Method for detecting mura defects on display screen, electronic device and computer readable storage medium

Abstract

The application relates to the technical field of image processing, and discloses a display screen Mura defect detection method, electronic equipment and a computer readable storage medium. The detection method comprises the following steps: calculating an integral image of a to-be-detected area on a display screen; sliding a detection window in the integral image, and calculating the sum of the gray values of the corresponding area in the to-be-detected area of each detection window in the sliding process as the area gray value of each detection window; determining the absolute value of the difference between the area gray values of two side detection windows located on both sides of each middle detection window and the middle detection window, generating a defect response image; performing gray threshold segmentation on the defect response image to obtain the position of the Mura defect in the defect response image; and determining the Mura defect in the to-be-detected area based on the corresponding relationship between the defect response image and the to-be-detected area. The detection method of the application can improve the efficiency and accuracy of display screen Mura defect detection.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the technical field of image processing. More particularly, the present application relates to a method for detecting Mura defects on a display screen, an electronic device and a computer readable storage medium. BACKGROUND

[0002] With the wide application of electronic devices such as smart phones, tablets, desktop displays and vehicle-mounted display screens, the display screen as a core human-computer interaction component, its display quality directly determines the user experience. In the production and manufacturing process of the display screen, due to the influence of factors such as panel material uniformity and process precision, various display defects are prone to occur, among which the uneven brightness and darkness (Mura) defect is particularly common. The horizontal or vertical distribution of strip-shaped bright and dark Mura defects will cause the screen to form a significant brightness difference between the local strip-shaped area and the surrounding normal gray scale area, which seriously damages the consistency of the display picture, and becomes a key control object for quality detection before the display screen is shipped. At present, display screen production enterprises have included the detection of strip-shaped bright and dark Mura defects into the core quality control process, and the demand for detection technology with high efficiency and accuracy is increasingly urgent.

[0003] Therefore, it is urgent to provide a detection scheme for Mura defects on a display screen in order to improve the detection efficiency and accuracy of strip-shaped bright and dark Mura defects, so as to meet the efficient and reliable quality detection needs in the production process of the display screen. SUMMARY

[0004] In order to at least solve one or more technical problems mentioned above, the present application provides a method for detecting Mura defects on a display screen, an electronic device and a computer readable storage medium in multiple aspects.

[0005] In a first aspect, the present application provides a method for detecting Mura defects on a display screen, comprising: calculating an integral image of a to-be-detected area on the display screen, wherein the integral value of any point in the integral image is the sum of the gray scale values of all points in the rectangular area surrounded by the top left corner of the to-be-detected area and the point; according to a set step size, sliding a detection window in the integral image, and calculating the sum of the gray scale values of the corresponding area in the to-be-detected area of each detection window in the sliding process as the area gray scale value of each detection window; determining the absolute value of the difference between the area gray scale values of two side detection windows located on both sides of each intermediate detection window and the intermediate detection window, and generating a defect response image, wherein the position direction of the two side detection windows relative to the intermediate detection window is perpendicular to the extension direction of the to-be-detected Mura defect; performing gray threshold segmentation on the defect response image to obtain the position of the Mura defect in the defect response image; determining the Mura defect in the to-be-detected area based on the correspondence between the defect response image and the to-be-detected area.

[0006] In some embodiments, the area gray value of each detection window is determined based on the following formula: Sum(D) = ii(4) + ii(l) - ii(2) - ii(3); wherein Sum(D) represents the sum of the gray values of the corresponding areas D of the detection window in the area to be detected, ii(4) represents the integral value of the point at the lower right corner of area D, ii(l) represents the integral value of the point at the upper left corner of area D, ii(2) represents the integral value of the point at the upper right corner of area D, and ii(3) represents the integral value of the point at the lower left corner of area D.

[0007] In other embodiments, determining the absolute value of the difference between the area gray values of the two side detection windows located on both sides of each intermediate detection window and the intermediate detection window comprises: determining the positions of the two side detection windows based on the step length, wherein the non-overlapping areas of the two side detection windows and the intermediate detection window can be spliced into the size of one detection window; and calculating the difference and the absolute value of the sum of the area gray values of the two side detection windows and twice the area gray value of the intermediate detection window.

[0008] In yet other embodiments, the detection method further comprises: determining the type of the Mura defect based on the positive or negative of the difference value, wherein, in response to the difference value being positive, determining that the Mura defect is a dark defect; and in response to the difference value being negative, determining that the Mura defect is a bright defect.

[0009] In some embodiments, the position direction of the two side detection windows relative to the intermediate detection window is perpendicular to the extension direction of the Mura defect to be detected, comprising: when the extension direction of the Mura defect to be detected is a vertical direction, the position direction of the two side detection windows relative to the intermediate detection window is a horizontal direction; and when the extension direction of the Mura defect to be detected is a horizontal direction, the position direction of the two side detection windows relative to the intermediate detection window is a vertical direction.

[0010] In other embodiments, the difference value is calculated by the following formula: diff = grey(i-b) + grey(i+b) - 2 x grey(i); wherein diff represents the difference value, grey(i) represents the area gray value of the intermediate detection window i, grey(i-b) and grey(i+b) respectively represent the area gray values of the b-th detection window located on both sides of the intermediate detection window i, wherein b = 1 / (2 x S), and S represents the ratio of the step length to the window width of the detection window.

[0011] In yet some embodiments, the window width of the detection window is greater than the defect width of the Mura defect to be detected; the detection window is set as a square; and the step length is set as 1 / 4 of the window width.

[0012] In some embodiments, the gray scale threshold segmentation is performed on the defect response map, and the position of the Mura defect in the defect response map is obtained by setting the gray scale value of the position with a gray scale value greater than a threshold value in the defect response map to 255 and setting the gray scale value of the position with a gray scale value less than the threshold value in the defect response map to 0. The detection method further comprises length screening of the Mura defect in the defect response map to obtain the Mura defect of interest.

