Display screen defect detection method and system based on machine vision

Abstract

The present invention relates to the field of image data processing technology, and more specifically, to a display screen defect detection method and system based on machine vision, the method comprising: obtaining an image to be detected and a plurality of comparison images, constructing an exponential function based on the light sensitivity of the comparison image and the difference between the grayscale value of any pixel point in the image to be detected and the grayscale mean value of the pixel points in all the comparison images, determining the abnormality of the pixel point, correcting the grayscale value of the pixel point based on the correction amplitude determined by the abnormality and a preset positive feedback rule, obtaining a correction value, calculating the cosine similarity between the one-dimensional vector of the correction value of each pixel point in the image to be detected and the one-dimensional vector of the grayscale mean value of each pixel position in all the comparison images, and performing defect detection on the display screen to be detected according to the comparison result of the cosine similarity and the preset threshold value. The present invention can improve the accuracy of defect detection on the display screen.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and more specifically, to a display screen defect detection method and system based on machine vision. Background Art

[0002] The production of display screens is affected by a variety of processes and environmental factors, and defective products are inevitable. At present, defect detection mainly relies on manual visual recognition, which not only easily causes visual fatigue of inspectors, but also reduces efficiency and affects the detection effect. Machine vision technology can automatically identify defects through image processing and analysis, thereby significantly improving the efficiency and accuracy of detection.

[0003] In the related art, a method and device for detecting display defects of a display screen is disclosed in a patent application document with publication number CN114331975A, the method comprising: obtaining a sampling image, the sampling image being an image obtained by photographing a display screen displaying a sample image; obtaining each pixel in the sampling image; performing brightness removal processing on each pixel to obtain a target image; performing defect detection on the target image to obtain the defect position in the target image; wherein the brightness removal processing comprises: calculating the ambient light brightness value at the pixel point, the ambient light brightness value being the brightness value of the ambient light outside the display screen at the pixel point, and performing brightness compensation on the brightness of the pixel point according to the ambient light brightness value.

[0004] However, the above scheme only performs defect detection on the target image obtained by brightness cleaning to obtain the defect position. There is a lack of contrast data of the sampled image, so that the brightness cleaning may not fully represent the normal state, and the effect of the brightness cleaning depends on the accuracy of brightness compensation. When brightness compensation is performed on the brightness of the pixel point, it may lead to the loss of detail information when the defect area is small, which may reduce the accuracy of defect detection on the display screen to a certain extent. Summary of the invention

[0005] In order to solve the problem of low accuracy when performing defect detection on a display screen, the present invention provides a display screen defect detection method and system based on machine vision.

[0006] According to a first aspect of the present invention, a display screen defect detection method based on machine vision is provided, comprising:

[0007] Obtaining a grayscale image of the display screen to be inspected to obtain an image to be inspected, and obtaining grayscale images of an intact display screen at different times to obtain multiple comparison images;

[0008] Calculate the light sensitivity of the contrast images. Light sensitivity reflects the effect of illumination on each contrast image. Determine the difference between the gray value of any pixel in the image to be detected and the gray mean of the pixels in all contrast images. Use the sum of the light sensitivity and 1 as the power of an exponential function with the normalized value of the difference as the base, and perform power operations to obtain the degree of abnormality of any pixel.

[0009] Using the correction amplitude of any pixel point, the gray value of any pixel point is corrected according to the preset positive feedback rule to obtain the correction value. , For the The abnormality of each pixel, is the grayscale mean of the eight neighboring pixels of the pixel, is the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, is the absolute value symbol;

[0010] The cosine similarity between the one-dimensional vector of the correction value of each pixel point in the image to be detected and the one-dimensional vector of the grayscale mean of each pixel position in all the comparison images is calculated, so as to perform defect detection on the display screen to be detected based on the comparison result of the cosine similarity and the preset threshold.

