A method, device, system and storage medium for extracting abnormal pixels from a display screen

Abstract

The present application discloses a method, device, system and storage medium for extracting abnormal pixels from a display screen, which are used to improve the accuracy of pixel positioning and pixel extraction. A first sub-pixel brightness map is obtained; a second sub-pixel brightness map of a standard screen is obtained; several standard corner point position coordinates are determined on the second sub-pixel brightness map, and several standard templates are generated; the second sub-pixel brightness map is superimposed on the first sub-pixel brightness map; the position of the standard template is displaced as a whole, so that the standard template overlaps with the corner point area to be measured respectively, and the standard corner point position coordinates are also displaced on the first sub-pixel brightness map; after each displacement, corresponding correlation coefficients are generated for several standard corner point position coordinates, and a correlation coefficient set is generated; according to the correlation coefficient set, a corner point positioning coordinate set of the first sub-pixel brightness map is determined; according to the corner point positioning coordinate set, the screen area range data on the first sub-pixel brightness map is determined; according to the screen area range data, pixel extraction is performed.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of pixel extraction, and in particular to a method, device, system and storage medium for extracting abnormal pixels from a display screen. Background Art

[0002] With the rapid development of industrial and Internet technologies, the industry's requirements for industrial product quality are becoming more and more refined and standardized. As far as the display panel industry is concerned, the quality of products directly affects its product competitiveness. Among them, the accuracy of product shape and size specifications is the most basic process indicator of product quality. As one of the display components of the device, the display screen is used in various high-end devices, such as mobile phones, TVs, tablets, etc. As people's requirements for screen display continue to increase, the display screen has gradually become a technologically sophisticated product.

[0003] However, due to the complex manufacturing process of display screens, abnormal pixels often appear on the display screen, forming Mura defects, which are the problem of uneven screen brightness. Therefore, Demura (brightness uniformity correction) technology has become a key method to repair such defects.

[0004] Before using Demura technology, the screen area needs to be located before pixel extraction can be performed. Correct acquisition of the screen area determines the accuracy of pixel location extraction, and thus determines the defect compensation effect of Demura technology. Because the accuracy of pixel coordinates in the screen area depends largely on the accuracy of screen area detection. Generally speaking, the conventional method of obtaining the screen area is mainly to perform edge detection on the captured display area, so as to obtain the four boundary corner points of the screen area, and use the boundary corner points to determine the range of the screen area.

[0005] However, with the wide variety of display screens nowadays and different application fields, the distribution and display methods of pixels on the display screens are different, and the scope of application of this extraction method has decreased.

[0006] Nowadays, micro-displays are gradually being put into use. Micro-displays are suitable for small devices and require high clarity. The distance between sub-pixels of micro-displays has become closer, making the crosstalk between pixels more serious. In addition, the pixel arrangement of such micro-displays is varied and relatively complex. In the process of collecting with industrial cameras, the pixels at the edge of the display will easily have a large area of ​​dead spots and dark spots. If the screen area of ​​the micro-display obtained by edge detection methods is obtained, there will be edge pixel missing and overlap. Coupled with camera noise and sensor noise, the abnormal situation of dead spots and dark spots is aggravated, which in turn exacerbates the situation of missing and overlapping pixels, resulting in reduced pixel positioning and pixel extraction accuracy. Summary of the invention

[0007] The present application discloses a method, device, system and storage medium for extracting abnormal pixels from a display screen, which are used to improve the accuracy of pixel positioning and pixel extraction.

[0008] The first aspect of the present application discloses a method for extracting abnormal pixels from a display screen, comprising:

[0009] Acquire a first sub-pixel brightness map, where the first sub-pixel brightness map is an industrial camera-captured image of a display screen to be detected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map has a corner point area to be detected determined by the initial corner point detection;

[0010] Obtain a second sub-pixel brightness image of a standard screen of the same type as the display screen to be tested, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness image are clear and have no abnormalities;

[0011] Determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and define a standard corner point region as a region center according to each standard corner point position coordinate to generate a plurality of standard templates;

[0012] superimposing the second sub-pixel brightness map on the first sub-pixel brightness map;

[0013] The position of the standard template is shifted as a whole, so that the standard template overlaps with the corner point area to be measured, and the position coordinates of the standard corner point are shifted on the first sub-pixel brightness map;

[0014] After each displacement, corresponding correlation coefficients are generated for the position coordinates of several standard corner points according to the position relationship between the standard template and the corner point area to be measured, and a correlation coefficient set is generated. The correlation coefficient set includes the current position coordinates of each standard corner point after each displacement and the corresponding correlation coefficient;

[0015] Determine a set of corner point positioning coordinates of the first sub-pixel brightness map according to the set of correlation coefficients;

[0016] Determine the screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set;

[0017] Pixel extraction is performed based on screen area range data.

[0018] Optionally, the screen area range data includes relevant row coordinates and relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map;

[0019] Determining screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set includes:

[0020] Calculate the projection transformation matrix based on the corner point positioning coordinate set and several standard corner point position coordinates;

[0021] Get the first origin coordinate, the second origin coordinate and the related constant array, the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin;

[0022] The relevant row coordinates and the relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map are calculated according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

[0023] Optionally, pixel extraction is performed according to the screen area range data, including:

[0024] Generate sub-pixel positioning initial coordinates according to the arrangement mode of the display screen to be tested and the related row coordinates and the related column coordinates;

[0025] The initial sub-pixel positioning coordinates are mapped and transformed according to a preset affine transformation matrix to obtain the final sub-pixel positioning coordinates;

[0026] A pixel extraction operation is performed on the first sub-pixel brightness map according to the final positioning coordinates of the sub-pixels.

