Background subtraction method and device based on display screen and storage medium

Abstract

The invention discloses a background subtraction method and device based on a display screen and a storage medium, which are used for reducing the influence of ambient light and invisible light. Placing the target display screen in a darkroom environment to enable the target display screen to display a preset background picture; generating a to-be-segmented exposure time set; performing image acquisition on the target display screen to generate a to-be-segmented image set; performing quasi-linear interval segmentation on the to-be-segmented exposure time set according to the gray data in the to-be-segmented image set to generate a quasi-linear exposure interval set; generating a background subtraction exposure time set according to the collection exposure time set and the quasi-linear exposure interval set; collecting a target display screen to generate a background image set; using the target display screen to display each display screen Pattern picture; performing image acquisition on the target display screen to generate a display screen acquisition image set; and performing background subtraction processing on the images in the display screen acquisition image set by using the background image set.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of display screen detection, and in particular to a method, device and storage medium for background subtraction based on a display screen. Background Art

[0002] In the field of display inspection, it is usually necessary to inspect different types of defects on the display. Defect detection and defect compensation are important production links. When different display structures generate the same defect, its manifestation is also different. In order to improve the quality of display products, De-Mura is still an indispensable link in the panel process.

[0003] The De-Mura compensation process is based on the grayscale values ​​of each pattern screen captured by a high-resolution camera. Before starting the algorithm processing, the program will go through a series of image preprocessing to make the grayscale difference of each screen more realistic to restore the mura form of the panel itself. However, as the structure of the display screen is constantly updated and iterated, in order to adapt to the improved functionality of the display screen, the pixel points of the display screen are usually arranged more closely and more complexly, and a thin-film circuit under the screen layer is added under the pixel layer of the display screen. These new display screens have affected the De-Mura compensation process. Specifically, in order to capture the true mura form of the panel, the high-resolution camera of the De-Mura system must work in a darkroom. When the De-Mura system is operating normally, there will inevitably be many indicator lights of other devices. In addition, our research found that the panel mura form captured by the camera is not only affected by visible light, but also by invisible light. The structure of the new display screen happens to increase the influence of visible and invisible light on the production line.

[0004] Whether it is the LCD Demura process or the Oled Demura process, when observing the De-Mura compensation effect, low grayscale mura is always more difficult to repair or shows more complexity. Low grayscale mura is more difficult to repair, mainly because low grayscale brightness is relatively low and mura is more diverse, and is more easily affected by ambient light. Therefore, in order to capture the mura of low grayscale images more clearly, the camera needs to set a longer exposure time when shooting low grayscale images, but it will also amplify the impact of ambient light on the low grayscale mura state, further increasing the impact of visible and invisible light. Summary of the invention

[0005] The present application discloses a method, device and storage medium for background subtraction based on a display screen, which are used to reduce the influence of ambient light and invisible light.

[0006] In a first aspect, an embodiment of the present application provides a method for background subtraction based on a display screen, comprising: In the scenario of display screen production process, a target display screen is placed in a darkroom environment so that the target display screen displays a preset background picture; a set of exposure times to be segmented is generated according to the maximum grayscale saturation of a sampling camera; images of the target display screen are collected according to the exposure time set to be segmented to generate a set of images to be segmented; the exposure time set to be segmented is segmented into quasi-linear intervals according to the grayscale data in the set of images to be segmented to generate a set of quasi-linear exposure intervals, wherein the set of quasi-linear exposure intervals includes at least two quasi-linear exposure intervals; a background subtraction exposure time set is generated according to the collection exposure time set and the quasi-linear exposure interval set corresponding to the display screen pattern picture set of the current production process; the target display screen is collected using a sampling camera and according to the background subtraction exposure time set to generate a background image set; each display screen pattern picture is displayed using the target display screen; images of the target display screen are collected according to the collection exposure time set to generate a display screen collection image set; and background subtraction processing is performed on images in the display screen collection image set using the background image set.

[0007] Optionally, the step of generating a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera includes: setting a target grayscale upper limit and a target grayscale lower limit according to the maximum grayscale saturation of the acquisition camera; adjusting the exposure time of the acquisition camera so that the grayscale of the image acquired by the acquisition camera is the target grayscale upper limit and the target grayscale lower limit, respectively, and recording the corresponding exposure time upper limit and exposure time lower limit; obtaining the exposure time interval; generating a set of exposure times to be segmented according to the exposure time upper limit, the exposure time lower limit and the exposure time interval.

[0008] Optionally, the target display screen does not include an on-screen circuit area; the exposure time set to be segmented is segmented into quasi-linear intervals according to the grayscale data in the image set to be segmented, and the step of generating a quasi-linear exposure interval set includes: determining the sampling area of ​​each image to be segmented in the image set to be segmented, and determining the grayscale mean of the sampling area; generating a segmentation interval value for each exposure time to be segmented according to the grayscale mean of the sampling area; sequentially performing interval analysis on the segmentation interval values ​​of all exposure times to be segmented; if two adjacent exposure times to be segmented belong to the same quasi-linear exposure time interval, the latter exposure time to be segmented is incorporated into the quasi-linear exposure time interval of the previous exposure time to be segmented; if two adjacent exposure times to be segmented do not belong to the same quasi-linear exposure time interval, the latter exposure time to be segmented is used as the starting point of a new quasi-linear exposure time interval; when the analysis of all exposure time intervals to be segmented is completed, a quasi-linear exposure interval set is generated.

[0009] Optionally, the target display screen includes an in-screen circuit area, a non-in-screen circuit area as an effective area, and an in-screen circuit area as a background area, and the camera center point of the sampling camera is aligned with the in-screen circuit area; in the scenario of the display screen production process, the target display screen is placed in a darkroom environment, and the step of making the target display screen display a preset background picture includes: in the scenario of the display screen production process, the target display screen is placed in a darkroom environment, and a first display screen Pattern picture is obtained, and the first display screen Pattern picture is a picture when the non-circuit area is used as the background area; target grayscale data is generated according to grayscale data of the first display screen Pattern picture and reflectivity data of the in-screen circuit area; arrangement rules are adjusted according to arrangement rules of pixels to be lit in the first display screen Pattern picture and reflectivity data of the in-screen circuit area; a second display screen Pattern picture is generated according to the adjusted arrangement rules and target grayscale data; image features of the circuit area of ​​the target display screen are obtained, and the circuit area image features are integrated into the second display screen Pattern picture through a feature fusion model to generate a background picture; and the target display screen displays the background picture.

[0010] Optionally, the camera center point of the sampling camera is aligned with the circuit area in the screen. The exposure time set to be segmented is segmented into linear intervals according to the grayscale data in the image set to be segmented, and the step of generating the linear exposure interval set includes: determining the sampling area of ​​each image to be segmented in the image set to be segmented according to the center of the camera optical axis, and detecting the grayscale mean of the sampling area, wherein the sampling area is located in the circuit area in the screen; generating a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean of the sampling area; sequentially performing interval analysis on the segmentation interval values ​​of all exposure times to be segmented; if two adjacent exposure times to be segmented belong to the same linear exposure time interval, the latter exposure time to be segmented is incorporated into the linear exposure time interval of the previous exposure time to be segmented; if two adjacent exposure times to be segmented do not belong to the same linear exposure time interval, the latter exposure time to be segmented is used as the starting point of a new linear exposure time interval; when all exposure time intervals to be segmented are analyzed, a linear exposure interval set is generated.

[0011] Optionally, the circuit type of the circuit area in the screen is a uniform circuit area; the step of generating a segmentation interval value for each exposure time to be segmented based on the reflectivity data of the circuit area in the screen and the grayscale mean of the sampling area includes: determining the reflectivity parameter corresponding to the uniform circuit area from the reflectivity data; adjusting the grayscale mean of the sampling area according to the reflectivity parameter to generate a first grayscale mean; generating a segmentation interval value for each exposure time to be segmented based on the first grayscale mean and the grayscale mean of the sampling area.