[0011] The present invention can increase the difference between the grayscale value of the pixel points in the image to be detected that are in a defective area and the grayscale value of the corresponding pixel points in the comparison image by correcting the grayscale value of each pixel point in the image to be detected. Then, based on the similarity between the one-dimensional vector composed of the correction value of each pixel point in the image to be detected and the one-dimensional vector composed of the grayscale mean value of each pixel position in the comparison image, the defect detection of the display screen to be detected can be accurately performed, thereby improving the accuracy of defect detection on the display screen; and when determining the abnormality degree of each pixel point, the influence of light can be reduced, thereby further ensuring the accuracy of the determined abnormality degree.

[0012] Preferably, the abnormality degree of any pixel satisfies the following relationship:

[0013] ;

[0014] In the formula, is the first The abnormality degree of each pixel; is the grayscale mean of all pixels in the comparison images; is the first The gray value of each pixel; is the maximum grayscale value in the image to be detected; To compare the light sensitivity of the image.

[0015] In the calculation formula of the abnormality degree provided by the present invention, the range of the base of the deformed exponential function can be limited to 0-1, thereby reducing the difficulty of calculation.

[0016] Preferably, the method for obtaining photosensitivity includes:

[0017] Performing vectorization processing on each comparison image to obtain multiple comparison vectors, and obtaining the extreme difference and minimum value of the cosine similarity between each comparison vector;

[0018] The light sensitivity of the contrast image is calculated, and the light sensitivity is positively correlated with the normalized value of the extreme value and negatively correlated with the normalized value of the minimum value.

[0019] The present invention determines the light sensitivity of the contrast image according to the difference between the contrast images under different illumination conditions, thereby ensuring the accuracy of the light sensitivity.

[0020] Preferably, the light sensitivity of the contrast image satisfies the following relationship:

[0021] ;

[0022] In the formula, To compare the light sensitivity of the image; , are the minimum and maximum values ​​of the cosine similarity between the comparison vectors respectively; The natural constant An exponential function with base .

[0023] Preferably, each comparison image is vectorized to obtain a plurality of comparison vectors, including:

[0024] The grayscale values ​​of the pixels in any contrast image are spliced ​​into a one-dimensional vector in the order of rows or columns to obtain the contrast vector corresponding to any contrast image, and then the contrast vector corresponding to each contrast image is determined.

[0025] Preferably, the method for obtaining the correction value includes:

[0026] If the grayscale value of any pixel in the image to be detected is less than or equal to half of the sum of the maximum grayscale value and the minimum grayscale value, the difference between the grayscale value of any pixel and the correction amplitude is used as the correction value of the grayscale value of any pixel;

[0027] If it is greater than half of the sum of the maximum grayscale value and the minimum grayscale value, the sum of the grayscale value of any pixel and the correction amplitude is used as the correction value of the grayscale value of any pixel.

[0028] The present invention can utilize a positive feedback rule to increase the difference between the grayscale value of a pixel point in an area that is more likely to be a defect and the grayscale value of a corresponding pixel point in a comparison image.

[0029] Preferably, according to the comparison result of the cosine similarity and the preset threshold, defect detection is performed on the display screen to be detected, including:

[0030] If the cosine similarity is greater than or equal to the preset threshold, it is determined that the display screen to be tested is intact;

[0031] If the cosine similarity is less than a preset threshold, it is determined that the display screen to be inspected has defects.

[0032] The present invention can accurately perform defect detection on a display screen to be detected.

[0033] According to the second aspect of the present invention, a display screen defect detection system based on machine vision is provided. The display screen defect detection system based on machine vision includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the first aspect of the present invention.

[0034] The present invention has the following effects:

[0035] When calculating the abnormality degree of each pixel in the image to be detected, the present invention uses the sum of the light sensitivity of the comparison image and 1 to downwardly correct the difference between the grayscale value of the pixel in the image to be detected and the grayscale mean value of the pixel in all the comparison images, thereby reducing the influence of light, thereby ensuring the accuracy of the determined abnormality degree; and uses the correction amplitude and a preset positive feedback rule to correct the grayscale value of the pixel in the image to be detected, thereby increasing the difference between the pixel that is more likely to be in the defective area and the corresponding pixel in the comparison image, so that based on the similarity between the one-dimensional vector composed of the correction value of each pixel in the image to be detected and the one-dimensional vector composed of the grayscale mean value of each pixel position in the comparison image, the defective area can be accurately detected, thereby improving the accuracy of defect detection on the display screen. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0037] Figure 1 It is a schematic diagram of the steps of a display screen defect detection method based on machine vision according to an embodiment of the present invention;