[0027] Optionally, after each displacement, corresponding correlation coefficients are generated for the position coordinates of a number of standard corner points according to the positional relationship between the standard template and the corner point area to be measured, and a set of correlation coefficients is generated, including:

[0028] After completing one displacement, the real-time position of each standard corner point position coordinate in the first sub-pixel brightness image is recorded;

[0029] Calculate the overlapping area of ​​each standard template and its corresponding corner point area to be measured, and generate an overlapping area set;

[0030] A set of correlation coefficients is generated for each real-time position according to the set of overlapping areas and the overlapping coefficients.

[0031] Optionally, obtaining a first sub-pixel brightness map includes:

[0032] Lighting up sub-pixels of a preset color grayscale on the display screen to be detected according to a preset lighting method;

[0033] The first sub-pixel brightness image of the display screen to be tested is collected by an industrial camera, and the pixels on the display screen to be tested are closely arranged;

[0034] A plurality of corner point regions to be detected are generated on the first sub-pixel brightness map by using an edge detection method.

[0035] The second aspect of the present application discloses a device for extracting abnormal pixels of a display screen, comprising:

[0036] A first acquisition unit is used to acquire a first sub-pixel brightness map, where the first sub-pixel brightness map is an industrial camera-captured image of the display screen to be detected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map has a corner point area to be detected determined by the initial corner point detection;

[0037] The second acquisition unit is used to acquire a second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities;

[0038] A first generating unit is used to determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and to define a standard corner point region with each standard corner point position coordinate as the region center to generate a plurality of standard templates;

[0039] A stacking unit, used to stack the second sub-pixel brightness map on the first sub-pixel brightness map;

[0040] A displacement unit is used to displace the position of the standard template as a whole, so that the standard template overlaps with the corner point area to be measured, and also displaces the position coordinates of the standard corner point on the first sub-pixel brightness map;

[0041] A second generating unit is used to generate corresponding correlation coefficients for a number of standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be measured after each displacement, and generate a correlation coefficient set, wherein the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient;

[0042] A first determining unit, configured to determine a set of corner point positioning coordinates of a first sub-pixel brightness map according to a set of correlation coefficients;

[0043] A second determining unit, configured to determine screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set;

[0044] The extraction unit is used to extract pixels according to the screen area range data.

[0045] Optionally, the screen area range data includes relevant row coordinates and relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map;

[0046] The second determining unit includes:

[0047] Calculate the projection transformation matrix based on the corner point positioning coordinate set and several standard corner point position coordinates;

[0048] Get the first origin coordinate, the second origin coordinate and the related constant array, the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin;

[0049] The relevant row coordinates and the relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map are calculated according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

[0050] Optionally, the extraction unit comprises:

[0051] Generate sub-pixel positioning initial coordinates according to the arrangement mode of the display screen to be tested and the related row coordinates and the related column coordinates;

[0052] The initial sub-pixel positioning coordinates are mapped and transformed according to a preset affine transformation matrix to obtain the final sub-pixel positioning coordinates;

[0053] A pixel extraction operation is performed on the first sub-pixel brightness map according to the final positioning coordinates of the sub-pixels.

[0054] Optionally, the second generation unit includes:

[0055] After completing one displacement, the real-time position of each standard corner point position coordinate in the first sub-pixel brightness image is recorded;

[0056] Calculate the overlapping area of ​​each standard template and its corresponding corner point area to be measured, and generate an overlapping area set;

[0057] A set of correlation coefficients is generated for each real-time position according to the set of overlapping areas and the overlapping coefficients.

[0058] Optionally, the first acquisition unit includes:

[0059] Lighting up sub-pixels of a preset color grayscale on the display screen to be detected according to a preset lighting method;

[0060] The first sub-pixel brightness image of the display screen to be tested is collected by an industrial camera, and the pixels on the display screen to be tested are closely arranged;

[0061] A plurality of corner point regions to be detected are generated on the first sub-pixel brightness map by using an edge detection method.

[0062] A third aspect of the present application provides a system for extracting abnormal pixels from a display screen, comprising:

[0063] Processor, memory, input-output unit, and bus;

[0064] The processor is connected to the memory, the input and output unit, and the bus;

[0065] The memory stores a program, and the processor calls the program to execute the first aspect and any optional method of the first aspect.

[0066] A fourth aspect of the present application provides a computer-readable storage medium, on which a program is stored. When the program is executed on a computer, the program executes the first aspect and any optional method of the first aspect.