[0013] In a second aspect, the present application provides an electronic device, comprising a processor configured to execute program instructions, and a memory configured to store the program instructions, when the program instructions are loaded and executed by the processor, the processor executes the detection method according to any one of the first aspect of the present application.

[0014] In a third aspect, the present application provides a computer readable storage medium, wherein the program instructions are stored, when the program instructions are loaded and executed by the processor, the processor executes the detection method according to any one of the first aspect of the present application.

[0015] By the detection method of Mura defect on a display screen as provided above, the embodiments of the present application calculate the integral graph of the to-be-detected area of the display screen (the integral value is the sum of the gray scale values of all points in the rectangular area surrounded by the top-left corner of the to-be-detected area and the point), slide the detection window in the integral graph by a set step length and calculate the area gray scale value of each detection window corresponding to the to-be-detected area, determine the absolute value of the difference between the area gray scale values of the two side detection windows and the middle detection window perpendicular to the extension direction of the to-be-detected Mura defect to generate a defect response map, perform gray scale threshold segmentation on the defect response map, and determine the position of the Mura defect based on the correspondence between the response map and the to-be-detected area, which can avoid the tedious operation of traditional pixel-by-pixel traversal and accumulation calculation of the area gray scale value, and improve the calculation efficiency in the Mura defect detection process. At the same time, by adapting the setting of the two side detection windows corresponding to the extension direction of the to-be-detected Mura defect, the detection adaptability to Mura defects with different extension directions is enhanced, and by means of the defect response map generated by the absolute value of the difference and the gray scale threshold segmentation, the defect area is effectively highlighted, the accuracy of Mura defect positioning is improved, and the efficiency and accuracy requirements of display screen Mura defect detection are met. BRIEF DESCRIPTION OF DRAWINGS

[0016] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description read in conjunction with the accompanying drawings, in which several embodiments of the present application are shown. In the drawings, the same or corresponding parts are denoted by the same or corresponding reference numerals, and:

[0017] Figure 1 An exemplary flowchart of a method for detecting Mura defects on a display screen according to some embodiments of the present application is shown;

[0018] Figure 2 An exemplary diagram of obtaining a region to be detected according to some embodiments of the present application is shown;

[0019] Figure 3 An exemplary diagram of sliding a window in an integral diagram according to an embodiment of the present application is shown;

[0020] Figure 4 An exemplary diagram of calculating a gray value of a region according to an embodiment of the present application is shown;

[0021] Figure 5 A gray value diagram of a region according to some embodiments of the present application is shown;

[0022] Figure 6 An exemplary diagram of a relationship between a middle detection window and two side detection windows according to an embodiment of the present application is shown;

[0023] Figure 7 A defect response diagram according to some embodiments of the present application is shown;

[0024] Figure 8 An exemplary diagram of determining a Mura defect position based on a defect response diagram according to some embodiments of the present application is shown;

[0025] Figure 9 An exemplary block diagram of a system for detecting Mura defects on a display screen according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be apparently and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person skilled in the art without any creative work fall within the scope of protection of the present application.

[0027] It should be understood that the terms “include” and “contain” used in the specification and claims of the present application indicate the existence of the described features, integers, steps, operations, elements and / or components, but do not exclude the existence or addition of one or more other features, integers, steps, operations, elements, components and / or sets thereof.

[0028] It should also be understood that the terms used herein are for the purpose of describing particular embodiments and are not intended to limit the application. As used in this specification and claims, the singular forms "a," "an" and "the" include plural referents unless the context clearly dictates otherwise. It should also be further understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items, as well as the lack of combinations when interpreted in the alternative.

[0029] As used in this specification and claims, the term "if' can be construed to mean "when" or "once," or "in response to a determination" or "in response to the occurrence of an event" that is detected. Similarly, the phrase "if it is determined" or "if [a described condition or event] is detected" can be construed to mean "once it is determined" or "in response to a determination" or "once [the described condition or event] is detected" or "in response to the detection of [the described condition or event]," depending on the context.

[0030] Mura defect is a common uneven brightness defect in the production and manufacturing process of display screens, especially gray scale screens. The core performance of the defect is that there is a significant brightness difference between the local area of the screen and the normal gray scale area around it, and the defect is typically in the form of horizontal or vertical strips, which can destroy the consistency of the display picture and directly affect the user experience. Mura defect has three core characteristics: first, the distribution direction is fixed, usually horizontal or vertical; second, the form is strip-shaped with a certain width, which is different from the line defect without width; third, there is an identifiable brightness difference with the adjacent normal area. According to the brightness attribute, it can be divided into dark Mura (the brightness of the defect area is lower than that of the normal area) and bright Mura (the brightness of the defect area is higher than that of the normal area), and according to the distribution direction, it can be divided into horizontal strip Mura and vertical strip Mura.

[0031] The present application provides a new detection method for the above characteristics of Mura defect, which can not only improve the detection efficiency and accuracy of Mura defect, but also adapt to the detection of Mura defects with various attributes (dark Mura, bright Mura) and various distribution directions (horizontal direction, vertical direction). The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0032] Figure 1 An exemplary flowchart of a method for detecting Mura defects on a display screen is shown. As shown in FIG. 1, the method includes the following steps: Figure 1As shown in FIG. 1, the detection method 100 can include: in step S101, calculating an integral image of a to-be-detected region on the display screen, wherein the integral value of any point in the integral image is the sum of the gray values of all points in the rectangular region formed by the top-left corner of the to-be-detected region and the point; in step S102, sliding the detection window in the integral image according to a set step, and calculating the sum of the gray values of the corresponding region of each detection window in the to-be-detected region during the sliding process as the region gray value of each detection window; in step S103, determining the absolute value of the difference between the region gray values of the two side detection windows located on the two sides of each middle detection window and the middle detection window, and generating a defect response image, wherein the position direction of the two side detection windows relative to the middle detection window is perpendicular to the extension direction of the to-be-detected Mura defect; in step S104, performing gray threshold segmentation on the defect response image to obtain the position of the Mura defect in the defect response image; and in step S105, determining the Mura defect in the to-be-detected region based on the correspondence between the defect response image and the to-be-detected region.