[0038] Figure 2 It is a schematic diagram of a conversion result of performing vectorization processing on any comparison image to obtain a corresponding comparison vector in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0040] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0041] Reference Figure 1 , a display screen defect detection method based on machine vision includes steps S1 to S4, which are specifically as follows:

[0042] S1: Obtain a grayscale image of the display screen to be detected to obtain an image to be detected, and obtain grayscale images of an intact display screen at different times to obtain multiple comparison images.

[0043] Specifically, a collection device, such as a camera, may be placed at a certain distance in front of the intact display screen to collect complete images of the intact display screen at multiple times, and then grayscale the collected images to obtain multiple comparison images. This embodiment does not specifically limit the number of comparison images collected and the sampling time.

[0044] Furthermore, the state of the acquisition device when acquiring the comparison image can be maintained, a complete image of the display screen to be detected can be acquired right in front of the display screen to be detected, and the acquired image can be gray-scaled to obtain the image to be detected.

[0045] S2: Calculate the light sensitivity of the contrast images. The light sensitivity reflects the effect of light on each contrast image.

[0046] In an example embodiment of the present invention, light sensitivity refers to data that can reflect the effect of lighting on each contrast image. For example, when the difference between contrast images under different lighting conditions is large, it means that the effect of lighting on each contrast image is large, and the corresponding contrast image has a larger light sensitivity. Conversely, the light sensitivity of the contrast image is small.

[0047] In an exemplary embodiment of the present invention, the light sensitivity of the contrast image may be determined by the following steps:

[0048] Step 1: vectorize each comparison image to obtain multiple comparison vectors, and obtain the range and minimum value of the cosine similarity between the comparison vectors;

[0049] In an exemplary embodiment of the present invention, the determination of the contrast vector corresponding to the contrast image can be achieved by the following steps:

[0050] The grayscale values ​​of the pixels in any contrast image are spliced ​​into a one-dimensional vector in the order of rows or columns to obtain the contrast vector corresponding to any contrast image, and then the contrast vector corresponding to each contrast image is determined.

[0051] Exemplary, reference Figure 2 As shown, if any contrast image is a 4×4 image, and the grayscale values ​​of the pixels in the contrast image are 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, and 16 respectively, then the contrast vector corresponding to the contrast image is: a one-dimensional vector formed by concatenating the grayscale values ​​of the pixels in each row in row order.

[0052] Optionally, the grayscale values ​​of the pixels in each column in any comparison image may be concatenated in sequence according to the column order to form a comparison vector corresponding to the comparison image. This embodiment does not specifically limit the method for determining the comparison vector.

[0053] Further, the cosine similarity between each comparison vector can be calculated, and then the difference between the maximum and minimum values ​​of the cosine similarity between each comparison vector is determined as the range of the cosine similarity between each comparison vector. It should be noted that the cosine similarity between any two comparison vectors is the cosine value of the angle between the two comparison vectors, which can measure the similarity between the two comparison vectors. The calculation method of cosine similarity is a prior art and is not described in detail in this embodiment.

[0054] Step 2: Calculate the light sensitivity of the contrast image. The light sensitivity is positively correlated with the normalized value of the extreme value and negatively correlated with the normalized value of the minimum value.

[0055] Specifically, the light sensitivity of the contrast image satisfies the following relationship:

[0056] ;

[0057] In the formula, To compare the light sensitivity of the image; , are the minimum and maximum values ​​of the cosine similarity between the comparison vectors respectively; The natural constant An exponential function with base .

[0058] in, It is the normalized value of the minimum cosine similarity between all contrast vectors, and its value range is [0,1]. The larger the value, the greater the similarity between the contrast vectors, indicating that the difference between the contrast images collected under different lighting conditions is smaller, that is, the influence of lighting on the contrast images is smaller, and the corresponding contrast images have lower light sensitivity.