[0067] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0068] In the present application, a first sub-pixel brightness map is first obtained, and the first sub-pixel brightness map is an industrial camera acquisition image of a display screen to be detected that lights up a sub-pixel of a preset color grayscale. The first sub-pixel brightness map has a corner point area to be detected determined by the initial corner point detection. A second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected is obtained, and the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities. Several standard corner point position coordinates are determined on the second sub-pixel brightness map, and a standard corner point area is delineated as the center of the area according to each standard corner point position coordinate to generate several standard templates. The second sub-pixel brightness map is superimposed on the first sub-pixel brightness map. The position of the standard template is shifted as a whole, so that the standard template overlaps with the corner point area to be detected respectively, and the standard corner point position coordinates are also shifted on the first sub-pixel brightness map. After each displacement, corresponding correlation coefficients are generated for several standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be detected, and a correlation coefficient set is generated, and the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient. Determine the corner point positioning coordinate set of the first sub-pixel brightness map according to the correlation coefficient set. Determine the screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set. Perform pixel extraction according to the screen area range data.

[0069] A standard template with clear pixel and corner point positions is generated through a standard screen. The standard template is placed on the first sub-pixel brightness map for displacement. Each displacement is analyzed through the positional relationship between the standard template and the corner point area to be measured, so as to calculate the correlation coefficient of the coordinate position of the current set of standard corner point position coordinates on the first sub-pixel brightness image, that is, the weight coefficient. After multiple displacements, multiple sets of correlation coefficients are collected, and then the optimal corner point positioning coordinates are determined by the correlation coefficients. The screen area range data on the first sub-pixel brightness map is determined by the corner point positioning coordinates, and finally pixels are extracted based on the screen area range data. The above-mentioned method of using a standard template can deal with the abnormal situation of dead spots and dark spots on the edge pixels caused by dense pixels and complex arrangements, and at the same time solves the problem that the edge detection method cannot accurately detect accurately, thereby improving the accuracy of pixel positioning and pixel extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0071] Figure 1 A schematic diagram of an embodiment of the method for extracting abnormal pixels from a display screen of the present application;

[0072] Figure 2 A schematic diagram of an embodiment of a method for determining screen area range data of the present application;

[0073] Figure 3 A schematic diagram of an embodiment of the pixel extraction method of the present application;

[0074] Figure 4 A schematic diagram of an embodiment of a method for generating correlation coefficients of the present application;

[0075] Figure 5 A schematic diagram of an embodiment of a method for generating a first sub-pixel brightness map according to the present application;

[0076] Figure 6 A schematic diagram of an embodiment of a device for extracting abnormal pixels from a display screen according to the present application;

[0077] Figure 7 A schematic diagram of an embodiment of a system for extracting abnormal pixels from a display screen according to the present application;

[0078] Figure 8 A schematic diagram of edge detection and template detection for this application;

[0079] Fig. 9 Another schematic diagram of edge detection and template detection for this application;

[0080] Fig.10 A schematic diagram of edge detection and template detection for this application;

[0081] Fig.11 A schematic diagram of the final positioning of sub-pixels under different sub-pixel abnormal conditions of the present application;

[0082] Fig.12 Another schematic diagram of the final positioning of sub-pixels under different sub-pixel abnormal conditions of the present application. DETAILED DESCRIPTION

[0083] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0084] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0085] It should also be understood that the terms "and, or" used in the specification and appended claims refer to and include any and all possible combinations of one or more of the associated listed items.

[0086] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0087] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0088] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0089] In the prior art, micro-display screens are gradually being put into use. Micro-display screens are suitable for small devices and require high definition. The distance between sub-pixels of micro-display screens becomes closer, which makes the crosstalk between pixels more serious. In addition, the pixel arrangement of such micro-display screens is varied and relatively complex. In the process of collecting data using industrial cameras, the pixels at the edge of the display screen will easily have a large number of abnormal dead spots and dark spots. If the screen area of ​​the micro-display screen obtained using the edge detection method will have edge pixel missing and overlap. Coupled with camera noise and sensor noise, the abnormal situation of dead spots and dark spots is aggravated, which in turn exacerbates the situation of missing and overlapping pixels, resulting in reduced pixel positioning and pixel extraction accuracy.

[0090] Based on this, the present application discloses a method, device, system and storage medium for extracting abnormal pixels from a display screen, which are used to improve the accuracy of pixel positioning and pixel extraction.

[0091] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0092] The method of the present application can be applied to a server, a device, a terminal or other devices with logic processing capabilities, and the present application does not limit this. For the convenience of description, the following description is made by taking the execution subject as an example of a terminal.

[0093] See also Figure 1 The present application provides an embodiment of a method for adjusting consistency of multiple cameras of a spliced ​​screen, including:

[0094] 101. Obtain a first sub-pixel brightness map, where the first sub-pixel brightness map is an image captured by an industrial camera of a display screen to be inspected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map contains a corner point region to be inspected determined by initial corner point detection;

[0095] In this embodiment, the terminal lights up the display screen to be detected, specifically, lights up the sub-pixels of a certain color grayscale, for example: RGB pixels, only the R pixels are lit, the grayscale value is A, and the arrangement of the R pixels is a cross-mixed arrangement.

[0096] After the terminal lights up the pixel to be detected, it takes a picture through the industrial camera to obtain the first sub-pixel brightness image, and then can generate 4 corner points to be corrected according to the algorithm. With the initial corner point as the center, a corner point area to be detected with a known length and width is generated. The display screen to be detected in this embodiment, taking a conventional rectangular display screen as an example, has 4 corner points, and 4 corner point areas to be detected will be generated on the first sub-pixel brightness image.