[0033] In some embodiments, the to-be-detected region can be the entire screen region of the display screen. In other embodiments, the to-be-detected region can be cropped from a product image containing the display screen. For example, Figure 2 FIG. 2 shows a schematic diagram of obtaining a to-be-detected region according to some embodiments of the present application. As shown in FIG. 2, a product image 210 is obtained, and a screen region of the product image 210 is cropped as a to-be-detected region 220. Figure 2 As shown in FIG. 2, the four corner points 211 (white circles in the figure) of the screen region in the product image 210 can be located using a corner point detection technique (such as the Harris corner point detection technique), and then the screen region is cropped from the background of the product image 210 using affine transformation to obtain the to-be-detected region 220. In yet other embodiments, the to-be-detected region can be a partial screen region on the display screen, for example, a region of interest is cropped from the screen region of the product image 210 as the to-be-detected region. Figure 2 As shown in FIG. 2, the four corner points 211 (white circles in the figure) of the screen region in the product image 210 can be located using a corner point detection technique (such as the Harris corner point detection technique), and then the screen region is cropped from the background of the product image 210 using affine transformation to obtain the to-be-detected region 220. In yet other embodiments, the to-be-detected region can be a partial screen region on the display screen, for example, a region of interest is cropped from the screen region of the product image 210 as the to-be-detected region.

[0034] The integral image is a matrix representation method for quickly calculating the sum of the pixel gray values of a local region of an image, and its core attribute is that for any point (x, y) in the integral image, the integral value of the point (denoted as ii(x, y)) is equal to the sum of the gray values of all pixels in the rectangular region formed by the top-left corner of the to-be-detected region and the point (x, y). In other words, the rectangular region here refers to the region of all pixels above and to the left of the point (x, y), including the point (x, y).

[0035] After the integral image is calculated in step S101, a detection window can be set to slide in the integral image, and the sum of the gray scale values of the corresponding regions in the detection region at each stop of the sliding detection window, i.e., the region gray scale value of each detection window, can be calculated. For ease of understanding, exemplary descriptions will be given below in combination with Figure 3 and Figure 4 .

[0036] Figure 3 A schematic diagram of the sliding window in the integral image according to an embodiment of the present application is shown in FIG. 3. As shown in FIG. 3, the detection window 320 can slide along the X-axis direction (horizontal direction) and the Y-axis direction (vertical direction) from the upper left corner of the integral image 310 according to a set step (labeled as step in the figure). The set step should not be greater than the window width winSz of the detection window 320. In some embodiments, the step can be set to 1 / 16~1 of the window width winSz. In some preferred embodiments, the step can be set to 1 / 4 of the window width winSz. Figure 3 Specifically, in some embodiments, the detection window 320 (with a window width of winSz) can first move along the X-axis direction from left to right (e.g., from the red detection window at the upper left corner to the green detection window, to the blue detection window, and to the yellow detection window in the figure) from the upper left corner of the integral image 310 with the set step step as the moving interval, stop at each step, and then move to the next position after the sum of the gray scale values of the corresponding regions in the detection region at the position is calculated. When the detection window 320 moves to the upper right corner in the first row of the integral image, it can move to the next position downward along the Y-axis direction from the upper left corner of the integral image 310 with the step step, and then continue to move rightward along the X-axis direction. This cycle is repeated until all the regions in the integral image that can have the strip-shaped bright-dark Mura defect are traversed. This sliding manner is more suitable for detecting the Mura defect distributed along the vertical direction when steps S102 and S103 are executed in parallel.

[0037] In other embodiments, the detection window 320 can first move along the Y-axis direction from the upper left corner of the integral image 310 step by step from top to bottom, and then move to the next position rightward along the X-axis direction from the upper left corner of the integral image with the step step when it moves to the lower left corner of the integral image 310, and then continue to move downward along the Y-axis direction. This cycle is repeated. This sliding manner is more suitable for detecting the Mura defect distributed along the horizontal direction when steps S102 and S103 are executed in parallel.

[0038]

[0039] ​The sliding mode ensures the systematicness and comprehensiveness of detection by a fixed step length, ensures that each potential defect area can be covered, and balances the detection efficiency and accuracy by reasonable setting of the step length, thereby providing a data collection basis for subsequent Mura defect recognition based on window pixel sum.

[0040] In some embodiments, the window width winSz of the detection window 320 is greater than the defect width of the Mura defect to be detected. For example, the defect width of the Mura defect to be detected can be estimated according to an empirical value, and the window width winSz of the detection window 320 can be set to 1.1 times to 2 times the estimated defect width. In some preferred embodiments, the window width winSz of the detection window 320 can be set to 1.5 times the estimated defect width. According to the setting of 1.5 times, it can be ensured that one detection window can cover the defect width of the Mura defect to be detected, and the surrounding normal area can not be excessively included when the detection window covers the position of the Mura defect, thereby facilitating the improvement of the accuracy of defect edge detection.

[0041] In some embodiments, the detection window can be set to a rectangle or a square. In some preferred embodiments, the detection window can be set to a square to facilitate the movement of the detection window and subsequent calculation.

