of a point (x, y) on the input image T is defined as the sum of all pixel values above and to the left of (x, y); the integral value of the (x, y) position can be written as:
s42: by integrating the image, calculating the local average m (x, y) for any window size can be obtained by two additions and two subtractions, without having to go through each pixel value in turn; let the current pixel point be (x, y), the area of the point be M × n, T (x, y) represents the gray value at (x, y), the average value M (x, y) of gray values in the M × n area and the local variance S 2 (x, y) is represented as:
s43: the dynamic threshold t at pixel point (x, y) is then:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210867075.4A CN115131336B (en) | 2022-07-22 | 2022-07-22 | Dark line defect detection method under four-color background picture of display screen |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210867075.4A CN115131336B (en) | 2022-07-22 | 2022-07-22 | Dark line defect detection method under four-color background picture of display screen |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115131336A true CN115131336A (en) | 2022-09-30 |
| CN115131336B CN115131336B (en) | 2023-11-14 |
Family
ID=83383629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210867075.4A Active CN115131336B (en) | 2022-07-22 | 2022-07-22 | Dark line defect detection method under four-color background picture of display screen |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115131336B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115171A (en) * | 2023-10-25 | 2023-11-24 | 苏州视达讯远电子科技有限公司 | Slight bright point defect detection method applied to subway LCD display screen |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015194567A (en) * | 2014-03-31 | 2015-11-05 | 三菱電機株式会社 | display device |
| CN106204459A (en) * | 2014-09-02 | 2016-12-07 | 台北科技大学 | The method of image visibility reparation and image processing apparatus thereof |
| CN106650770A (en) * | 2016-09-29 | 2017-05-10 | 南京大学 | Mura defect detection method based on sample learning and human visual characteristics |
| CN109493291A (en) * | 2018-10-29 | 2019-03-19 | 南京邮电大学 | A kind of method for enhancing color image contrast ratio of adaptive gamma correction |
| CN111161246A (en) * | 2019-12-30 | 2020-05-15 | 歌尔股份有限公司 | Product defect detection method, device and system |
| CN114757913A (en) * | 2022-04-15 | 2022-07-15 | 电子科技大学 | Display screen defect detection method |
-
2022
- 2022-07-22 CN CN202210867075.4A patent/CN115131336B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015194567A (en) * | 2014-03-31 | 2015-11-05 | 三菱電機株式会社 | display device |
| CN106204459A (en) * | 2014-09-02 | 2016-12-07 | 台北科技大学 | The method of image visibility reparation and image processing apparatus thereof |
| CN106650770A (en) * | 2016-09-29 | 2017-05-10 | 南京大学 | Mura defect detection method based on sample learning and human visual characteristics |
| CN109493291A (en) * | 2018-10-29 | 2019-03-19 | 南京邮电大学 | A kind of method for enhancing color image contrast ratio of adaptive gamma correction |
| CN111161246A (en) * | 2019-12-30 | 2020-05-15 | 歌尔股份有限公司 | Product defect detection method, device and system |
| CN114757913A (en) * | 2022-04-15 | 2022-07-15 | 电子科技大学 | Display screen defect detection method |
Non-Patent Citations (2)
| Title |
|---|
| 戴声奎 等: "无色恒常性原理与CLC白平衡方法", 信号处理, no. 04 * |
| 贾坤昊;夹尚丰;杨栩;余振军;蔡丽杰;李志国;孙林;: "改进Niblack算法及其在不均匀光照条件下的应用", 软件导刊, no. 04 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115171A (en) * | 2023-10-25 | 2023-11-24 | 苏州视达讯远电子科技有限公司 | Slight bright point defect detection method applied to subway LCD display screen |
| CN117115171B (en) * | 2023-10-25 | 2024-01-26 | 苏州视达讯远电子科技有限公司 | Slight bright point defect detection method applied to subway LCD display screen |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115131336B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
| CN113450307B (en) | Product edge defect detection method | |
| CN107507173B (en) | No-reference definition evaluation method and system for full-slice image | |
| CN111161222B (en) | Printing roller defect detection method based on visual saliency | |
| CN104537634B (en) | The method and system of raindrop influence is removed in dynamic image | |
| CN111915704A (en) | Apple hierarchical identification method based on deep learning | |
| CN112132196B (en) | Cigarette case defect identification method combining deep learning and image processing | |
| CN114118144A (en) | Anti-interference accurate aerial remote sensing image shadow detection method | |
| CN112583999B (en) | Method for detecting lens dirt of camera module | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN112862744B (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
| CN116363140B (en) | Method, system and device for detecting defects of medium borosilicate glass and storage medium | |
| CN115131336B (en) | Dark line defect detection method under four-color background picture of display screen | |
| CN111223110A (en) | Microscopic image enhancement method and device and computer equipment | |
| Ma et al. | An automatic detection method of Mura defects for liquid crystal display | |
| CN115578660A (en) | Land block segmentation method based on remote sensing image | |
| CN114332081A (en) | Textile surface abnormity determination method based on image processing | |
| US20240086661A1 (en) | Method and apparatus for processing graphic symbol and computer-readable storage medium | |
| CN112801141B (en) | Heterogeneous image matching method based on template matching and twin neural network optimization | |
| CN112802074A (en) | Textile flaw detection method based on illumination correction and visual saliency characteristics | |
| Yusoff et al. | Enhancement and Segmentation of Ziehl Neelson Sputum Slide Images using Contrast Enhancement and Otsu Threshold Technique | |
| Zhang et al. | A LCD screen Mura defect detection method based on machine vision | |
| CN114092441A (en) | Product surface defect detection method and system based on dual neural network | |
| CN115187788A (en) | Crop seed automatic counting method based on machine vision |
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 |