[0097] 102. Obtain a second sub-pixel brightness image of a standard screen of the same type as the display screen to be detected, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness image are clear and have no abnormalities;

[0098] The terminal obtains a second sub-pixel brightness image of a standard screen of the same type as the display screen to be tested, that is, obtains a standard defect-free display screen, which is at least consistent in size and pixel arrangement, and preferably the model is exactly the same. The pixels on the standard screen in this embodiment are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness image are clear and have no abnormalities. The image can be collected in advance.

[0099] 103. Determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and define a standard corner point region with each standard corner point position coordinate as the region center to generate a plurality of standard templates;

[0100] The terminal determines the standard corner point position coordinates on the second sub-pixel brightness map. This step can also be completed in advance, and then defines a standard corner point area as the center of the area according to each standard corner point position coordinate to generate several standard templates. For example, if there are 4 corner points, 4 standard templates can be generated.

[0101] It should be noted that the size of the standard template here is the same as the size of the corner point area to be measured.

[0102] 104. Overlay the second sub-pixel brightness map on the first sub-pixel brightness map;

[0103] The terminal overlays the second sub-pixel brightness map on the first sub-pixel brightness map, which actually places the position of the standard template and the coordinates of the standard corner point position on the first sub-pixel image.

[0104] Since the display screen to be detected may have dark spots or one line missing, the position of the corner point area to be detected should generally correspond to the position of the standard template at most, otherwise it can only be indented inward as a whole or in part.

[0105] 105. Overall shifting the position of the standard template so that the standard template overlaps with the corner point regions to be measured, and also shifting the position coordinates of the standard corner points on the first sub-pixel brightness map;

[0106] In this embodiment, the terminal as a whole shifts the position of the standard template so that the standard template overlaps with the corner point regions to be measured, and also shifts the position coordinates of the standard corner points on the first sub-pixel brightness map.

[0107] 106. After each displacement, generate corresponding correlation coefficients for a number of standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be measured, and generate a correlation coefficient set, wherein the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient;

[0108] After each displacement, the terminal generates corresponding correlation coefficients for the position coordinates of several standard corner points according to the positional relationship between the standard template and the corner point area to be measured, thereby generating a set of correlation coefficients.

[0109] Specifically, it is necessary to generate a correlation coefficient for the current position of each standard corner point position coordinate according to the overlapping area between each standard template and the corner point area to be measured, and the orientation of the area where the standard template exceeds the part.

[0110] The size of the overlapping area is proportional to the correlation coefficient. The terminal calculates the correlation coefficient for the center point of the standard template (the position after the standard corner point position coordinates are translated) based on the overlapping area of ​​each standard template and its corresponding corner point area to be measured. This displacement can generate a correlation coefficient for each standard corner point position coordinate, or a correlation coefficient for the entire group of points.

[0111] 107. Determine a set of corner point positioning coordinates of the first sub-pixel brightness image according to the set of correlation coefficients;

[0112] Specifically, taking a rectangular screen as an example: the standard corner point position coordinates of the four standard template centers are (x1, y1), (x2, y2), (x3, y3) and (x4, y4), and the correlation coefficients of the second sub-pixel brightness map and the first sub-pixel brightness map at each position are used. The point with the largest correlation value is used as the precise corner point positioning coordinates of the first sub-pixel brightness map (x1', y1'), (x2', y2'), (x3', y3') and (x4', y4').

[0113] 108. Determine screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set;

[0114] Next, the terminal can determine the screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set.

[0115] 109. Extract pixels based on screen area range data.

[0116] The terminal finally extracts pixels based on the screen area range data of the positioning number.

[0117] In this embodiment, firstly, a first sub-pixel brightness map is obtained, which is an industrial camera captured image of a display screen to be detected with sub-pixels of a preset color grayscale lit, and the first sub-pixel brightness map has a corner point area to be detected determined by the initial corner point detection. A second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected is obtained, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities. Several standard corner point position coordinates are determined on the second sub-pixel brightness map, and a standard corner point area is delineated as the center of the area according to each standard corner point position coordinate to generate several standard templates. The second sub-pixel brightness map is superimposed on the first sub-pixel brightness map. The position of the standard template is shifted as a whole, so that the standard template overlaps with the corner point area to be detected respectively, and the standard corner point position coordinates are also shifted on the first sub-pixel brightness map. After each displacement, corresponding correlation coefficients are generated for several standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be detected, and a correlation coefficient set is generated, and the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient. Determine the corner point positioning coordinate set of the first sub-pixel brightness map according to the correlation coefficient set. Determine the screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set. Perform pixel extraction according to the screen area range data.

[0118] A standard template with clear pixel and corner point positions is generated through a standard screen. The standard template is placed on the first sub-pixel brightness map for displacement. Each displacement is analyzed through the positional relationship between the standard template and the corner point area to be measured, so as to calculate the correlation coefficient of the coordinate position of the current set of standard corner point position coordinates on the first sub-pixel brightness image, that is, the weight coefficient. After multiple displacements, multiple sets of correlation coefficients are collected, and then the optimal corner point positioning coordinates are determined by the correlation coefficients. The screen area range data on the first sub-pixel brightness map is determined by the corner point positioning coordinates, and finally pixels are extracted based on the screen area range data. The above-mentioned method of using a standard template can deal with the abnormal situation of dead spots and dark spots on the edge pixels caused by dense pixels and complex arrangements, and at the same time solves the problem that the edge detection method cannot accurately detect accurately, thereby improving the accuracy of pixel positioning and pixel extraction.

