A method, device, and storage medium for correcting uneven viewing angles of display screen images

Abstract

The present application discloses a method, device and storage medium for correcting uneven viewing angles of display screen images, which are used to improve the optical measurement accuracy of the Mura inspection and compensation system of the display screen. Align the center position of the sampling camera with the center position of the calibration screen; input the uneven viewing angle calibration screen with a gray level of g into the calibration screen; detect the ROI area on the calibration screen to generate a brightness array; input the pure color screen with a gray level of g into the calibration screen and collect it through the sampling camera to generate a pure color display screen image; calculate the gray level mean value of the pure color display screen image in the ROI area to generate a gray level mean value array; perform centering normalization processing on the brightness array and the gray level mean value array to generate normalized data; calculate a set of correction coefficients according to the normalized brightness mean value data and the normalized gray level mean value data; generate a set of distance arrays; construct an uneven viewing angle model of the calibration screen at gray level g according to the distance array and the set of correction coefficients.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of display screen detection, and in particular, to a method, device, and storage medium for correcting uneven viewing angles of display screen images. Background Art

[0002] With the update and iteration of electronic products, display screens are used more and more frequently, and consumers have higher requirements for product quality. This has led to an increasing amount of detection content for display screens and an increasing detection accuracy.

[0003] In the automatic AOI system and optical De-Mura system of display screens, high-precision industrial cameras are used to capture images of the lit display screens. However, for a lit display screen, the optical brightness and chromaticity observed at different viewing angles are inherently uneven, and the Mura quality inspection standards for the display screen itself almost all give optical parameters at a 0° field of view angle, which brings great systematic errors to the optical Mura inspection and Mura compensation of the display screen, resulting in a significant reduction in the inspection effect or repair effect. At the same time, the quantitative analysis of this viewing angle-related non-uniformity is inconsistent at different display gray levels. Generally, lower gray levels have more serious viewing angle non-uniformity than higher gray levels; the phenomenon of viewing angle non-uniformity is also different for display screens with different pixel arrangements or manufacturing processes. It can be concluded that as the precision and functionality of the display screen increase, the more precise the display screen, the more complex the situation of display screen viewing angle non-uniformity, the greater the difficulty in detecting display screen viewing angle non-uniformity, and thus the greater the difficulty in correcting the operation of display screen viewing angle non-uniformity, reducing the correction effect of display screen viewing angle non-uniformity.

[0004] Currently, manufacturers of optical inspection and compensation systems for specific display screens give correction models by manually trying different strength correction coefficients under specific shooting conditions and confirming the compensation or inspection effect through human eye observation. This method has many defects: First, the method of manual trial is time-consuming and laborious, and it is difficult to obtain an ideally theoretical correction effect. Second, the method of human eye observation to confirm the effect has too high a dependence on human subjective judgment, introducing a large subjective error into the system. Sometimes, for the optical Mura inspection and Mura compensation of display screens, it is necessary to first model the light-emitting characteristics at different gray levels and then detect or compensate. The errors introduced by the method of manual trial and manual confirmation will be further transmitted to the subsequent optical modeling process, resulting in the model being greatly affected by the viewing angle non-uniformity factor, and the final inspection and compensation effects are poor, reducing the correction effect of display screen viewing angle non-uniformity, and thus reducing the optical measurement accuracy of the display screen Mura inspection and compensation system. Summary of the Invention

[0005] The present application discloses a method, device, and storage medium for correcting uneven viewing angles of a display screen image, which are used to improve the correction effect of uneven viewing angles of the display screen.

[0006] The first aspect of the present application discloses a method for correcting uneven viewing angles of a display screen image, including:

[0007] Align the central position of the sampling camera with the central position of the calibration screen, where the calibration screen and the display screen to be corrected are of the same type;

[0008] Input a viewing angle uneven calibration image with a gray level of g into the calibration screen. There are several ROI regions distributed on the viewing angle uneven calibration image, and the ROI regions are determined by a chromometer for implementation detection;

[0009] Use a chromometer to detect the ROI regions on the calibration screen and generate a brightness array;

[0010] Input a solid color image with a gray level of g into the calibration screen and collect it through a sampling camera to generate a solid color display screen image;

[0011] Calculate the gray level mean value of the solid color display screen image in the ROI regions to generate a gray level mean value array;

[0012] Perform centering normalization processing on the brightness array and the gray level mean value array to generate normalized data, where the normalized data includes normalized brightness mean value data and normalized gray level mean value data;

[0013] Calculate a set of correction coefficients according to the normalized brightness mean value data and the normalized gray level mean value data;

[0014] Generate a set of distance arrays according to the relationship between the position of each ROI region and the central position;

[0015] Construct a viewing angle uneven model of the calibration screen at gray level g according to the distance array and the set of correction coefficients.

[0016] Optionally, the calibration screen is a bendable transparent touch screen, and the chromometer includes a convex chromometer and a concave chromometer;

[0017] Using a chromometer to detect the ROI regions on the calibration screen and generate a brightness array includes:

[0018] Bend the bendable transparent touch screen with a preset curvature so that the front side of the transparent touch screen is located in the inner bent part;

[0019] Use a convex chromometer to perform brightness detection on the ROI regions on the front side of the transparent touch screen to generate a first initial brightness array;

[0020] Use a concave chromometer to perform brightness detection on the ROI regions on the back side of the transparent touch screen to generate a second initial brightness array;

[0021] Generate a final brightness array based on the first initial brightness array, the second initial brightness array, and in combination with the light transmission parameters of the transparent touch screen.