[0042] Using the integral image calculated in step S101, the regional gray value of the local area in the area to be detected corresponding to each stay of the detection window can be quickly calculated based on the integral values of the positions of the four corner points of the detection window. For ease of understanding, the following will be described by way of example. Figure 4

[0043] Figure 4 The principle diagram for calculating the regional gray value of the embodiment of the present application is shown. As shown in Figure 4 , the upper left corner of the area to be detected is taken as the origin O, the right is the X-axis direction, and the downward is the Y-axis direction. In some embodiments, the regional gray value of each detection window is determined based on the following formula:

[0044] Sum(D) = ii(4) + ii(1) - ii(2) - ii(3) (Formula One);

[0045] Wherein, Sum(D) represents the sum of the gray values of the detection window in the corresponding area D in the area to be detected, ii(4) represents the integral value of the pixel point 4 at the lower right corner of the area D, ii(1) represents the integral value of the pixel point 1 at the upper left corner of the area D, ii(2) represents the integral value of the pixel point 2 at the upper right corner of the area D, and ii(3) represents the integral value of the pixel point 3 at the lower left corner of the area D. Figure 4 Figure 4 Figure 4 ​​​the integral value of pixel point 3 in the integral image. Figure 5 the integral value of pixel point 3 in the integral image.

[0046] Since the integral image and the to-be-detected region are of the same size, the positions of the pixels in the two are also corresponding, and thus the sliding of the detection window in the integral image can be equivalent to the sliding in the to-be-detected region. Based on the set step and the window width of the detection window, the coordinates of the four corner points of each detection window can be determined, and thus the region gray value of each detection window can be directly calculated based on the above formula one. Here, each detection window refers to the position of the detection window when it stays once in the sliding process.

[0047] In the traditional calculation manner of the pixel sum of a sub-region, all the pixels in the sub-region need to be traversed and added one by one, and the calculation amount will significantly increase with the increase of the area of the sub-region (for example, 10000 additions are needed for a 100x100 region). In comparison, in the calculation manner of the region gray value based on the integral image, no matter how large the area of the sub-region is, only "one addition + two subtractions" are needed to perform the above formula one, and thus the pixel sum (i.e., the sum of the gray values of the sub-region) can be obtained instantaneously. The larger the area of the to-be-detected region is, the more sub-regions need to be calculated (such as the sliding detection window detection in the detection method of the present application), and the more significant the efficiency improvement brought by the integral image is.

[0048] In some implementation scenarios, to simplify the subsequent calculation, the region gray image can be generated based on the region gray value of each detection window calculated in step S102. For example, Figure 3 The region gray image of some embodiments of the present application is shown. By performing the sliding window calculation on the integral image as shown in Figure 5 , the region gray value of each detection window is obtained, and each region gray value is sequentially filled into the image in which the initial pixel gray value is 0, and thus the region gray image as shown in Figure 5 is obtained. As shown in Figure 3 , the gray value of each pixel point in the region gray image corresponds to the region gray value of a corresponding detection window.

[0049] Specifically, it is assumed that Figure 3The integral image has integralImgCols columns and integralImgRows rows. The number of times the detection window moves in the horizontal direction and the vertical direction can be calculated based on the window width winSz and the step size step. For example, if the step size step is set to be 1 / 4 of the window width winSz, then the number of times the detection window moves in the horizontal direction is (integralImgCols / step-3), i.e., the integer obtained by dividing integralImgCols by step minus 3; similarly, the number of times the detection window moves in the vertical direction is (integralImgRows / step-3), i.e., the integer obtained by dividing integralImgRows by step minus 3. Thus, based on the above, the number of times the detection window moves in the horizontal direction is equal to saCol, and the number of times the detection window moves in the vertical direction is equal to saRow. Figure 5 The generated region gray image has saRow rows and saCol columns, where saRow=integralImgRows / step-3 and saCol=integralImgCols / step-3. As can be seen from the number of times the detection window moves in the horizontal direction and the vertical direction, saRow is equal to the number of times the detection window moves in the vertical direction, and saCol is equal to the number of times the detection window moves in the horizontal direction. Then, the region gray values calculated for each movement of the detection window are sequentially filled in from the top left corner of the image, where the initial pixel gray values are all 0, to obtain Figure 6 The region gray image is shown above.

[0050] After obtaining the region gray values of each detection window, the detection method 100 can perform step S103. Of course, the detection method 100 can also perform step S103 in parallel after calculating the region gray values of each intermediate detection window and the region gray values of the two side detection windows during the execution of step S102, without waiting for the region gray values of all detection windows in the detection region to be calculated before performing step S103, which is conducive to further improving the detection speed.

[0051] In step S103, the middle detection window is a window used for core calculation in the sliding detection process, and is located at a non-edge position of the integral image / detection region. The two-side detection windows are detection windows located on both sides of the middle detection window. In implementation, the position direction of the two-side detection windows relative to the middle detection window is determined according to the extension direction (i.e., distribution direction) of the Mura defect to be detected. In some embodiments, the position direction of the two-side detection windows relative to the middle detection window is perpendicular to the extension direction of the Mura defect to be detected, including: when the extension direction of the Mura defect to be detected is a vertical direction, the position direction of the two-side detection windows relative to the middle detection window is a horizontal direction; and when the extension direction of the Mura defect to be detected is a horizontal direction, the position direction of the two-side detection windows relative to the middle detection window is a vertical direction.

[0052] That is, if the Mura defect to be detected extends vertically (e.g., in a vertical bar shape), the two-side detection windows need to be distributed in the horizontal direction of the middle detection window (i.e., on the left and right sides); and if the Mura defect to be detected extends horizontally (e.g., in a horizontal bar shape), the two-side detection windows need to be distributed in the vertical direction of the middle detection window (i.e., on the top and bottom sides).