[0059] It is the normalized value of the extreme difference of the cosine similarity between all contrast vectors. The larger the value, the greater the difference in similarity between the contrast vectors, which means that the gap between the contrast images collected under different lighting conditions is greater, that is, the influence of lighting on each contrast image is greater, and the corresponding contrast image has a greater light sensitivity.

[0060] In another embodiment, the light sensitivity of the contrast image may also be calculated by other calculation formulas, such as by a summation calculation formula instead of a product calculation formula.

[0061] S3: Determine the difference between the grayscale value of any pixel in the image to be detected and the grayscale mean value of the pixels in all the comparison images, and use the sum of the light sensitivity and 1 as the power of an exponential function with the normalized value of the difference as the base, and perform a power operation to obtain the abnormality degree of any pixel.

[0062] It should be noted that, since the image of the display screen contains a large number of pixels, when the display screen to be detected contains only a small range of defects, only a small number of pixels in the image to be detected may have certain differences with the pixels at the corresponding positions in the comparison image. Even if there are certain defects in the image to be detected, it still has a high similarity with the comparison image, and under different lighting conditions, the grayscale value of the pixel in the image to be detected may also be different from the grayscale value of the comparison image. Therefore, this embodiment uses the sum of the light sensitivity of the comparison image and 1 to correct the difference between the grayscale value of the pixel in the image to be detected and the grayscale mean of the pixel in all the comparison images, which can reduce the influence of light, so as to accurately measure the possibility that the position of each pixel in the image to be detected is a defect position.

[0063] In an example embodiment of the present invention, the degree of abnormality refers to data that measures the possibility that the location of a pixel point in the image to be detected is a defect location. For example, when the degree of abnormality of any pixel point in the image to be detected is large, it means that the possibility that the location of the pixel point is a defect location is large. Conversely, the possibility that the location of the pixel point is a defect location is small.

[0064] Specifically, the abnormality degree of any pixel in the image to be detected satisfies the following relationship:

[0065] ;

[0066] In the formula, is the first The abnormality degree of each pixel; is the grayscale mean of the pixels in all the contrast images. Specifically, the grayscale mean may be obtained by calculating the average grayscale value of all the pixels in all the contrast images. is the first The gray value of each pixel; is the maximum grayscale value in the image to be detected. , where the maximum grayscale value is the maximum grayscale value of all pixels in the image to be detected; To compare the light sensitivity of the image.

[0067] in, is the first The normalized value of the gray value of the pixel and the gray value mean of all the pixels in the comparison image. The larger the value, the higher the gray value of the pixel in the image to be detected. The greater the possibility that the location of the pixel is a defective location.

[0068] In another embodiment, the normalized value of the difference may also be determined by other means, such as by The normalization function is determined. This embodiment does not specifically limit the method for determining the normalized value of the difference.

[0069] S4: using the correction amplitude of any pixel point, correcting the grayscale value of any pixel point according to a preset positive feedback rule to obtain a correction value.

[0070] In an exemplary embodiment of the present invention, the correction amplitude refers to the degree of correction when the grayscale value of a pixel in the image to be detected is corrected; specifically, the correction amplitude of any pixel in the image to be detected satisfies the following relationship:

[0071] ;

[0072] In the formula, is the first The correction amplitude of each pixel; For the The abnormality degree of each pixel; is the grayscale mean of the eight neighboring pixels of the pixel; It is the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected. Half of this value reflects the average grayscale value of the pixels in the image to be detected. The maximum grayscale value is the maximum grayscale value of all pixels in the image to be detected, and the minimum grayscale value is the minimum grayscale value of all pixels in the image to be detected. is the absolute value symbol.

[0073] in, It reflects the first The difference between the grayscale mean of the surrounding pixels of a pixel and the average grayscale value of the pixels in the image to be detected, the larger the value is and the greater the abnormality of the pixel, the more likely the location of the pixel is to be a defective location. Therefore, the difference between the pixel and the corresponding pixel in the comparison image can be increased to ensure that the location of the pixel can be identified as a defective location.