[0012] Optionally, the circuit type of the circuit area in the screen is a gradient circuit area; the step of generating a segmentation interval value for each exposure time to be segmented based on the reflectivity data of the circuit area in the screen and the grayscale mean of the sampling area includes: determining the maximum reflectivity parameter, minimum reflectivity parameter and circuit ratio corresponding to the gradient circuit area from the reflectivity data; adjusting the grayscale mean of the sampling area according to the maximum reflectivity parameter, the minimum reflectivity parameter and the circuit ratio to generate a second grayscale mean; generating a segmentation interval value for each exposure time to be segmented based on the second grayscale mean and the grayscale mean of the sampling area.

[0013] Optionally, the step of generating a background subtraction exposure time set based on the acquisition exposure time set and the quasi-linear exposure interval set corresponding to the display pattern screen set of the current production process includes: determining the acquisition exposure time corresponding to each display pattern screen in the display pattern screen set of the current production process; classifying each acquisition exposure time into a corresponding quasi-linear exposure interval; screening the acquisition exposure time in each quasi-linear exposure interval, determining a background subtraction exposure time from each quasi-linear exposure interval, and generating a background subtraction exposure time set, wherein each background subtraction exposure time corresponds to at least one display pattern screen.

[0014] In a second aspect, an embodiment of the present application provides a device for background subtraction based on a display screen, comprising: a setting unit, used to place a target display screen in a darkroom environment in a scenario of a display screen production process, so that the target display screen displays a preset background picture; a first generating unit, used to generate a set of exposure times to be segmented according to the maximum grayscale saturation of a sampling camera; a second generating unit, used to perform image acquisition on the target display screen according to the set of exposure times to be segmented, and generate a set of images to be segmented; a third generating unit, used to perform quasi-linear interval segmentation on the set of exposure times to be segmented according to grayscale data in the set of images to be segmented, and generate a set of quasi-linear exposure intervals, wherein the set of quasi-linear exposure intervals includes at least two Quasi-linear exposure interval; a fourth generating unit, used to generate a background subtraction exposure time set according to the acquisition exposure time set corresponding to the display screen Pattern picture set of the current production process and the quasi-linear exposure interval set; a fifth generating unit, used to use the sampling camera and to acquire the target display screen according to the background subtraction exposure time set to generate a background image set; a display unit, used to use the target display screen to display each display screen Pattern picture; a sixth generating unit, used to acquire images of the target display screen according to the acquisition exposure time set to generate a display screen acquisition image set; a subtraction unit, used to use the background image set to perform background subtraction processing on the images in the display screen acquisition image set.

[0015] Optionally, the first generation unit includes: setting a target grayscale upper limit and a target grayscale lower limit according to the maximum grayscale saturation of the acquisition camera; adjusting the exposure time of the acquisition camera so that the grayscale of the image acquired by the acquisition camera is the target grayscale upper limit and the target grayscale lower limit, respectively, and recording the corresponding exposure time upper limit and exposure time lower limit; obtaining the exposure time interval; generating a set of exposure times to be segmented according to the exposure time upper limit, the exposure time lower limit and the exposure time interval.

[0016] Optionally, the target display screen does not include an on-screen circuit area; the third generation unit includes: determining a sampling area of ​​each image to be segmented in the set of images to be segmented, and determining a grayscale mean of the sampling area; generating a segmentation interval value for each exposure time to be segmented according to the grayscale mean of the sampling area; sequentially performing interval analysis on the segmentation interval values ​​of all exposure times to be segmented; if two adjacent exposure times to be segmented belong to the same type of linear exposure time interval, merging the latter exposure time to be segmented into the quasi-linear exposure time interval of the previous exposure time to be segmented; if two adjacent exposure times to be segmented do not belong to the same type of linear exposure time interval, taking the latter exposure time to be segmented as the starting point of a new quasi-linear exposure time interval; when the analysis of all exposure time intervals to be segmented is completed, generating a quasi-linear exposure interval set.

[0017] Optionally, the target display screen includes an in-screen circuit area, a non-in-screen circuit area as an effective area, and an in-screen circuit area as a background area, and the camera center point of the sampling camera is aligned with the in-screen circuit area. The setting unit includes: in the scenario of the display screen production process, placing the target display screen in a darkroom environment, obtaining a first display screen Pattern screen, and the first display screen Pattern screen is a screen when the non-circuit area is used as the background area; generating target grayscale data according to the grayscale data of the first display screen Pattern screen and the reflectivity data of the in-screen circuit area; adjusting the arrangement rules according to the arrangement rules of the pixels to be lit in the first display screen Pattern screen and the reflectivity data of the in-screen circuit area; generating a second display screen Pattern screen according to the adjusted arrangement rules and target grayscale data; obtaining the circuit area image features of the target display screen, integrating the circuit area image features into the second display screen Pattern screen through a feature fusion model to generate a background screen; and causing the target display screen to display the background screen.

[0018] Optionally, the camera center point of the sampling camera is aligned with the circuit area in the screen. The third generation unit includes: a determination module, which determines the sampling area of ​​each image to be segmented in the set of images to be segmented according to the center of the camera optical axis, and detects the grayscale mean of the sampling area, and the sampling area is located in the circuit area in the screen. The first generation module is used to generate a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean of the sampling area. The analysis module is used to sequentially perform interval analysis on the segmentation interval values ​​of all exposure times to be segmented; the merging module is used to merge the latter exposure time to be segmented into the linear exposure time interval of the previous exposure time to be segmented if two adjacent exposure times to be segmented belong to the same linear exposure time interval; the division unit is used to use the latter exposure time to be segmented as the starting point of the new linear exposure time interval if two adjacent exposure times to be segmented do not belong to the same linear exposure time interval; the second generation module is used to generate a linear exposure interval set when all exposure time intervals to be segmented are analyzed.

[0019] Optionally, the circuit type of the circuit area within the screen is a uniform circuit area; the steps of the first generation module include: determining the reflectivity parameters corresponding to the uniform circuit area from the reflectivity data; adjusting the grayscale mean of the sampling area according to the reflectivity parameters to generate a first grayscale mean; generating a segmentation interval value for each exposure time to be segmented according to the first grayscale mean and the grayscale mean of the sampling area.

[0020] Optionally, the circuit type of the circuit area in the screen is a gradient circuit area; the steps of the first generation module include: determining the maximum reflectivity parameter, minimum reflectivity parameter and circuit ratio corresponding to the gradient circuit area from the reflectivity data; adjusting the grayscale mean of the sampling area according to the maximum reflectivity parameter, the minimum reflectivity parameter and the circuit ratio to generate a second grayscale mean; generating a segmentation interval value for each exposure time to be segmented according to the second grayscale mean and the grayscale mean of the sampling area.

[0021] Optionally, the fourth generation unit includes: determining the acquisition exposure time corresponding to each display screen Pattern screen in the display screen Pattern screen set of the current production process; classifying each acquisition exposure time into a corresponding quasi-linear exposure interval; screening the acquisition exposure time in each quasi-linear exposure interval, determining a background subtraction exposure time from each quasi-linear exposure interval, and generating a background subtraction exposure time set, each background subtraction exposure time corresponding to at least one display screen Pattern screen.

[0022] In a third aspect, an embodiment of the present application provides a device for background subtraction based on a display screen, comprising: 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 first aspect and any optional method of the first aspect.

[0023] In a fourth aspect, an embodiment 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.