[0053] In the process of sliding the detection windows along the set step, for each middle detection window, the respective area gray values of the middle detection window and the two-side detection windows are quickly calculated by using the integral image, and then the absolute values of the gray value differences between the two-side detection windows and the middle detection window are obtained, and these absolute values are taken as the response values of the positions of the corresponding middle detection windows. After traversing all the detection window positions, the image composed of these response values is the defect response image. The area with a greater response value in the defect response image indicates that the gray value difference between the middle detection window and the two-side detection windows is more significant, and the possibility of the presence of the Mura defect is higher.

[0054] In some embodiments, the two-side detection windows can be adjacent to the middle detection window. In other embodiments, the two-side detection windows can have an overlapping area with the middle detection window, and the non-overlapping areas of the two-side detection windows and the middle detection window can be spliced into the size of one detection window.

[0055] The overlapping area retains the continuity information of the gray correlation between the detection windows, making the gray difference calculation more consistent with the actual scene; and the non-overlapping area between the two side detection windows and the middle detection window can be spliced into a detection window size design, which not only fully covers the area around the middle detection window where Mura defects may exist, but also, with the aid of the "overlapping + non-overlapping" structure, accurately captures the gray change difference between the middle detection window and the two side detection windows under the premise of using integral graph for fast calculation, thereby improving the accuracy of Mura defect detection and the complete identification ability of defect features while ensuring the detection efficiency.

[0056] In order to accurately obtain the response value, the above difference value can be the difference between the sum of the area gray values of the two side detection windows and twice the area gray value of the middle detection window. According to such a setting, not only can the interference of the overlapping area be offset, the influence of the non-target area on the detection result can be excluded, and it can be ensured that the comparison object conforms to the technical idea of "two normal areas and a middle area", but also the gray difference of weak defects can be amplified, and the twice calculation can convert the gray difference of weak dark / bright defects into a recognizable difference value, avoiding missing weak defects.

[0057] In some other embodiments, the two side detection windows are symmetrically distributed relative to the middle detection window. By letting the two side detection windows be symmetrically distributed relative to the middle detection window, the gray comparison between the middle detection window and the two side detection windows can be more balanced, reducing the calculation error caused by position deviation.

[0058] In yet some other embodiments, the detection method 300 can further include determining the type of Mura defect based on the positive and negative of the difference value, wherein, in response to the difference value being positive, determining that the Mura defect is a dark defect; and in response to the difference value being negative, determining that the Mura defect is a bright defect.

[0059] Herein, the dark defect is a Mura defect whose corresponding area gray of the middle detection window is lower than that of the two side normal areas. The bright defect is a Mura defect whose middle area gray is higher than that of the two side normal areas. When implemented, if the difference value of the area gray values of the aforementioned two side detection windows and the middle detection window is positive, it means that the sum of the area gray values of the two side detection windows is greater than twice the area gray value of the middle detection window, reflecting that the middle area gray is lower, and therefore a dark defect is determined; if the difference value is negative, it means that the sum of the area gray values of the two side detection windows is less than twice the area gray value of the middle detection window, i.e. the middle area gray is higher, and therefore a bright defect is determined, so that the type of Mura defect can be quickly distinguished.

[0060] In other implementations, determining the absolute value of the difference between the gray values ​​of the regions of the two detection windows located on either side of each intermediate detection window and the intermediate detection window includes: determining the positions of the two detection windows based on the step size, wherein the non-overlapping regions of the two detection windows and the intermediate detection window can be stitched together to form the size of a single detection window; and calculating the difference and absolute value of the sum of the gray values ​​of the regions of the two detection windows and twice the gray value of the region of the intermediate detection window.

[0061] To facilitate a further understanding of the calculation of the above difference, the following will combine... Figure 6 An exemplary description is provided. Figure 6 A schematic diagram showing the relationship between the middle detection window and the two side detection windows in an embodiment of this application is shown.

[0062] like Figure 6 As shown, the positions of the two detection windows (shown by dashed boxes) can be determined based on the step size. The two detection windows (322-1, 322-2) have overlapping areas (regions 02 and 03) and non-overlapping areas (regions 01 and 04) with the middle detection window 321. Regions 01 and 04 can be combined to form a single detection window. Specifically, the middle detection window 321 covers regions 02 and 03, the left detection window 322-1 covers regions 01 and 02, and the right detection window 322-2 covers regions 03 and 04. The combined region 01 and 04 is equivalent in size to a single detection window. Based on this, the sum of the grayscale values ​​of the two detection windows and twice the grayscale value of the middle detection window are used to calculate the difference and absolute value. This cancels out the grayscale interference from the overlapping areas (regions 02 and 03), accurately capturing the grayscale difference between the middle area and the two non-overlapping normal areas, thus achieving accurate detection of strip-shaped Mura defects.

[0063] In one specific embodiment, the aforementioned difference can be calculated using the following formula:

[0064] diff=grey(ib)+grey(i+b)-2×grey(i) (formula 2);

[0065] Where, diff represents the difference in grayscale values ​​between the two detection windows and the middle detection window, grey(i) represents the grayscale value of the middle detection window i, grey(ib) and grey(i+b) represent the grayscale values ​​of the b-th detection windows on both sides of the middle detection window i, respectively, where b=1 / (2×S), and S represents the ratio of the sliding step size of the detection window to the width of the detection window.