[0074] Among them, the preset positive feedback rule is: after determining the correction amplitude of any pixel point in the image to be detected, if the grayscale value of the pixel point is large, the grayscale value of the pixel point is increased; if the grayscale value of the pixel point is small, the grayscale value of the pixel point is reduced, thereby determining the correction value of each pixel point in the image to be detected based on the positive feedback mechanism.

[0075] In an exemplary embodiment of the present invention, the correction value can be determined by the following steps:

[0076] (1) If the grayscale value of any pixel in the image to be detected is less than or equal to half of the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, the difference between the grayscale value of any pixel and the correction amplitude is used as the correction value of the grayscale value of any pixel;

[0077] Optionally, when the grayscale value of any pixel in the image to be detected is less than or equal to half of the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, it means that the grayscale value of any pixel is closer to the minimum grayscale value in the image to be detected. Therefore, the grayscale value of the pixel can be corrected downward, and when the correction amplitude of the pixel is larger, the difference between the pixel and the corresponding pixel in the comparison image is larger.

[0078] Specifically, the grayscale value of the image to be detected, the pixel points close to the minimum grayscale value of the image to be detected are corrected downward to satisfy the following relationship:

[0079] ;

[0080] In the formula, is the first When a pixel is close to the minimum gray value in the image to be detected, the gray value correction value of the pixel; is the gray value of the pixel; is the abnormality degree of the pixel; is the grayscale mean of the eight neighboring pixels of the pixel; is the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected; is the absolute value symbol.

[0081] (2) If it is greater than half of the sum of the maximum grayscale value and the minimum grayscale value, the sum of the grayscale value of any pixel and the correction amplitude is used as the correction value of the grayscale value of any pixel.

[0082] Optionally, when the grayscale value of any pixel in the image to be detected is greater than half of the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, it means that the grayscale value of any pixel is closer to the maximum grayscale value in the image to be detected. Therefore, the grayscale value of the pixel can be corrected upward, and when the correction amplitude of the pixel is large, the difference between the pixel and the corresponding pixel in the comparison image is larger, so that the defective area in the image to be detected can be identified.

[0083] Specifically, the grayscale values ​​of the image to be detected and the pixel points close to the maximum grayscale value of the image to be detected are corrected upward to satisfy the following relationship:

[0084] ;

[0085] In the formula, is the first When a pixel is close to the maximum gray value in the image to be detected, the gray value correction value of the pixel; is the gray value of the pixel; is the abnormality degree of the pixel; is the grayscale mean of the eight neighboring pixels of the pixel; is the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected; is the absolute value symbol.

[0086] S5: Calculate the cosine similarity between the one-dimensional vector of the correction value of each pixel point in the image to be detected and the one-dimensional vector of the grayscale mean value of each pixel position in all the comparison images, so as to perform defect detection on the display screen to be detected according to the comparison result of the cosine similarity and the preset threshold.

[0087] In an exemplary embodiment of the present invention, a one-dimensional vector of correction values ​​of each pixel in the image to be detected and a one-dimensional vector of grayscale mean values ​​of each pixel position in all the comparison images are determined in accordance with Figure 2 The determination method shown is exactly the same and will not be described in detail in this embodiment.

[0088] It should be noted that the grayscale mean value of any pixel position in all the contrast images is: the average value of the grayscale values ​​of all pixels at the pixel position in all the contrast images.

[0089] In an exemplary embodiment of the present invention, defect detection of a display screen to be inspected can be achieved by the following steps:

[0090] If the cosine similarity is greater than or equal to the preset threshold, the display screen to be tested is determined to be intact; if the cosine similarity is less than the preset threshold, the display screen to be tested is determined to be defective.

[0091] Optionally, the preset threshold can be set to 0.98. When the cosine similarity between the one-dimensional vector of the correction value of each pixel point in the image to be detected and the one-dimensional vector of the grayscale mean of each pixel position in all the comparison images is greater than or equal to 0.98, the display screen to be detected is judged to be a complete display screen; if the cosine similarity is less than 0.98, it is judged that the display screen to be detected has defects, thereby realizing defect detection of the display screen to be detected.