[0024] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages: The present application first places the target display screen in a darkroom environment under the scenario of the display screen production process, so that the target display screen displays a preset background picture to create a real scene environment. A set of exposure times to be segmented is generated according to the maximum grayscale saturation of the sampling camera. Then, the target display screen is imaged according to the set of exposure times to be segmented to generate a set of images to be segmented. The set of exposure times to be segmented is segmented into quasi-linear intervals according to the grayscale data in the set of images to be segmented to generate a set of quasi-linear exposure intervals, wherein the set of quasi-linear exposure intervals includes at least two quasi-linear exposure intervals. A set of background subtraction exposure times is generated according to the set of acquisition exposure times and the set of quasi-linear exposure intervals corresponding to the set of display screen pattern pictures of the current production process. The target display screen is acquired using a sampling camera and according to the set of background subtraction exposure times to generate a set of background images. Each display screen pattern picture is displayed using the target display screen. The target display screen is imaged according to the set of acquisition exposure times to generate a set of display screen acquisition images. The images in the set of display screen acquisition images are subjected to background subtraction processing using the background image set.

[0025] By pre-processing the target display screen placed in the production line, displaying the background picture and performing image acquisition through the exposure time set to be segmented generated by the maximum grayscale saturation, the background picture is an exposure time sensitive image generated for the current detection project. The grayscale data of the image obtained by acquisition is used to classify the exposure time in the exposure time set to be segmented to generate a quasi-linear exposure interval, which represents the approximate degree data of different exposure times of this type of target display screen when using the current system (acquisition camera and lens) to perform the detection project. Next, it is only necessary to obtain the display screen Pattern picture corresponding to the detection project according to the current production process, and select the corresponding quasi-linear exposure interval according to the exposure time matched by the display screen Pattern picture, so that all the display screen Pattern pictures are matched with the corresponding quasi-linear exposure interval, and then determine the effective and minimum background subtraction exposure time within the range according to the quasi-linear exposure interval and the matched exposure time. The background subtraction exposure time is used to acquire the background image for background subtraction of the detection project, and finally the background image set is used to perform background subtraction processing on the images in the display screen acquisition image set. This solution allows all display pattern images to use the most appropriate background image for background subtraction, which can better reduce the impact of visible and invisible light on the display image in this detection project. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] 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 creative labor.

[0027] Figure 1 A schematic diagram of an embodiment of a method for background subtraction based on a display screen of the present application; Figure 2 A schematic diagram of an embodiment of a method for generating a set of exposure times to be segmented according to the present application; Figure 3 A schematic diagram of an embodiment of a method for generating a set of quasi-linear exposure intervals for the present application; Figure 4 A schematic diagram of an embodiment of a method for generating a background image for the present application; Figure 5 A schematic diagram of an embodiment of a method for generating a set of quasi-linear exposure intervals for the present application; Figure 6 A schematic diagram of an embodiment of a method for segmenting interval values ​​of the present application; Figure 7A schematic diagram of another embodiment of the method for segmenting interval values ​​of the present application; Figure 8 A schematic diagram of an embodiment of a method for generating a background subtraction exposure time set for the present application; Fig. 9 A schematic diagram of an embodiment of a device for background subtraction based on a display screen of the present application; Fig.10 It is a schematic diagram of another embodiment of the device for background subtraction based on a display screen of the present application. DETAILED DESCRIPTION

[0028] 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.

[0029] 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.

[0030] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] In the prior art, the De-Mura compensation process is based on the grayscale values ​​of each pattern screen captured by a high-resolution camera. Before starting the algorithm processing, the program will undergo a series of image preprocessing to allow the grayscale difference of each screen to more realistically restore the mura form of the panel itself. However, as the structure of the display screen is continuously updated and iterated, in order to adapt to the improved functionality of the display screen, the pixel points of the display screen are usually arranged more closely and more complexly, and a thin-film circuit under the screen is added under the pixel layer of the display screen. These new display screens have affected the De-Mura compensation process. Specifically, in order to capture the true mura form of the panel, the high-resolution camera of the De-Mura system must work in a darkroom. When the De-Mura system is operating normally, there will inevitably be many indicator lights of other devices. In addition, our research has found that the panel mura form captured by the camera is not only affected by visible light, but also by invisible light, and the structure of the new display screen happens to increase the impact of visible and invisible light on the production line. Whether it is the LcdDemura process or the Oled Demura process, when observing the De-Mura compensation effect, low grayscale mura is always more difficult to repair or shows more complexity. Low grayscale mura is more difficult to repair, mainly because low grayscale brightness is relatively low and mura is more diverse, and is more easily affected by ambient light. Therefore, in order to capture the mura of low grayscale images more clearly, the camera needs to set a longer exposure time when shooting low grayscale images, but it will also amplify the impact of ambient light on the low grayscale mura state, further increasing the impact of visible and invisible light.

[0035] Based on this, the present application discloses a method, device and storage medium for background subtraction based on a display screen, which are used to reduce the influence of ambient light and invisible light.

[0036] 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.

[0037] 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.

[0038] See also Figure 1 The present application provides an embodiment of a method for background subtraction based on a display screen, comprising: 101. In a display screen production process scenario, a target display screen is placed in a darkroom environment so that the target display screen displays a preset background image.

[0039] The De-Mura system of the embodiment of the present application is composed of an industrial camera (i.e., a sampling camera, which includes dark field correction and flat field correction functions and is turned on during calibration and measurement), a telephoto industrial lens, a human-computer interaction interface, a computer, and a display controller. The entire process of background subtraction is performed in a darkroom. The background subtraction of the embodiment of the present application is mainly used in inspection projects on production lines.

[0040] Place the De-Mura system and the target display screen (with Pgamma adjusted) in a darkroom working environment, keep the sampling camera gain, sampling camera working distance, lens aperture and lens focal length, and keep the above parameters unchanged in the working state of the corresponding display screen model (the same type as the target display screen) during normal Demura. The computer sends power-on and image cutting instructions to the display screen controller to light up the target display screen so that the display screen displays the W0 screen. The W0 screen is the background screen corresponding to the current detection item of the target display screen. The detection items in the embodiments of the present application are defect detection items, which are divided into screen defect detection and display defect detection in the detection items of the display screen. Screen defect detection is mainly for physical trauma caused to the screen during the production process, such as: edge collapse, scratches, etc. Display defects are mainly for display damage to pixels, such as: highlight, poor display, etc. The embodiments of the present application are mainly for display defects, and the background screen is generated according to the specific type that needs to be detected, and it is necessary to design it in combination with the scene and Pattern screen of the detection item.

[0041] In this step, after placing the target display screen in a darkroom environment, a background image needs to be generated before displaying the display defect of the current inspection item.

[0042] The background image can be generated based on the pattern image. In the De-Mura process, for a display screen without a complex layer, the structures of all parts of the entire display screen area are similar, and the difficulty of detecting display screen defects is relatively low. By determining the pattern image with the same exposure time in the same interval, the background image with the same exposure time in the area can be used as the background image.

[0043] 102. Generate a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera.

[0044] The terminal generates a set of exposure times to be divided according to the maximum grayscale saturation of the sampling camera, that is, it is necessary to generate a valid exposure time interval (upper and lower limits of exposure time) according to the maximum grayscale saturation of the sampling camera, and then generate multiple exposure times to be divided that can be used as the interval range according to the preset step size. The exposure time to be divided here may be the upper and lower limits of the subsequent linear exposure interval, so it is necessary to limit the range by the maximum grayscale saturation of the sampling camera, and then design the step size to generate the exposure time to be divided.

[0045] The step size can be set according to the content of the detection project, in order to ensure that the intervals generated subsequently are more accurate, so that the background image can better perform background subtraction on the display screen captured image.

[0046] Specifically, when the effective exposure time interval is determined to be [t0, th], t0 is set as the starting exposure time of the sampling camera, th is set as the ending exposure time of the camera, and the linear exposure interval values ​​T={t0, t0+∆t, t0+2*∆t…th} (the set of exposure times to be divided) of the background image are determined at time intervals of the calculated step length ∆t (in milliseconds ms).

[0047] 103. Capture images of the target display screen according to the exposure time set to be segmented, and generate a set of images to be segmented.