[0066] For ease of understanding, if the window width of the detection window is winSz, taking the step size as 1 / 4 of the window width as an example for illustration. The step size step is winSz / 4, then S=step / winSz=(winSz / 4) / winSz=1 / 4, and further b=1 / (2x1 / 4)=2. At this time, i-b is i-2, and i+b is i+2, then grey(i-2) represents the area gray value of the second detection window on the left side of the middle detection window i (i.e., the detection window moving two steps to the left), such as the area gray value of the left detection window 322-1 in the illustration (the sum of the gray values of the 01 area and the 02 area); grey(i+2) represents the area gray value of the second detection window on the right side of the middle detection window i (i.e., the detection window moving two steps to the right), such as the area gray value of the right detection window 322-2 in the illustration (the sum of the gray values of the 03 area and the 04 area), and grey(i) corresponds to the area gray value of the middle detection window 321 (the sum of the gray values of the 02 area and the 03 area).

[0067] Assuming that the sum of the gray values of the 01 area is sum(01), the sum of the gray values of the 02 area is sum(02), the sum of the gray values of the 03 area is sum(03), and the sum of the gray values of the 04 area is sum(04), then grey(i-2)=sum(01)+sum(02), grey(i)=sum(02)+sum(03), and grey(i+2)=sum(03)+sum(04). Substituting into Formula Two can obtain diff=(sum(01)+sum(02))+(sum(03)+sum(04))-2x(sum(02)+sum(03))=sum(01)+sum(04)-sum(02)-sum(03), which intuitively reflects the gray difference value of the middle overlapping area and the two sides of the non-overlapping normal area, and if there is a Mura defect (such as a middle area being dark or bright) in the middle detection window i, diff will present obvious positive and negative differences and / or numerical fluctuations, so as to determine the defect type and degree.

[0068] It should be noted that, since the positions of the two-side detection windows are determined based on the step size, the size of the step size has a double influence on the Mura defect detection: the smaller the step size is set, the more accurate the calculation result is, but there is an adverse trend of increasing the calculation amount and reducing the calculation efficiency; the larger the step size is set, the faster the calculation speed is, but the accuracy of the detection result may be affected due to insufficient window coverage. Compared with each other, when the step size is set to 1 / 4 of the window width, the two-side detection windows and the middle detection window form a reasonable overlapping area (each overlapping half-window size) by using the step size, which provides a basis for the calculation of the difference between the sum of the gray values of the two-side detection window regions and twice the gray value of the middle detection window region, and through offsetting the redundant gray information of the overlapping area, the gray difference between the middle region (which may contain defects) and the two-side non-overlapping normal regions is accurately captured, thereby ensuring the detection accuracy; and a good balance between the calculation amount and the detection efficiency is achieved, avoiding the problem of too low calculation efficiency caused by too small step size or insufficient accuracy caused by too large step size, so that the Mura defect detection is efficiently and accurately realized.

[0069] It can be understood that, Figure 6 The position direction of the two-side detection window is not limited to Figure 7 The horizontal direction (i.e., the X-axis direction) shown in the middle, but can also be a vertical direction (i.e., the Y-axis direction), and the difference calculation principle is similar, which will not be described here.

[0070] Further, after the absolute value of each difference value is calculated as described above, the response value corresponding to each middle detection window can be obtained, and these response values are generated into a defect response map according to the positional relationship between each other. For example, Figure 7 The defect response map of some embodiments of the present application is shown. As Figure 7 The gray value of each pixel point in the defect response map, except for the pixel points on the two-side edges (such as the two columns of pixel points on the left and right sides in the figure), represents a response value corresponding to a middle detection window.

[0071] In some embodiments, Figure 5 The defect response map shown is the same size as the region gray map shown in Figure 5 By using the region gray map shown in Figure 8 The corresponding defect response value can be directly obtained by calculating the absolute value of the difference between the gray values of the middle pixel points and the two-side pixel points of the region gray map, so that the difference operation of the region gray value is converted into the difference operation of the pixel gray value, and the calculation is more convenient.

[0072] Next, after obtaining the defect response map, the detection method 100 can continue to perform steps S104 and S105. For ease of understanding, the following will be combined with Figure 8An illustrative example is provided.

[0073] Figure 8 This illustration shows a schematic diagram of determining the location of a Mura defect based on a defect response map, according to some embodiments of this application. Figure 8 As shown, the defect response map is first segmented using a grayscale threshold. By setting an appropriate threshold, regions below or above the threshold are distinguished, resulting in a segmentation map 810 (containing highlighted areas of multiple suspected defects). In some embodiments, performing grayscale threshold segmentation on the defect response map to obtain the location of Mura defects in the defect response map includes: setting the grayscale value of locations in the defect response map with a grayscale value greater than the threshold to 255; and setting the grayscale value of locations in the defect response map with a grayscale value less than the threshold to 0. The threshold can be determined empirically.

[0074] The defect response map is an image reflecting the grayscale differences in the area to be detected, obtained through preliminary difference calculations. The Mura defect area exhibits a higher grayscale value due to its significant grayscale difference, while the background or normal areas have lower grayscale values. Grayscale thresholding is a processing method that simplifies the image by setting a critical grayscale value (threshold). Specifically, it involves judging the grayscale value of each pixel in the defect response map. If the value is greater than the preset threshold, its grayscale value is uniformly set to 255 (white); if it is less than the threshold, it is set to 0 (black), forming a binary image containing only black and white. The advantage of this method is that binarization significantly simplifies the image data, creating a sharp contrast between the Mura defect area (white) and the background area (black), facilitating rapid identification and location of defects. It also filters out low-grayscale noise interference, reducing the computational load of subsequent processing and improving the efficiency and accuracy of Mura defect detection.

[0075] Segmentation image 810 presents Mura defect regions (white) of different shapes and sizes, exhibiting varying degrees of Mura defects. Based on the size relationship between the defect response image and the aforementioned region grayscale image and integral image, the positional correspondence between the defect response image and each pixel in the area to be detected can be determined. Therefore, the positions of the Mura defects in segmentation image 810 can be mapped to their corresponding positions in the area to be detected, thus determining the location of the Mura defects in the area to be detected (or the product image). In other words, based on the window width and step size of the detection window, the defect positions in the defect response image can be restored to the area to be detected by adjusting the step size ratio between the detection window and the aforementioned region grayscale image (or defect response image).