[0119] The following is a demonstration and analysis of the effects of the template matching method and the traditional edge detection method. Figure 8 As shown, Figure 8 This is a schematic diagram of edge detection and template matching for a portion of the display screen when there are many missing pixels in the upper edge row of the screen. Normally lit pixels will form a pixel cluster area, and its center position cannot be directly determined. Direct edge detection (dark blue line box) of the screen area is fitted to obtain a range of approximately one row of sub-pixels missing in the screen area. The template matching method of the present application can obtain a relatively accurate upper left corner point of the screen area (green line intersection point). The light blue box here is the corner point positioning box generated after the intersection point is determined; Fig. 9 This is a schematic diagram of edge detection and template matching for a portion of the display screen when there are many missing columns of pixels on the left edge of the screen. Direct edge detection (dark blue line box) of the screen area is fitted to obtain the range of the screen area where a column of sub-pixels will be missing, and the correct lower left corner point of the screen area (green line intersection point) can be obtained through the template matching method. The green box here is the corner point positioning box generated after the intersection point is determined, and the light blue line here is the auxiliary line for corner point positioning; Fig.10The diagram is a schematic diagram of edge detection and template matching for part of the display screen area when there are many missing row and column pixels in the upper right corner of the screen. Direct edge detection (dark blue line box) of the screen area is fitted to obtain the range of the screen area where some row and column sub-pixels will be missing, and the template matching method can obtain the correct corner point of the upper right corner of the screen area (green line intersection point). The green box here is the corner point positioning box generated after the intersection point is determined, and the light blue line here is the auxiliary line for corner point positioning. Therefore, when there are many abnormalities in the number of sub-pixels on the edge of the screen, the template matching method can adapt to various abnormal situations of edge pixels. Its effect is better than that of traditional edge detection, and the correct screen area range is obtained, the accuracy of pixel extraction is improved, and the problem of incorrect positioning coordinates of the four corner points caused by the missing screen area after direct edge detection of the screen area can be avoided.

[0120] See also Figure 2 The present application provides an embodiment of a method for determining screen area range data, comprising:

[0121] 201. Calculate a projection transformation matrix according to a set of corner point positioning coordinates and a number of standard corner point position coordinates;

[0122] In this embodiment, the terminal calculates the projection transformation matrix according to the corner point positioning coordinate set and a number of standard corner point position coordinates. The projection affine transformation matrix calculation formula is:

[0123]

[0124] Among them, Point.x is the row coordinate set of the four standard corner point position coordinates corresponding to the display screen area to be tested of the first sub-pixel brightness map, and Point.y is the column coordinate set of the four standard corner point position coordinates corresponding to the display screen area to be tested of the first sub-pixel brightness map; Point.x` is the row coordinate set (x1', x2', x3', x4') of the four corner point positioning of the display screen to be tested obtained by template matching, Point.y` is the column coordinate set (y1', y2', y3', y4') of the four corner point positioning of the display screen to be tested obtained by template matching, and Matrix is ​​a 3*3 projection affine transformation matrix.

[0125] 202. Obtain the first origin coordinate, the second origin coordinate and a related constant array, wherein the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin;

[0126] 203. Calculate relevant row coordinates and relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

[0127] The terminal obtains the first origin coordinate, the second origin coordinate and the related constant array. The first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin. The second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin. The specific calculation formula for determining the current screen area range through the projection transformation matrix is:

[0128]

[0129] Among them, Px is the actual row coordinate of the display screen to be tested in the camera's field of view with the upper left corner of the screen as the origin, Py is the actual vertical coordinate of the display screen to be tested in the industrial camera's field of view with the upper left corner of the display screen to be tested as the origin, Pz is a constant array, Qx, Qy, and Qz are the relevant row and column coordinates and coefficients of the display screen area to be tested in the target image calculated according to the projection matrix.

[0130] See also Figure 3 , the present application provides an embodiment of a pixel extraction method, comprising:

[0131] 301. Generate sub-pixel positioning initial coordinates according to the arrangement mode of the display screen to be tested and the relevant row coordinates and relevant column coordinates;

[0132] 302. Mapping and transforming the initial sub-pixel positioning coordinates according to a preset affine transformation matrix to obtain final sub-pixel positioning coordinates;

[0133] 303. Perform a pixel extraction operation on the first sub-pixel brightness map according to the final sub-pixel positioning coordinates.