[0048] The terminal first controls the exposure time of the sampling camera so that it can collect images of the target display screen according to the exposure time to be segmented in the exposure time set to be segmented. At this time, the target display screen displays the background image, and the grayscale data are different because of the different exposure times.

[0049] 104. Perform quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented, and generate a quasi-linear exposure interval set, wherein the quasi-linear exposure interval set includes at least two quasi-linear exposure intervals.

[0050] The terminal performs quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented, that is, performs joint analysis based on the grayscale data of each image to be segmented and its corresponding exposure time to be segmented, determines the exposure times to be segmented belonging to the same interval, and then integrates the exposure times to be segmented in the same interval to generate a quasi-linear exposure interval set. The specific method of generating the image set to be segmented is described in detail in the subsequent embodiments.

[0051] The core purpose of this step is to generate a comparison coefficient based on the grayscale data of the image to be segmented and the corresponding exposure time to be segmented. Through this comparison coefficient, the influence of different exposure times on the display capability of the image on the target display screen can be better determined, and then the images to be segmented with similar display capabilities are integrated, so that the corresponding exposure times are also integrated, and then a set of quasi-linear exposure intervals is generated. However, the quasi-linear exposure intervals of different types of display screens are not the same. This is because the density and arrangement of pixels of different display screens are different, and the hierarchical structure of some display screens makes the background area different from the regular display area, so that different exposure times have different effects on the display screen.

[0052] 105. Generate a background subtraction exposure time set according to the acquisition exposure time set and the quasi-linear exposure interval set corresponding to the display screen Pattern screen set of the current production process.

[0053] After obtaining the set of quasi-linear exposure intervals, the exposure time corresponding to the set of display screen pattern images on the production line can be classified into intervals. Specifically, it is necessary to match each display screen pattern image and its corresponding exposure time with all quasi-linear exposure intervals, and find the quasi-linear exposure interval to which each display screen pattern image and its corresponding exposure time belong. This makes it possible for a quasi-linear exposure interval to be associated with and matched to multiple display screen pattern images and their corresponding exposure times, while some quasi-linear exposure intervals may not have display screen pattern images and their corresponding exposure times.

[0054] The subsequent terminal collects all the associated quasi-linear exposure intervals, selects an exposure time from each of these associated quasi-linear exposure intervals as the background subtraction exposure time according to a preset rule, and generates a background subtraction exposure time set. The specific method of selecting the background subtraction exposure time is described in detail in the subsequent embodiments.

[0055] 106. Use a sampling camera to collect images of a target display screen according to a background subtraction exposure time set to generate a background image set.

[0056] 107. Use the target display screen to display each display pattern screen.

[0057] 108. Capture images of the target display screen according to the collection exposure time set to generate a display screen collection image set.

[0058] 109. Use the background image set to perform background subtraction processing on the images in the display screen captured image set.

[0059] In step 106 to step 109, the terminal sets the sampling camera according to the background subtraction exposure time in the background subtraction exposure time set, and then shoots the target display screen to generate a number of background images, the number of background images being the same as the number of background subtraction exposure times.

[0060] After the segmented background subtraction scheme of the De-Mura scheme is determined, each time the De-Mura process is started, the terminal will first display the W0 screen (background screen) on the target display screen, and take a background image at each exposure time of the background subtraction exposure time T={T1, T2...Tm} to generate m background images. Then the program will continue to take pictures of each display screen Pattern screen in the order of W={w1,w2,w3......wn} and its acquisition exposure time set. After each display screen Pattern screen is taken, the corresponding display screen acquisition image is obtained. The terminal will perform background subtraction processing based on the corresponding background image. Each display screen acquisition image is associated with a background image through the acquisition exposure time.

[0061] In an embodiment of the present application, first, in the scenario of the display screen production process, the target display screen is placed in a darkroom environment, so that the target display screen displays a preset background picture to create a real scene environment. A set of exposure times to be segmented is generated according to the maximum grayscale saturation of the sampling camera. Then, the target display screen is imaged according to the exposure time set to be segmented to generate a set of images to be segmented. The exposure time set to be segmented is segmented into quasi-linear intervals according to the grayscale data in the set of images to be segmented to generate a set of quasi-linear exposure intervals, and the set of quasi-linear exposure intervals includes at least two quasi-linear exposure intervals. A set of background subtraction exposure times is generated according to the set of acquisition exposure times and the set of quasi-linear exposure intervals corresponding to the set of display screen pattern pictures of the current production process. The target display screen is collected using the sampling camera and according to the set of background subtraction exposure times to generate a set of background images. Each display screen pattern picture is displayed using the target display screen. The target display screen is imaged according to the acquisition exposure time set to generate a set of display screen acquisition images. The images in the set of display screen acquisition images are subjected to background subtraction processing using the background image set.

[0062] By pre-processing the target display screen placed in the production line, displaying the background picture and performing image acquisition through the exposure time set to be segmented generated by the maximum grayscale saturation, the background picture is an exposure time sensitive image generated for the current detection project. The grayscale data of the image obtained by acquisition is used to classify the exposure time in the exposure time set to be segmented to generate a quasi-linear exposure interval, which represents the approximate degree data of different exposure times of this type of target display screen when using the current system (acquisition camera and lens) to perform the detection project. Next, it is only necessary to obtain the display screen Pattern picture corresponding to the detection project according to the current production process, and select the corresponding quasi-linear exposure interval according to the exposure time matched by the display screen Pattern picture, so that all the display screen Pattern pictures are matched with the corresponding quasi-linear exposure interval, and then determine the effective and minimum background subtraction exposure time within the range according to the quasi-linear exposure interval and the matched exposure time. The background subtraction exposure time is used to acquire the background image for background subtraction of the detection project, and finally the background image set is used to perform background subtraction processing on the images in the display screen acquisition image set. This solution allows all display pattern images to use the most appropriate background image for background subtraction, which can better reduce the impact of visible and invisible light on the display image in this detection project.

[0063] See also Figure 2 The present application provides an embodiment of a method for generating a set of exposure times to be segmented, comprising: 201. Set a target grayscale upper limit and a target grayscale lower limit according to the maximum grayscale saturation of the acquisition camera.

[0064] 202. Adjust the exposure time of the acquisition camera so that the grayscale of the image acquired by the acquisition camera is respectively the target grayscale upper limit and the target grayscale lower limit, and record the corresponding exposure time upper limit and exposure time lower limit.

[0065] 203. Obtain exposure time interval.

[0066] 204. Generate a set of exposure times to be segmented according to the exposure time upper limit, the exposure time lower limit, and the exposure time interval.

[0067] In the embodiment of the present application, the target grayscale upper limit and the target grayscale lower limit are set according to the maximum grayscale saturation of the acquisition camera.

[0068] The following is an example: First, turn on the sampling camera to take a picture of the target display screen. The target grayscale of the sampling camera needs to be set to 1% and 90% of the maximum grayscale saturation (such as 255 for 8-bit images) (this value can be adjusted according to the test item). Use the camera's automatic exposure function or manually adjust the camera exposure to make the image grayscale collected by the camera close to the target grayscale set above. The camera exposure time corresponding to these two target grayscales is recorded as t0 and th, respectively, in ms. The terminal sets t0 as the camera's starting exposure time and th as the camera's ending exposure time, and starts to search for each linear exposure interval point T={t0,t0+∆t,t0+2*∆t……th} of the background image with an exposure time interval (step length) ∆t as the time interval.

[0069] The method of obtaining the exposure time interval is mainly determined by the target defect of the inspection project. One specific method can be as follows: first determine the range of the target defect that is most affected by the exposure time, which is called the defect exposure time range. This data can be set based on human experience, and then set the exposure time detection so that the exposure time point in the linear exposure interval point T={t0,t0+∆t,t0+2*∆t……th} is as close as possible to the upper and lower limits of the defect exposure time range.