[0076] In some other embodiments, in order to obtain the Mura defects that are really interested in actual application, length screening can be performed on the Mura defects in the defect response map to obtain the Mura defects of interest, so as to screen out the positions of the Mura defects that are really required to be detected. As shown in FIG. 8B, by performing pixel length screening on the Mura defect regions in the segmentation map 810, a defect screening map 820 can be obtained, in which a small number of longer Mura defects are retained. The length screening can be realized by setting a length threshold, and the Mura defects longer than the length threshold are retained, and the Mura defects shorter than the length threshold are removed. Figures 1-8

[0077] Further, based on the spatial correspondence (i.e., pixel position correspondence) between the defect response map and the to-be-detected region, the positions of the Mura defects determined in the defect screening map 820 are mapped to the to-be-detected region, and finally the Mura defects are accurately marked in the to-be-detected region; or can be mapped to the display screen image (or product image) where the to-be-detected region is located, to obtain a defect detection map 830 (as shown by the solid rectangular frame in the defect detection map 830), so as to realize the positioning and identification of the Mura defects in the display screen.

[0078] By the above exemplary description of the method for detecting the Mura defects on the display screen according to the embodiments of the present application in combination with the plurality of drawings, those skilled in the art can understand that, by using the integral map to obtain the region gray value, the embodiments of the present application can avoid the tedious operation of traditional pixel-by-pixel traversal and accumulation calculation of the region gray value, and improve the calculation efficiency in the Mura defect detection process; meanwhile, by using the defect response map generated by the absolute value of the sliding window difference value, the accurate detection of the Mura defect position is realized, and the efficiency and accuracy requirements of the display screen Mura defect detection are met.

[0079] The embodiments of the present application also provide an electronic device, which comprises a processor configured to execute program instructions, and a memory configured to store the program instructions, when the program instructions are loaded and executed by the processor, the processor executes the detection method according to any one of the preceding Figure 9 described in the embodiments of the present application. The following will be described in combination with the system shown in Figure 9 .

[0080] Figures 1-8 A schematic block diagram of a detection system for Mura defects on a display screen according to the embodiments of the present application is shown. The detection system 900 can comprise an electronic device 901 according to the embodiments of the present application and its peripheral devices and external network, wherein the electronic device 901 is used to detect the Mura defects on the display screen, so as to realize the technical solutions of the embodiments of the present application described in any one of the preceding Figure 9 .

[0081] ​like Figures 1-8 As shown, the electronic device 901 may include a CPU 9011, which may be a general-purpose CPU, a dedicated CPU, or other information processing and program execution unit. Furthermore, the electronic device 901 may also include a large-capacity memory 9012 and a read-only memory (ROM) 9013. The large-capacity memory 9012 can be configured to store various types of data, including product images, areas to be inspected, defect response diagrams, and various programs required to run the inspection method. The ROM 9013 can be configured to store data required for the electronic device 901's power-on self-test, the initialization of various functional modules in the system, the system's basic input / output drivers, and the data required to boot the operating system.

[0082] Furthermore, the electronic device 901 also includes other hardware platforms or components, such as the TPU 9014, GPU 9015, FPGA 9016, and MLU 9017 shown. It is understood that although various hardware platforms or components are shown in the electronic device 901, they are merely exemplary and not limiting, and those skilled in the art can add or remove corresponding hardware as needed. For example, the electronic device 901 may include only a CPU as a known hardware platform and another hardware platform as the test hardware platform of this application.

[0083] The electronic device 901 of this application also includes a communication interface 9018, through which it can connect to a local area network / wireless local area network (LAN / WLAN) 905, and further connect to a local server 906 or the Internet 907 via the LAN / WLAN. Alternatively or additionally, the electronic device 901 of this application can also directly connect to the Internet or a cellular network via the communication interface 9018 based on wireless communication technology, such as third-generation ("3G"), fourth-generation ("4G"), or fifth-generation ("5G") wireless communication technology. In some application scenarios, the electronic device 901 of this application can also access a server 908 on an external network and a possible database 909 as needed to obtain various known product images, defect width statistics, etc., and can remotely store various parameters or intermediate data.

[0084] The peripherals of the electronic device 901 can include a display device 902, an input device 903, and a data transmission interface 904. In an embodiment, the display device 902 can include, for example, one or more speakers and / or one or more visual displays configured to audibly and / or visually display the operation process or detection result of the device of the present application. The input device 903 can include, for example, a keyboard, a mouse, a microphone, a gesture capture camera, or other input buttons or controls configured to receive input of raw images or images to be detected or user instructions. The data transmission interface 904 can include, for example, a serial interface, a parallel interface, or a universal serial bus interface (“USB”), a small computer system interface (“SCSI”), Serial ATA, FireWire, PCI Express, and a high-definition multimedia interface (“HDMI”), etc., configured to transmit and interact data with other devices or systems. According to the scheme of the present application, the data transmission interface 904 can receive images of the region to be detected and the like, and transmit various types of data and results to the electronic device 901.

[0085] The above-mentioned CPU 9011, mass storage 9012, read-only memory ROM 9013, TPU 9014, GPU 9015, FPGA 9016, MLU 9017, and communication interface 9018 of the electronic device 901 of the present application can be connected to each other through a bus 9019, and realize data interaction with peripherals through the bus. In an embodiment, through the bus 9019, the CPU 9011 can control other hardware components and their peripherals in the electronic device 901.