[0134] First, the terminal generates the initial sub-pixel positioning coordinates according to the display area range of the current display screen to be tested and the distribution mode of the screen sub-pixels; secondly, according to the affine transformation matrix, the initial sub-pixel positioning coordinates are mapped and transformed to obtain the final sub-pixel positioning coordinates; finally, the pixel extraction operation is performed on the target image according to the final sub-pixel positioning coordinates. Fig.11 and Fig.12 As shown, Fig.11 This is a schematic diagram of the final positioning coordinates when a large number of sub-pixels are missing at the upper left corner of the screen area. Fig.12 This is a schematic diagram of the final positioning coordinates when the sub-pixel is missing and there is a dark spot in the lower right corner of the screen area of ​​this application. It can be seen that there are blurred pixels (dark spots, unclear pixel clusters) and missing pixels (pixel clusters can no longer be detected) in both figures. The green dot in the pixel cluster is the final positioning coordinate of the sub-pixel on the target image. When there are many abnormal pixels at the edge of the screen area (including but not limited to sub-pixel missing, dark spots and other abnormal situations), accurate positioning can also be achieved.

[0135] See also Figure 4 The present application provides an embodiment of a method for generating a correlation coefficient, comprising:

[0136] 401. After completing one displacement, record the real-time position of each current standard corner point position coordinate in the first sub-pixel brightness image;

[0137] 402. Calculate the overlapping area between each standard template and its corresponding corner point area to be measured, and generate an overlapping area set;

[0138] 403. Generate a correlation coefficient set for each real-time position according to the overlapping area set and the overlapping coefficient.

[0139] In this embodiment, after completing a displacement, the terminal records the real-time position of each standard corner point position coordinate in the first sub-pixel brightness image. Then the terminal calculates the overlapping area of ​​each standard template and its corresponding corner point area to be measured, generates an overlapping area set, and then generates independent coefficients in respective directions according to the overlapping area set. Then, a correlation coefficient set is generated for each real-time position through the independent coefficient and the overlapping coefficient.

[0140] It should be noted that when a standard template is completely separated from the corresponding corner point area to be measured, the overlap coefficient is set to 0. Secondly, if the outermost edge of a standard template is inside the outermost edge of the corresponding corner point area to be measured, the overlap coefficient is set to 0. When the standard template and the corresponding corner point area to be measured overlap, the corner point area to be measured corresponding to the standard template is located on the inner side, or the outer edge of the corner point area to be measured and the outermost edge of the standard template are on the same straight line, U is set to 1.

[0141] Generate independent coefficients for each direction based on the overlapping area set. Specifically, assume that the current point is the target point. According to the abnormal area of ​​each standard module in each direction, determine the number of dark spots and missing pixels on each side based on the abnormal area. For example, the standard template in the upper left corner has non-overlapping abnormal areas on the upper and left sides. Cut out the area on the upper side and divide it by the length of the upper side to generate the missing length. Then determine the pixel row spacing based on the arrangement of the pixels. Divide the missing length by the pixel row spacing to determine the missing row value. In this way, determine the missing row values ​​in the four directions of up, down, left, and right, and calculate the independent coefficients based on the missing row values.

[0142] d1=t0 / t1

[0143] d2=t0 / t2

[0144] d3=t0 / t3

[0145] d4=t0 / t4

[0146] d1, d2, d3, and d4 are independent coefficients in the four directions of up, down, left, and right. t1, t2, t3, and t4 are missing row values. It can be seen that the larger the missing row value, the smaller the independent coefficient. t0 is the average missing row value when the display screen under test is abnormal.

[0147] Finally, a set of correlation coefficients is generated for each real-time position through independent coefficients and overlapping coefficients, in the following format:

[0148] s1=U*d1*R0

[0149] s2=U*d2*R0

[0150] s3=U*d3*R0

[0151] s4=U*d4*R0

[0152] s1, s2, s3, and s4 generate correlation coefficients for each real-time position. R0 is the average area of ​​the overlapping areas of the four standard templates and the respective corner point areas to be measured.

[0153] R0=(R1+R2+R3+R4) / 4

[0154] R1, R2, R3, and R4 are the overlapping areas of the four standard templates and their respective corner point regions to be measured.

[0155] This set of standard corner point position coordinates can also produce an overall correlation coefficient in the following format:

[0156] S0=s1+s2+s3+s4

[0157] In this embodiment, the correlation coefficient is calculated by using the average missing row value when the display screen to be tested is abnormal and the current actual missing row value, and the overlap coefficient is integrated to improve the calculation precision and accuracy of the correlation coefficient.

[0158] See also Figure 5 The present application provides an embodiment of a method for generating a first sub-pixel brightness map, comprising:

[0159] 501. Light up sub-pixels of a preset color grayscale on the display screen to be detected according to a preset lighting method;

[0160] The terminal lights up the sub-pixels of preset color grayscale on the display screen to be detected according to a preset lighting method. The sub-pixel screen lighting method of the target image can be full lighting, alternate row and alternate column lighting, etc., which is not limited here.

[0161] 502. Collecting a first sub-pixel brightness image of the display screen to be tested by an industrial camera, where the pixels on the display screen to be tested are closely arranged;

[0162] The sub-pixel brightness image of the display screen at different preset color grayscales is obtained through an industrial camera. Specifically, the terminal collects the first sub-pixel brightness image of the display screen to be tested through the industrial camera, and the pixels on the display screen to be tested are closely arranged.

[0163] 503. Generate a plurality of corner point regions to be detected on the first sub-pixel brightness map by using an edge detection method.