[0070] See also Figure 3 The present application provides an embodiment of a method for generating a set of quasi-linear exposure intervals, comprising: 301. Determine a sampling area of ​​each image to be segmented in the set of images to be segmented, and determine a grayscale mean value of the sampling area.

[0071] 302. Generate a segmentation interval value for each exposure time to be segmented according to the grayscale mean value of the sampling area.

[0072] 303. Performing interval analysis on all the segmentation interval values ​​of the exposure time to be segmented in sequence; 304. If two adjacent exposure times to be divided belong to the same linear exposure time interval, the latter exposure time to be divided is merged into the linear exposure time interval of the former exposure time to be divided; 305. If two adjacent exposure times to be divided do not belong to the same linear exposure time interval, the latter exposure time to be divided is used as the starting point of a new linear exposure time interval; 306. When all exposure time intervals to be segmented are analyzed, a set of quasi-linear exposure intervals is generated.

[0073] The terminal first shoots the target display screen showing the W0 image with the exposure time t0 to be segmented, and then takes the central area of ​​the image to be segmented as the sampling area. Specifically, a rectangular area with a field of view of 2 relative to the center of the camera optical axis is taken to calculate the grayscale mean Gt0 Then, the W0 image is captured with the exposure time t0+∆t to be segmented, and the grayscale mean G of the corresponding area of ​​the image to be segmented is calculated. t0+∆t , take the W0 image with the exposure time t0+2*∆t to be segmented, and calculate the grayscale mean G of the corresponding area of ​​the image to be segmented t0+2*∆t , and so on, each exposure time to be segmented generates an image to be segmented, and a grayscale mean is calculated.

[0074] Next, the terminal generates a segmentation interval value for each exposure time to be segmented according to the grayscale mean value of the sampling area, compares and analyzes the segmentation interval values ​​of two adjacent exposure times to be segmented, and generates a quasi-linear exposure interval set.

[0075] When the target display screen does not contain the circuit area within the screen and is a uniformly distributed flat screen, the segmentation interval value k is recorded as t0 =G t0 / t0, record the segmentation interval value k t0+∆t =G t0+∆t / (t0+∆t), and so on, to generate multiple segmentation interval values. Then analyze, if abs(k t0 -k t0+∆t )<∆k, it means that t0 and t0+∆t are in a linear exposure interval. First, mark the interval as [t0,t0+∆t] and abs() is the absolute value operation.

[0076] Then for t0+2*∆t, record k t0+2*∆t =G t0+2*∆t / (t0+2*∆t), then determine abs(k t0+∆t -k t0+2*∆t )<∆k, if the condition is still met, update the current linear exposure area to [t0, t0+2*∆t], and so on, until abs(k t0+(N-1)*∆t -k t0+N*∆t )<∆k condition is satisfied, and abs(k t0+N*∆t -k t0+(N+1)*∆t )<∆k is not satisfied, the optimization of the current linear interval [t0, t0+N*∆t] is completed.

[0077] The new linear exposure interval starts from (t0+N*∆t,t0+(N+1)*∆t] (note that the starting interval is the left open and right closed interval), and the slope difference of two adjacent exposure time intervals is determined in sequence with ∆t as the time interval, until abs(k t0+(M-1)*∆t -k t0+M*∆t )<∆k condition is satisfied, and abs(k t0+M*∆t -k t0+(M+1)*∆t)When ∆k is not satisfied, the optimization of the current linear interval (t0 + N*∆t, t0 + M*∆t) is completed; The new linear exposure interval starts from (t0 + M*∆t, t0 + (M + 1)*∆t]. Referring to the above steps, until [t0, th] is divided into several quasi-linear exposure intervals [t0, t0 + N*∆t], (t0 + N*∆t, t0 + M*∆t]......(t0 + X*∆t, th] (N < M <...... < X). So far, the division of the quasi-linear interval of the background picture in its exposure area [t0, th] is completed.

[0078] The above quasi-linear interval division method is the most efficient and accurate for a flat screen without complex layers (such as internal circuit layers).

[0079] Please refer to Figure 4 , this application provides an embodiment of a method for generating a background picture, including: 401. In the scenario of the display screen production process, place the target display screen in a darkroom environment to obtain the first display screen Pattern picture, where the first display screen Pattern picture is the picture when the non-circuit area is used as the background area.

[0080] 402. Generate target gray-scale data according to the gray-scale data of the first display screen Pattern picture and the reflectivity data of the in-screen circuit area.

[0081] 403. Adjust the arrangement rule according to the arrangement rule of the pixel points to be lit in the first display screen Pattern picture and the reflectivity data of the in-screen circuit area.

[0082] 404. Generate the second display screen Pattern picture according to the adjusted arrangement rule and the target gray-scale data.

[0083] 405. Obtain the image features of the circuit area of the target display screen, and integrate the circuit area image features into the second display screen Pattern picture through the feature fusion model to generate the background picture.

[0084] 406. Make the target display screen display the background picture.

[0085] In the embodiment of the present application, the target display screen is a screen body with a thin-film circuit (in-screen circuit) provided at the pixel layer. For display screens provided with internal circuits, it is more difficult to detect defects. Such display screens usually provide thin-film circuits at the pixel layer, and are usually not provided at the center of the display screen, but are usually distributed around or in a corner of the display screen. Since the thin-film circuit is mainly made of metal and has a high degree of integration, it has a certain reflective ability, which brings difficulties to the detection of display defects, because the detection of display defects usually requires the lighting of the display screen Pattern at the pixel layer, and the thin-film circuit may reflect the light of the pixel points, thereby affecting the detection of surrounding defects. In the embodiment of the present application, a display screen with an internal thin-film circuit is provided, and the circuit area is used as the background area in the detection project (i.e., defect detection in the non-circuit area). For this display screen and display screen Pattern, the generation of the background screen needs to be designed, and the design scheme is as follows: First, the terminal uses the grayscale data of the first display screen Pattern when the non-circuit area is used as the background area, and then adjusts the grayscale data according to the reflectivity of the thin-film circuit in the screen to generate the target grayscale data when the circuit area is used as the background area. After obtaining the target grayscale data, the arrangement rules of the pixels that need to be lit in the first display screen Pattern are obtained. It is necessary to determine whether the reflectivity of the thin-film circuit in the screen is greater than the preset value. If it is greater than the preset value, the arrangement rules of the original first display screen Pattern need to be adjusted to be sparser. For example, if the first display screen Pattern is fully lit, the new arrangement method is to light up alternate lines or select 2 diagonal pixels from 3*3 pixels to not light up.

[0086] Next, the second display screen Pattern image is generated according to the new arrangement rule and the target grayscale data. Then, it is determined to obtain the captured image of the thin-film circuit on the target display screen, extract the circuit area image features from the captured image, and then fuse the circuit area image features into the first display screen Pattern image to generate at least one background image. Specifically, the first display screen Pattern image and the circuit area image features after grayscale adjustment are feature-fused using an image fusion model. The background image will incorporate a small part of the thin-film circuit features, that is, the grayscale of the pixel points will be adjusted again. This type of background image can be more in line with the circuit area as the background area. The background image is photographed, and the collected image reacts more closely to the change in exposure time to the target display screen (there is a circuit area). The increase in the quasi-linear exposure interval is highly targeted. For the circuit area as the background area, the background subtraction effect of the subsequent defect detection captured image (the captured image of the actual detection process, the detection item is to detect non-circuit area defects) is further improved.

[0087] See also Figure 5The present application provides an embodiment of a method for generating a set of quasi-linear exposure intervals, comprising: 501. Determine a sampling area of ​​each image to be segmented in the set of images to be segmented according to the center of the camera optical axis, and detect the grayscale mean of the sampling area, where the sampling area is located in the circuit area within the screen.

[0088] 502. Generate a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean value of the sampling area.