[0086] In work, the processor CPU 9011 of the electronic device 901 of the present application can receive images of the region to be detected through the input device 903 or the data transmission interface 904, and call the computer program instructions or codes stored in the storage 9012 to detect defects in the received images of the region to be detected, to obtain Mura defect detection results. After the CPU 9011 determines the defect detection result by executing the program instructions, the defect position and type can be displayed on the display device 902 or output in the form of voice prompt. In addition, the electronic device 901 can also upload the defect detection result to the network, such as a remote database 909, through the communication interface 9018.

[0087] It should also be understood that any module, unit, component, server, computer, terminal or device of the application examples described herein can include or otherwise have access to computer readable media, such as storage media, computer storage media, or data storage devices (removable and / or non-removable) such as, for example, magnetic disks, optical disks, or tape. Computer storage media can include volatile and non- volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

[0088] Based on the foregoing, the present application also provides a computer readable storage medium having stored thereon computer readable instructions which, when executed by one or more processors, implement the detection method as described in any of the preceding embodiments. ​

[0089] The computer readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as, for example, Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random-Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), and the like, or any other medium that can be used to store the desired information and that can be accessed by an application, module, or both. Any such computer storage media can be part of the device or accessible or connectable thereto. Any application or module described herein can be implemented using computer readable / executable instructions that can be stored or otherwise held by such computer readable media.

[0090] While several embodiments of the application have been shown and described herein, it is to be understood that all such embodiments are merely illustrative of the many possible embodiments of the application. Numerous modifications, changes, and adaptations will occur to those skilled in the art upon the reading and understanding of the present specification. It is intended to cover in the appended claims all such modifications and equivalents. The specification is intended to cover the applications as they exist today as well as in the future.​

Claims

1. A method of detecting Mura defects on a display screen, characterized by, The method comprises: calculating an integral image of a to-be-detected region on a display screen, wherein an integral value of any point in the integral image is the sum of the gray values of all points in a rectangular region surrounded by the to-be-detected region and the upper left corner of the to-be-detected region; sliding a detection window in the integral image according to a set step, and calculating the sum of the gray values of the corresponding region of each detection window in the to-be-detected region during the sliding process as the region gray value of each detection window; determining the absolute value of the difference between the region gray values of two side detection windows located on both sides of each intermediate detection window and the intermediate detection window, and generating a defect response image, wherein the position direction of the two side detection windows relative to the intermediate detection window is perpendicular to the extension direction of the to-be-detected Mura defect; performing gray threshold segmentation on the defect response image to obtain the position of the Mura defect in the defect response image; determining the Mura defect in the to-be-detected region based on the correspondence between the defect response image and the to-be-detected region; The detection method further comprises: determining the type of the Mura defect based on the positive and negative of the difference value, wherein, in response to the difference value being positive, the Mura defect is determined to be a dark defect; and in response to the difference value being negative, the Mura defect is determined to be a bright defect.

2. The detection method according to claim 1, characterized in that, The region gray value of each detection window is determined based on the following formula: Sum(D)=ii(4)+ii(1)-ii(2)-ii(3); wherein Sum(D) represents the sum of the gray values of the corresponding region D of the detection window in the to-be-detected region, ii(4) represents the integral value of the point at the lower right corner of region D, ii(1) represents the integral value of the point at the upper left corner of region D, ii(2) represents the integral value of the point at the upper right corner of region D, and ii(3) represents the integral value of the point at the lower left corner of region D.

3. The method of claim 1, wherein Determining the absolute value of the difference between the region gray values of the two side detection windows located on both sides of each intermediate detection window and the intermediate detection window comprises: based on the step, determining the positions of the two side detection windows, wherein the non-overlapping regions of the two side detection windows and the intermediate detection window can be spliced into the size of one detection window; calculating the difference and the absolute value of the sum of the region gray values of the two side detection windows and twice the region gray value of the intermediate detection window.

4. The method of claim 1, wherein The position direction of the two side detection windows relative to the intermediate detection window is perpendicular to the extension direction of the to-be-detected Mura defect, which comprises: when the extension direction of the to-be-detected Mura defect to be detected is a vertical direction, the position direction of the two side detection windows relative to the intermediate detection window is a horizontal direction; when the extension direction of the to-be-detected Mura defect to be detected is a horizontal direction, the position direction of the two side detection windows relative to the intermediate detection window is a vertical direction.

5. The detection method according to claim 3, characterized in that, The difference value is calculated by the following formula: diff=grey(i-b)+grey(i+b)-2×grey(i); wherein, diff represents the difference value, grey(i) represents the area gray value of the middle detection window i, grey(i-b) and grey(i+b) respectively represent the area gray value of the b-th detection window located on both sides of the middle detection window i, wherein b=1 / (2xS), S represents the ratio of the step length to the window width of the detection window.

6. The detection method according to any one of claims 1-5, characterized in that, the window width of the detection window is greater than the defect width of the Mura defect to be detected; the detection window is set as a square; the step length is set as 1 / 4 of the window width.

7. The method of claim 1, wherein, The gray threshold segmentation of the defect response map is performed to obtain the position of the Mura defect in the defect response map, including: setting the gray value of the position with the gray value greater than the threshold in the defect response map to 255; setting the gray value of the position with the gray value less than the threshold in the defect response map to 0; The detection method further includes: length screening of the Mura defect in the defect response map to obtain the Mura defect of interest.

8. An electronic device, comprising: including: a processor configured to execute program instructions; and a memory configured to store the program instructions, which, when loaded and executed by the processor, cause the processor to execute the detection method according to any one of claims 1-7.

9. A computer readable storage medium having stored therein program instructions, wherein, When the program instructions are loaded and executed by the processor, the processor executes the detection method according to any one of claims 1-7.

Images