[0092] The present invention also provides a display screen defect detection system based on machine vision. The system includes a memory and a processor, and a computer program is stored in the memory. The computer program integrates the functions of the display screen defect detection method based on machine vision. When the computer program is executed, the accuracy of defect detection on the display screen can be improved through the display screen defect detection method based on machine vision.

[0093] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0094] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A display screen defect detection method based on machine vision, characterized in that: include: Obtaining a grayscale image of the display screen to be inspected to obtain an image to be inspected, and obtaining grayscale images of an intact display screen at different times to obtain multiple comparison images; Calculate the light sensitivity of the contrast image, where the light sensitivity reflects the effect of illumination on each contrast image, determine the difference between the gray value of any pixel in the image to be detected and the gray mean of the pixels in all contrast images, and use the sum of the light sensitivity and 1 as the power of an exponential function with the normalized value of the difference as the base, and perform a power operation to obtain the abnormality degree of any pixel; Using the correction amplitude of any pixel point, the gray value of any pixel point is corrected according to the preset positive feedback rule to obtain a correction value, wherein the correction amplitude , For the The abnormality of each pixel, is the grayscale mean of the eight neighboring pixels of the pixel, is the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, is the absolute value symbol; The cosine similarity between the one-dimensional vector of the correction value of each pixel point in the image to be detected and the one-dimensional vector of the grayscale mean of each pixel position in all the comparison images is calculated, so as to perform defect detection on the display screen to be detected according to the comparison result of the cosine similarity and the preset threshold.

2. The display screen defect detection method based on machine vision according to claim 1, characterized in that: The abnormality degree of any pixel point satisfies the following relationship: ； In the formula, is the first The abnormality degree of each pixel; is the grayscale mean of all pixels in the comparison images; is the first The gray value of each pixel; is the maximum grayscale value in the image to be detected; To compare the light sensitivity of the image.

3. The display screen defect detection method based on machine vision according to claim 2, characterized in that: The method for obtaining the light sensitivity comprises: Performing vectorization processing on each of the comparison images to obtain a plurality of comparison vectors, and obtaining the range and minimum value of the cosine similarity between the comparison vectors; The light sensitivity of the contrast image is calculated, and the light sensitivity is positively correlated with the normalized value of the extreme difference and negatively correlated with the normalized value of the minimum value.

4. The display screen defect detection method based on machine vision according to claim 3, characterized in that: The light sensitivity of the contrast image satisfies the following relationship: ； In the formula, To compare the light sensitivity of the image; , are the minimum and maximum values ​​of the cosine similarity between the comparison vectors respectively; The natural constant An exponential function with base .

5. The display screen defect detection method based on machine vision according to claim 3, characterized in that: The vectorization process is performed on each of the comparison images to obtain a plurality of comparison vectors, including: The grayscale values ​​of the pixels in any contrast image are spliced ​​into a one-dimensional vector in the order of rows or columns to obtain the contrast vector corresponding to any contrast image, and then the contrast vector corresponding to each contrast image is determined.

6. The display screen defect detection method based on machine vision according to claim 1, characterized in that: The method for obtaining the correction value includes: If the grayscale value of any pixel in the image to be detected is less than or equal to half of the sum of the maximum grayscale value and the minimum grayscale value in the image to be detected, the difference between the grayscale value of any pixel and the correction amplitude is used as the correction value of the grayscale value of any pixel; If it is greater than half of the sum of the grayscale maximum value and the grayscale minimum value, the sum of the grayscale value of any pixel point and the correction amplitude is used as the correction value of the grayscale value of any pixel point.

7. The display screen defect detection method based on machine vision according to claim 1, characterized in that: The step of performing defect detection on the display screen to be detected according to the comparison result of the cosine similarity and the preset threshold value includes: If the cosine similarity is greater than or equal to the preset threshold, it is determined that the display screen to be detected is intact; If the cosine similarity is less than the preset threshold, it is determined that the display screen to be detected has defects.

8. A display screen defect detection system based on machine vision, characterized in that: The machine vision-based display screen defect detection system includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the steps of the method as described in any one of claims 1-7.

Images