[0089] 503. Performing interval analysis on all the segmentation interval values ​​of the exposure time to be segmented in sequence; 504. If two adjacent exposure times to be divided belong to the same linear exposure time interval, the latter exposure time to be divided is merged into the linear exposure time interval of the former exposure time to be divided; 505. If two adjacent exposure times to be divided do not belong to the same linear exposure time interval, the latter exposure time to be divided is used as the starting point of a new linear exposure time interval; 506. When all exposure time intervals to be segmented are analyzed, a set of quasi-linear exposure intervals is generated.

[0090] In the embodiment of the present application, since the detection item is a non-circuit area, when using the sampling camera for sampling, it is necessary to adjust the optical axis center of the sampling camera, and align the camera center point of the sampling camera with the circuit area on the screen. After the image to be segmented is captured, the terminal needs to determine the sampling area of ​​each image to be segmented in the image set to be segmented according to the center of the camera optical axis, and detect the grayscale mean of the sampling area. The sampling area is located in the circuit area on the screen. The sampling area in the embodiment of the present application selects the circuit area on the screen, and then detects the grayscale mean of the sampling area. The segmentation interval value is generated for each exposure time to be segmented according to the reflectivity data of the circuit area on the screen and the grayscale mean of the sampling area. Because the circuit area on the screen is the background area, its reflectivity can cause differences in display, so it is necessary to use the reflectivity data, the grayscale mean of the sampling area and the new calculation method to generate the segmentation interval value, and then compare and analyze the segmentation interval values ​​of the two adjacent exposure times to be segmented to generate a quasi-linear exposure interval set, so that the circuit area on the screen as the background can be better.

[0091] The specific method of generating the segmentation interval value needs to be discussed separately according to the differences in the circuit areas within the screen.

[0092] See also Figure 6 The present application provides an embodiment of a method for segmenting interval values, comprising: 601. Determine a reflectivity parameter corresponding to a uniform circuit area from reflectivity data.

[0093] 602. Adjust the grayscale mean of the sampling area according to the reflectivity parameter to generate a first grayscale mean.

[0094] 603. Generate a segmentation interval value for each exposure time to be segmented according to the first grayscale mean and the grayscale mean of the sampling area.

[0095] In the embodiment of the present application, when the circuit type of the circuit area in the screen is a uniform circuit area, the reflectivity is uniform, and the terminal determines the reflectivity parameter corresponding to the uniform circuit area from the reflectivity data. , the terminal adjusts the grayscale mean of the sampling area according to the reflectivity parameter to generate the first grayscale mean, the formula is as follows:

[0096] in, is the grayscale mean of the sampling area, is the first grayscale mean.

[0097] Next, the terminal calculates the grayscale mean according to the first , grayscale mean of sampling area Generate a segmentation interval value for each exposure time to be segmented. The formula is as follows:

[0098] is the segmentation interval value, For the corresponding exposure time to be segmented, the grayscale of the circuit area in the screen is compared with the grayscale of the non-circuit area (conventional area) by the square difference method, and then divided by the corresponding exposure time to be segmented. This method is more suitable for this type of display screen containing an in-screen circuit, improves the segmentation of the background area of ​​this type of display screen, and can better classify the exposure time to be segmented. The subsequent analysis method is similar to steps 301-303 in the aforementioned embodiment, and will not be repeated here.

[0099] See also Figure 7 The present application provides an embodiment of a method for generating a segmentation interval value, comprising: 701. Determine a maximum reflectivity parameter, a minimum reflectivity parameter, and a circuit ratio corresponding to a gradient circuit area from reflectivity data.

[0100] 702. Adjust the grayscale mean of the sampling area according to the maximum reflectivity parameter, the minimum reflectivity parameter and the line ratio to generate a second grayscale mean.

[0101] 703. Generate a segmentation interval value for each exposure time to be segmented according to the second grayscale mean and the sampling area grayscale mean.

[0102] In this embodiment, when the circuit type of the circuit area in the screen is a uniform circuit area, this type of gradient circuit is usually a dense metal line. This type of metal line has a fence-like arrangement and a decreasing thickness. It belongs to the circuit module for transmitting signals and needs to calculate the maximum reflectivity parameter. , minimum reflectivity parameter , and then because there are regular gaps between the lines, the line ratio of the circuit area can be obtained , and then obtain the grayscale mean of the non-circuit area. Next, the terminal calculates the grayscale mean of the sampling area according to the maximum reflectivity parameter, the minimum reflectivity parameter and the line ratio. Adjust to generate the second grayscale mean , the formula is as follows:

[0103] The subsequent terminal generates a segmentation interval value for each exposure time to be segmented according to the second grayscale mean and the grayscale mean of the sampling area. The specific details are similar to step 603 and will not be repeated here.

[0104] See also Figure 8 The present application provides an embodiment of a method for generating a background subtraction exposure time set, comprising: 801. Determine the acquisition exposure time corresponding to each display screen Pattern picture in the display screen Pattern picture set of the current production process.

[0105] 802. Classify each collected exposure time into a corresponding quasi-linear exposure interval.

[0106] 803. Screen the collected exposure time in each quasi-linear exposure interval, determine a background subtraction exposure time from each quasi-linear exposure interval, and generate a background subtraction exposure time set, each background subtraction exposure time corresponding to at least one display screen Pattern image.

[0107] In the embodiment of the present application, the segmented background scheme of the De-mura process is mainly based on the display screen Pattern screen W={w1,w2,w3...w n}(The De-mura process generally collects images in the order of high grayscale to low grayscale. After the user sets all the photo screens, the program will automatically sort them in the order of high to low grayscale), and each display pattern screen has a corresponding exposure time t={t1,t2,t3...t n} (Generally speaking, t1≦t2≦t3...≦tn, the program will ensure that this condition is met), find the display pattern images belonging to the same linear exposure interval and determine the exposure time of the background images of these images.

[0108] The specific steps of the segmented background scheme of the De-mura process are as follows: 1. First, starting from the w1 picture, use the exposure time t1 corresponding to the w1 picture to find the quasi-linear exposure area D1 where t1 is located, where D1ϵ{[t0,t0+N*∆t],(t0+N*∆t,t0+M*∆t),...(t0+X*∆t,th]}, specifically belongs to one of these quasi-linear exposure intervals. Each quasi-linear exposure interval generates a background image according to the exposure time point. At this time, the first background image exposure time point T1 is recorded, and its corresponding initial set is {w1}.

[0109] 2. Then determine whether the exposure time t2 of the w2 image is in the D1 area, that is, determine whether t2ϵD1 holds. If so, the Pattern image set corresponding to T1 is expanded to {w1,w2}, and so on, until all the display pattern images belonging to the D1 linear exposure interval are found, recorded as T1->{w1,w2...w i}; 3. Then use screen W i+1 The exposure time ti+1 of the image is in the linear exposure interval D2 (D2ϵ{[t0,t0+N*∆t],(t0+N*∆t,t0+M*∆t),...(t0+X*∆t,th]}), record the exposure time point T2 of the second background image, and find all the display pattern images with exposure time in area D2, recorded as T2->{w i+1 ,w i+2 ... j}; 4. Continue in this way until a set of exposure time sets T = {T1, T2...Tm} (generally m ≦ 3) is found, that is, the background subtraction exposure time set, so that all display pattern images in the W set (display pattern images W = {w1, w2, w3...wn}) have corresponding quasi-linear exposure intervals D = {D1, D2...D m}(D is a subset of the set {[t0,t0+N*∆t],(t0+N*∆t,t0+M*∆t...(t0+X*∆t,th]}), and W={w1,w2...w i}∪{w i+1 ,w i+ 2...w j}∪......∪{w h ,w h+1 ... n}.