[0164] The terminal first detects the edge through edge detection, generates an edge fitting line, and then determines the initial corner point according to the intersection of the edge fitting, and generates a corner point area to be detected with a known length and width centered on the initial corner point. The advantage of the edge detection algorithm is that it generates a fitting straight line, and then excludes points with large differences through point screening. Further fitting, 4 fitting straight lines can form a rectangle. After determining the initial corner point, the row coordinates or vertical coordinates of two adjacent corner points are relatively closer, and the calculated correlation coefficient will be more accurate.

[0165] See also Figure 6 The present application provides an embodiment of a device for extracting abnormal pixels from a display screen, comprising:

[0166] A first acquisition unit 601 is used to acquire a first sub-pixel brightness map, where the first sub-pixel brightness map is an industrial camera-captured image of a display screen to be detected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map has a corner point area to be detected determined by initial corner point detection;

[0167] Optionally, the first acquisition unit 601 includes:

[0168] Lighting up sub-pixels of a preset color grayscale on the display screen to be detected according to a preset lighting method;

[0169] The first sub-pixel brightness image of the display screen to be tested is collected by an industrial camera, and the pixels on the display screen to be tested are closely arranged;

[0170] A plurality of corner point regions to be detected are generated on the first sub-pixel brightness map by using an edge detection method.

[0171] The second acquisition unit 602 is used to acquire a second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities;

[0172] The first generating unit 603 is used to determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and define a standard corner point region with each standard corner point position coordinate as the region center to generate a plurality of standard templates;

[0173] An overlay unit 604, configured to overlay the second sub-pixel brightness map on the first sub-pixel brightness map;

[0174] A displacement unit 605 is used to displace the position of the standard template as a whole, so that the standard template overlaps with the corner point area to be measured, and also displaces the position coordinates of the standard corner point on the first sub-pixel brightness map;

[0175] A second generating unit 606 is used to generate corresponding correlation coefficients for a number of standard corner point position coordinates according to the position relationship between the standard template and the corner point area to be measured after each displacement, and generate a correlation coefficient set, wherein the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient;

[0176] Optionally, the second generating unit 606 includes:

[0177] After completing one displacement, the real-time position of each standard corner point position coordinate in the first sub-pixel brightness image is recorded;

[0178] Calculate the overlapping area of ​​each standard template and its corresponding corner point area to be measured, and generate an overlapping area set;

[0179] A set of correlation coefficients is generated for each real-time position according to the set of overlapping areas and the overlapping coefficients.

[0180] A first determining unit 607, configured to determine a set of corner point positioning coordinates of a first sub-pixel brightness map according to a set of correlation coefficients;

[0181] A second determining unit 608, configured to determine screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set;

[0182] The second determining unit 608 includes:

[0183] Calculate the projection transformation matrix based on the corner point positioning coordinate set and several standard corner point position coordinates;

[0184] Get the first origin coordinate, the second origin coordinate and the related constant array, the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin;

[0185] The relevant row coordinates and the relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map are calculated according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

[0186] The extraction unit 609 is used to extract pixels according to the screen area range data.

[0187] Optionally, the extraction unit 609 includes:

[0188] Generate sub-pixel positioning initial coordinates according to the arrangement mode of the display screen to be tested and the related row coordinates and the related column coordinates;

[0189] The initial sub-pixel positioning coordinates are mapped and transformed according to a preset affine transformation matrix to obtain the final sub-pixel positioning coordinates;

[0190] A pixel extraction operation is performed on the first sub-pixel brightness map according to the final positioning coordinates of the sub-pixels.

[0191] See also Figure 7 The present application provides a system for extracting abnormal pixels from a display screen, comprising:

[0192] Processor 701 , memory 702 , input / output unit 703 , and bus 704 .

[0193] The processor 701 is connected to the memory 702 , the input and output unit 703 , and the bus 704 .

[0194] The memory 702 stores a program, and the processor 701 calls the program to execute the following steps: Figure 1 , Figure 2 and Figure 3 , Figure 4 and Figure 5 The method in .

[0195] The present application provides a computer-readable storage medium, on which a program is stored, and when the program is executed on a computer, the program performs the following steps: Figure 1 , Figure 2 and Figure 3 , Figure 4 and Figure 5 The method in .

[0196] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0197] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0198] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0199] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0200] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A method for extracting abnormal pixels from a display screen, characterized in that: include: Acquire a first sub-pixel brightness map, where the first sub-pixel brightness map is an industrial camera-captured image of a display screen to be detected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map has a corner point area to be detected determined by initial corner point detection; Acquire a second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities; Determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and define a standard corner point region with each standard corner point position coordinate as the region center to generate a plurality of standard templates; Overlaying the second sub-pixel brightness map on the first sub-pixel brightness map; The position of the standard template is shifted as a whole, so that the standard template overlaps with the corner point area to be measured respectively, and the position coordinates of the standard corner point are shifted on the first sub-pixel brightness map; After each displacement, corresponding correlation coefficients are generated for a plurality of the standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be measured, and a correlation coefficient set is generated, wherein the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient; After each displacement, corresponding correlation coefficients are generated for a number of the standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be measured, and a correlation coefficient set is generated, including: after completing one displacement, recording the real-time position of each current standard corner point position coordinate in the first sub-pixel brightness map; calculating the overlapping area between each standard template and its corresponding corner point area to be measured, and generating an overlapping area set; generating a correlation coefficient set for each real-time position according to the overlapping area set and the overlapping coefficient; Determining a set of corner point positioning coordinates of the first sub-pixel brightness map according to the set of correlation coefficients; Determine the screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set; Pixel extraction is performed according to the screen area range data.

2. The method according to claim 1, characterized in that The screen area range data includes relevant row coordinates and relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map; Determining screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set includes: Calculating a projection transformation matrix according to the corner point positioning coordinate set and a plurality of standard corner point position coordinates; Obtain the first origin coordinate, the second origin coordinate and a related constant array, wherein the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin; The relevant row coordinates and the relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map are calculated according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

3. The method according to claim 2, characterized in that Pixel extraction is performed according to the screen area range data, including: Generating sub-pixel positioning initial coordinates according to the arrangement of the display screen to be tested and the relevant row coordinates and the relevant column coordinates; Mapping and transforming the sub-pixel positioning initial coordinates according to a preset affine transformation matrix to obtain the sub-pixel final positioning coordinates; A pixel extraction operation is performed on the first sub-pixel brightness map according to the final positioning coordinates of the sub-pixel.

4. The method according to any one of claims 1 to 3, characterized in that Obtain the first sub-pixel brightness map, including: Lighting up sub-pixels of a preset color grayscale on the display screen to be detected according to a preset lighting method; Collecting a first sub-pixel brightness image of a display screen to be tested by an industrial camera, wherein the pixels on the display screen to be tested are closely arranged; A plurality of corner point regions to be detected are generated on the first sub-pixel brightness map by using an edge detection method.

5. A device for extracting abnormal pixels from a display screen, characterized in that: include: A first acquisition unit is used to acquire a first sub-pixel brightness map, where the first sub-pixel brightness map is an industrial camera-captured image of a display screen to be detected that lights up sub-pixels of a preset color grayscale, and the first sub-pixel brightness map has a corner point area to be detected determined by initial corner point detection; A second acquisition unit is used to acquire a second sub-pixel brightness map of a standard screen of the same type as the display screen to be detected, the pixels on the standard screen are clear and have no abnormalities, and the pixels of the preset color grayscale collected on the second sub-pixel brightness map are clear and have no abnormalities; A first generating unit, configured to determine a plurality of standard corner point position coordinates on the second sub-pixel brightness map, and define a standard corner point region with each standard corner point position coordinate as the region center to generate a plurality of standard templates; a stacking unit, configured to stack the second sub-pixel brightness map on the first sub-pixel brightness map; A displacement unit, used for displacing the position of the standard template as a whole, so that the standard template overlaps with the corner point area to be measured respectively, and also displacing the position coordinates of the standard corner point on the first sub-pixel brightness map; A second generating unit is used to generate corresponding correlation coefficients for a plurality of the standard corner point position coordinates according to the positional relationship between the standard template and the corner point area to be measured after each displacement, and generate a correlation coefficient set, wherein the correlation coefficient set includes the current position coordinates of each standard corner point position after each displacement and the corresponding correlation coefficient; The second generation unit includes: After completing one displacement, the real-time position of each standard corner point position coordinate in the first sub-pixel brightness map is recorded; Calculate the overlapping area of ​​each standard template and its corresponding corner point area to be measured, and generate an overlapping area set; Generate a set of correlation coefficients for each real-time position based on the set of overlapping areas and the overlapping coefficients; A first determining unit, configured to determine a set of corner point positioning coordinates of the first sub-pixel brightness map according to the set of correlation coefficients; A second determining unit, configured to determine screen area range data on the first sub-pixel brightness map according to the corner point positioning coordinate set; An extraction unit is used to extract pixels according to the screen area range data.

6. The device according to claim 5, characterized in that The screen area range data includes relevant row coordinates and relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map; The second determining unit includes: Calculating a projection transformation matrix according to the corner point positioning coordinate set and a plurality of standard corner point position coordinates; Obtain the first origin coordinate, the second origin coordinate and a related constant array, wherein the first origin coordinate is the actual row coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin, and the second origin coordinate is the actual vertical coordinate of the display screen to be detected in the field of view of the industrial camera with the upper left corner of the display screen to be detected as the origin; The relevant row coordinates and the relevant column coordinates of the area of ​​the display screen to be tested in the first sub-pixel brightness map are calculated according to the first origin coordinates, the second origin coordinates, the relevant constant array and the projection transformation matrix.

7. The device according to claim 6, characterized in that Extraction unit, comprising: Generating sub-pixel positioning initial coordinates according to the arrangement of the display screen to be tested and the relevant row coordinates and the relevant column coordinates; Mapping and transforming the sub-pixel positioning initial coordinates according to a preset affine transformation matrix to obtain the sub-pixel final positioning coordinates; A pixel extraction operation is performed on the first sub-pixel brightness map according to the final positioning coordinates of the sub-pixel.

8. A system for extracting abnormal pixels from a display screen, characterized in that: include: Processor, memory, input-output unit, and bus; The processor is connected to the memory, the input and output unit, and the bus; The memory stores a program, and the processor calls the program to execute the method according to any one of claims 1 to 4.

9. A computer-readable storage medium having a program stored thereon, wherein the program, when executed on a computer, performs the method according to any one of claims 1 to 4.

Images