[0110] 5. For the background subtraction exposure time set T={T1, T2...Tm}, traverse each element Tx in T, and take the minimum exposure value of the quasi-linear exposure interval (Ta, Tb) corresponding to Tx. Ta and Tb are the upper and lower limits of the interval. If the left endpoint of the current interval is closed, directly take the value of the left endpoint Ta. If the left endpoint is not closed, take Ta+∆t as the final value of Tx. After that, the program will match each display pattern screen with Tx one by one.

[0111] See also Fig. 9 The present application provides an embodiment of a device for background subtraction based on a display screen, comprising: The setting unit 901 is used to place the target display screen in a darkroom environment in the scenario of the display screen production process, so that the target display screen displays a preset background picture.

[0112] Optionally, the target display screen includes an in-screen circuit area, the non-in-screen circuit area is used as the effective area, the in-screen circuit area is used as the background area, and the camera center point of the sampling camera is aligned with the in-screen circuit area.

[0113] The setting unit 901 includes: in the scenario of the display screen production process, placing the target display screen in a darkroom environment, obtaining a first display screen Pattern screen, and the first display screen Pattern screen is a screen when the non-circuit area is used as the background area; generating target grayscale data according to the grayscale data of the first display screen Pattern screen and the reflectivity data of the circuit area in the screen; adjusting the arrangement rule according to the arrangement rule of the pixels to be lit in the first display screen Pattern screen and the reflectivity data of the circuit area in the screen; generating a second display screen Pattern screen according to the adjusted arrangement rule and the target grayscale data; obtaining the circuit area image features of the target display screen, integrating the circuit area image features into the second display screen Pattern screen through a feature fusion model to generate a background screen; and causing the target display screen to display the background screen.

[0114] The first generating unit 902 is configured to generate a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera.

[0115] Optionally, the first generating unit 902 includes: Set the target grayscale upper limit and target grayscale lower limit according to the maximum grayscale saturation of the acquisition camera.

[0116] Adjust the exposure time of the acquisition camera so that the grayscale of the image acquired by the acquisition camera is the target grayscale upper limit and the target grayscale lower limit, and record the corresponding exposure time upper limit and exposure time lower limit.

[0117] Get the exposure time interval.

[0118] A set of exposure times to be segmented is generated according to an upper limit of the exposure time, a lower limit of the exposure time, and an exposure time interval.

[0119] The second generating unit 903 is used to collect images of the target display screen according to the exposure time set to be segmented, and generate a set of images to be segmented.

[0120] The third generating unit 904 is used to perform quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented, and generate a quasi-linear exposure interval set, wherein the quasi-linear exposure interval set includes at least two quasi-linear exposure intervals.

[0121] Optionally, the target display screen does not include an in-screen circuit area.

[0122] The third generating unit 904 includes: Determine the sampling area of ​​each image to be segmented in the set of images to be segmented, and determine the grayscale mean of the sampling area.

[0123] Generate a segmentation interval value for each exposure time to be segmented according to the grayscale mean value of the sampling area.

[0124] Sequentially perform interval analysis on all the segmentation interval values ​​of the exposure time to be segmented; If two adjacent exposure times to be divided belong to the same linear exposure time interval, the latter exposure time to be divided is merged into the linear exposure time interval of the former exposure time to be divided; If two adjacent exposure times to be divided do not belong to the same linear exposure time interval, the latter exposure time to be divided is used as the starting point of a new linear exposure time interval; When all exposure time intervals to be segmented are analyzed, a set of quasi-linear exposure intervals is generated.

[0125] Optionally, a camera center point of the sampling camera is aligned with an on-screen circuit area.

[0126] The third generation unit 904 includes: determining the sampling area of ​​each image to be segmented in the image set to be segmented according to the center of the camera optical axis, and detecting the grayscale mean of the sampling area, where the sampling area is located in the circuit area within the screen. Generate a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area within the screen and the grayscale mean of the sampling area. Perform interval analysis on all the segmentation interval values ​​of the exposure time to be segmented in sequence; if two adjacent exposure times to be segmented belong to the same linear exposure time interval, the latter exposure time to be segmented is incorporated into the linear exposure time interval of the previous exposure time to be segmented; if two adjacent exposure times to be segmented do not belong to the same linear exposure time interval, the latter exposure time to be segmented is used as the starting point of the new linear exposure time interval; when all exposure time intervals to be segmented are analyzed, a linear exposure interval set is generated.

[0127] Optionally, the target display screen includes an in-screen circuit area, the in-screen circuit area serves as a valid area, and the non-in-screen circuit area serves as a background area, and the camera center point of the sampling camera is aligned with the in-screen circuit area.

[0128] The third generating unit 904 includes: The determination module 9041 is used to determine the sampling area of ​​each image to be segmented in the set of images to be segmented according to the center of the camera optical axis, and detect the grayscale mean of the sampling area, and the sampling area is located in the circuit area within the screen.

[0129] The first generating module 9042 is used to generate a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean of the sampling area.

[0130] Optionally, the circuit type of the circuit area within the screen is a uniform circuit area.

[0131] The steps of the first generation module 9042 include: Reflectivity parameters corresponding to the uniform circuit area are determined from the reflectivity data.

[0132] The grayscale mean value of the sampling area is adjusted according to the reflectivity parameter to generate a first grayscale mean value.

[0133] A segmentation interval value is generated for each exposure time to be segmented according to the first grayscale mean value and the grayscale mean value of the sampling area.

[0134] Optionally, the circuit type of the circuit area within the screen is a gradient circuit area.

[0135] The steps of the first generation module 9042 include: The maximum reflectivity parameter, the minimum reflectivity parameter and the line ratio corresponding to the gradient circuit area are determined from the reflectivity data.

[0136] The grayscale mean value of the sampling area is adjusted according to the maximum reflectivity parameter, the minimum reflectivity parameter and the line ratio to generate a second grayscale mean value.

[0137] A segmentation interval value is generated for each exposure time to be segmented according to the second grayscale mean and the grayscale mean of the sampling area.

[0138] An analysis module 9043 is used to sequentially perform interval analysis on all segmentation interval values ​​of the exposure time to be segmented; A merging module 9044 is configured to merge the latter exposure time to be divided into the quasi-linear exposure time interval of the former exposure time to be divided if the two adjacent exposure times to be divided belong to the same quasi-linear exposure time interval; A division unit 9045 is used for taking the latter exposure time to be divided as the starting point of a new quasi-linear exposure time interval if two adjacent exposure times to be divided do not belong to the same quasi-linear exposure time interval; The second generating module 9046 is used to generate a set of quasi-linear exposure intervals when the analysis of all exposure time intervals to be divided is completed.

[0139] The fourth generating unit 905 is used to generate a background subtraction exposure time set according to the acquisition exposure time set and the quasi-linear exposure interval set corresponding to the display screen Pattern picture set of the current production process.

[0140] Optionally, the fourth generating unit 905 includes: Determine the acquisition exposure time corresponding to each display pattern screen in the display pattern screen set of the current production process.

[0141] Classify each collected exposure time into the corresponding quasi-linear exposure interval.

[0142] The collected exposure time in each quasi-linear exposure interval is screened, a background subtraction exposure time is determined from each quasi-linear exposure interval, and a background subtraction exposure time set is generated, and each background subtraction exposure time corresponds to at least one display screen Pattern screen.

[0143] The fifth generating unit 906 is configured to use the sampling camera to collect data from the target display screen according to the background subtraction exposure time set to generate a background image set.

[0144] The display unit 907 is used to display each display screen Pattern picture using the target display screen.

[0145] The sixth generating unit 908 is configured to perform image acquisition on the target display screen according to the acquisition exposure time set to generate a display screen acquisition image set.

[0146] The subtraction unit 909 is used to perform background subtraction processing on the images in the display screen captured image set using the background image set.

[0147] See also Fig.10 The present application provides a device for background subtraction based on a display screen, comprising: Processor 1001 , memory 1002 , input / output unit 1003 , and bus 1004 .

[0148] The processor 1001 is connected to the memory 1002 , the input and output unit 1003 , and the bus 1004 .

[0149] The memory 1002 stores a program, and the processor 1001 calls the program to execute the following steps: Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 and Figure 8 The method in .

[0150] 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 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 and Figure 8 The method in .

[0151] 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.

[0152] 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.

[0153] The units described as separate components may or may not be physically separated, and the components displayed 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 embodiments of the present application.

[0154] 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.

[0155] 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 background subtraction based on a display screen, characterized in that: include: In the scenario of the display screen production process, placing the target display screen in a darkroom environment so that the target display screen displays a preset background image; Generate a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera; Capturing images of the target display screen according to the exposure time set to be segmented, to generate a set of images to be segmented; Performing quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented, generating a quasi-linear exposure interval set, wherein the quasi-linear exposure interval set includes at least two quasi-linear exposure intervals; Generate a background subtraction exposure time set according to the acquisition exposure time set corresponding to the display screen Pattern screen set of the current production process and the quasi-linear exposure interval set; Using the sampling camera and according to the background subtraction exposure time set, the target display screen is captured to generate a background image set; Using the target display screen to display each display screen Pattern screen; Performing image acquisition on the target display screen according to the acquisition exposure time set to generate a display screen acquisition image set; The background image set is used to perform background subtraction processing on the images in the display screen captured image set.

2. The method according to claim 1, characterized in that The step of generating a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera comprises: Set the target grayscale upper limit and target grayscale lower limit according to the maximum grayscale saturation of the acquisition camera; Adjusting the exposure time of the acquisition camera so that the grayscale of the image acquired by the acquisition camera is respectively the target grayscale upper limit and the target grayscale lower limit, and recording the corresponding exposure time upper limit and exposure time lower limit; Get the exposure time interval; A set of exposure times to be divided is generated according to the exposure time upper limit, the exposure time lower limit and the exposure time interval.

3. The method according to claim 2, characterized in that The target display screen does not include an in-screen circuit area; The step of performing quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented to generate a quasi-linear exposure interval set comprises: Determine a sampling area of ​​each image to be segmented in the set of images to be segmented, and determine a grayscale mean value of the sampling area; Generate a segmentation interval value for each exposure time to be segmented according to the grayscale mean value of the sampling area; Sequentially perform interval analysis on all the segmentation interval values ​​of the exposure time to be segmented; If two adjacent exposure times to be divided belong to the same linear exposure time interval, the latter exposure time to be divided is merged into the linear exposure time interval of the former exposure time to be divided; If two adjacent exposure times to be divided do not belong to the same linear exposure time interval, the latter exposure time to be divided is used as the starting point of a new linear exposure time interval; When all exposure time intervals to be segmented are analyzed, a set of quasi-linear exposure intervals is generated.

4. The method according to claim 1, characterized in that The target display screen includes an in-screen circuit area, a non-in-screen circuit area is used as an effective area, and the in-screen circuit area is used as a background area, and the camera optical axis center of the sampling camera is located in the in-screen circuit area; In the scenario of the display screen production process, the step of placing the target display screen in a darkroom environment so that the target display screen displays a preset background image includes: In the scenario of the display screen production process, the target display screen is placed in a darkroom environment, and a first display screen pattern image is obtained, where the first display screen pattern image is an image when the non-circuit area is used as the background area; Generate target grayscale data according to the grayscale data of the first display screen Pattern image and the reflectivity data of the circuit area in the screen; Adjusting the arrangement rule according to the arrangement rule of the pixels that need to be lit in the first display screen Pattern and the reflectivity data of the circuit area in the screen; Generate a second display screen Pattern image according to the adjusted arrangement rule and target grayscale data; Acquire the circuit area image features of the target display screen, and integrate the circuit area image features into the second display screen Pattern screen through a feature fusion model to generate a background screen; The target display screen is made to display a background picture.

5. The method according to claim 4, characterized in that The camera center point of the sampling camera is aligned with the circuit area within the screen; The step of performing quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented to generate a quasi-linear exposure interval set comprises: Determine a sampling area of ​​each image to be segmented in the set of images to be segmented according to the center of the camera optical axis, and detect the grayscale mean of the sampling area, wherein the sampling area is located in the circuit area within the screen; Generate a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean value of the sampling area; Sequentially perform interval analysis on all the segmentation interval values ​​of the exposure time to be segmented; If two adjacent exposure times to be divided belong to the same linear exposure time interval, the latter exposure time to be divided is merged into the linear exposure time interval of the former exposure time to be divided; If two adjacent exposure times to be divided do not belong to the same linear exposure time interval, the latter exposure time to be divided is used as the starting point of a new linear exposure time interval; When all exposure time intervals to be segmented are analyzed, a set of quasi-linear exposure intervals is generated.

6. The method according to claim 5, characterized in that The circuit type of the circuit area within the screen is a uniform circuit area; The step of generating a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean value of the sampling area comprises: Determining a reflectivity parameter corresponding to a uniform circuit area from the reflectivity data; Adjusting the grayscale mean of the sampling area according to the reflectivity parameter to generate a first grayscale mean; A segmentation interval value is generated for each exposure time to be segmented according to the first grayscale mean value and the sampling area grayscale mean value.

7. The method according to claim 5, characterized in that The circuit type of the circuit area in the screen is a gradient circuit area; The step of generating a segmentation interval value for each exposure time to be segmented according to the reflectivity data of the circuit area in the screen and the grayscale mean value of the sampling area comprises: Determine the maximum reflectivity parameter, the minimum reflectivity parameter and the circuit ratio corresponding to the gradient circuit area from the reflectivity data; Adjusting the grayscale mean of the sampling area according to the maximum reflectivity parameter, the minimum reflectivity parameter and the line ratio to generate a second grayscale mean; A segmentation interval value is generated for each exposure time to be segmented according to the second grayscale mean and the sampling area grayscale mean.

8. The method according to any one of claims 1 to 7, characterized in that The step of generating a background subtraction exposure time set according to the acquisition exposure time set corresponding to the display screen Pattern screen set of the current production process and the quasi-linear exposure interval set comprises: Determine the acquisition exposure time corresponding to each display screen pattern screen in the display screen pattern screen set of the current production process; Classify each collected exposure time into a corresponding quasi-linear exposure interval; The collected exposure time in each quasi-linear exposure interval is screened, a background subtraction exposure time is determined from each quasi-linear exposure interval, and a background subtraction exposure time set is generated, and each background subtraction exposure time corresponds to at least one display screen Pattern screen.

9. A device for background subtraction based on a display screen, characterized in that: include: A setting unit, used to place a target display screen in a darkroom environment in a display screen production process, so that the target display screen displays a preset background picture; A first generating unit, used for generating a set of exposure times to be segmented according to the maximum grayscale saturation of the sampling camera; A second generating unit is used to collect images of the target display screen according to the exposure time set to be segmented, and generate a set of images to be segmented; A third generating unit is used to perform quasi-linear interval segmentation on the exposure time set to be segmented according to the grayscale data in the image set to be segmented, and generate a quasi-linear exposure interval set, wherein the quasi-linear exposure interval set includes at least two quasi-linear exposure intervals; A fourth generating unit, configured to generate a background subtraction exposure time set according to a collection exposure time set corresponding to a display screen Pattern picture set of a current production process and the quasi-linear exposure interval set; A fifth generating unit, configured to use the sampling camera to collect the target display screen according to the background subtraction exposure time set to generate a background image set; A display unit, used to display each display pattern screen using the target display screen; A sixth generating unit, configured to perform image acquisition on the target display screen according to the acquisition exposure time set, and generate a display screen acquisition image set; A subtraction unit is used to perform background subtraction processing on the images in the display screen captured image set using the background image set.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, and when the program is executed on a computer, the method according to any one of claims 1 to 8 is performed.

Images