[0022] Optionally, the solid color display screen image includes a front acquisition image and a back acquisition image;

[0023] Calculate the grayscale mean value of the solid color display screen image in the ROI region to generate a grayscale mean value array, including:

[0024] Perform flattening and correction processing on the pixel points on the front acquisition image and the back acquisition image in the solid color display screen image;

[0025] Calculate the grayscale mean value of the flattened front acquisition image in the ROI region to generate a first grayscale mean value array;

[0026] Calculate the grayscale mean value of the flattened back acquisition image in the ROI region to generate a second grayscale mean value array;

[0027] Generate a final grayscale mean value array based on the first grayscale mean value array and the second grayscale mean value array and in combination with the light transmission weight of the transparent touch screen.

[0028] Optionally, input a solid color screen with a gray level of g into the calibration screen, and collect it through a sampling camera to generate a solid color display screen image, including:

[0029] Input a solid color screen with a gray level of g into the transparent touch screen, and the transparent touch screen is bent with a preset curvature;

[0030] Perform backlight shielding processing on the transparent touch screen;

[0031] Use the acquisition camera to take a picture of the transparent touch screen to generate a front acquisition image;

[0032] Perform frontlight shielding processing on the transparent touch screen;

[0033] Use the acquisition camera to take a picture of the transparent touch screen to generate a back acquisition image, and determine the front acquisition image and the front acquisition image as the solid color display screen image.

[0034] Optionally, after constructing the viewing angle non-uniformity model of the calibration screen at gray level g according to the distance array and the correction coefficient set, the method further includes:

[0035] Generate a correction coefficient map according to the spatial position relationship of each pixel point and the viewing angle non-uniformity model;

[0036] Obtain the image to be corrected, and use the correction coefficient map to correct the image to be corrected to generate a corrected image;

[0037] Calculate the average gray value corresponding to each ROI region in the corrected image and perform centering normalization to obtain the centered and normalized average gray value after correction;

[0038] Evaluate the viewing angle non-uniformity model based on the relative brightness of the image to be corrected, the centered and normalized average gray value, and the reference brightness.

[0039] Optionally, align the center position of the sampling camera with the center position of the calibration screen, including:

[0040] Input the centered calibration Pattern screen into the calibration screen;

[0041] Adjust the sampling camera so that the center of the image formed by the sampling camera is aligned with the center of the crosshair on the calibration screen.

[0042] The second aspect of the present application discloses a device for correcting the viewing angle non-uniformity of a display screen image, including:

[0043] An alignment unit for aligning the center position of the sampling camera with the center position of the calibration screen, where the calibration screen and the display screen to be corrected are of the same type;

[0044] An input unit for inputting a viewing angle non-uniformity calibration screen with a gray level of g into the calibration screen, where several ROI regions are distributed on the viewing angle non-uniformity calibration screen, and the ROI regions are determined by a chromometer for implementation detection;

[0045] A first generation unit for detecting the ROI regions on the calibration screen using a chromometer and generating a brightness array;

[0046] A second generation unit for inputting a pure color screen with a gray level of g into the calibration screen and collecting it through a sampling camera to generate a pure color display screen image;

[0047] A third generation unit for calculating the average gray value of the pure color display screen image in the ROI regions and generating an average gray value array;

[0048] A fourth generation unit for performing centering normalization processing on the brightness array and the average gray value array to generate normalized data, where the normalized data includes normalized average brightness mean data and normalized average gray value data;

[0049] A calculation unit for calculating a set of correction coefficients based on the normalized average brightness mean data and the normalized average gray value data;

[0050] A fifth generation unit for generating a set of distance arrays according to the relationship between the position of each ROI region and the center position;

[0051] A construction unit for constructing a viewing angle non-uniformity model of the calibration screen at gray level g based on the distance array and the set of correction coefficients.

[0052] Optionally, the calibration screen is a bendable transparent touch screen, and the colorimeter includes a convex colorimeter and a concave colorimeter;

[0053] The first generation unit includes:

[0054] Bend the bendable transparent touch screen with a preset curvature so that the front of the transparent touch screen is located in the inner bent part;

[0055] Use the convex colorimeter to detect the brightness of the ROI area on the front of the transparent touch screen and generate a first initial brightness array;

[0056] Use the concave colorimeter to detect the brightness of the ROI area on the back of the transparent touch screen and generate a second initial brightness array;

[0057] Generate a final brightness array according to the first initial brightness array, the second initial brightness array and in combination with the light transmission parameter of the transparent touch screen.

[0058] Optionally, the pure color display screen image includes a front acquisition image and a back acquisition image;

[0059] The third generation unit includes:

[0060] Perform flattening and correction processing on the pixel points on the front acquisition image and the back acquisition image in the pure color display screen image;

[0061] Calculate the gray scale mean value of the flattened front acquisition image in the ROI area and generate a first gray scale mean value array;

[0062] Calculate the gray scale mean value of the flattened back acquisition image in the ROI area and generate a second gray scale mean value array;

[0063] Generate a final gray scale mean value array according to the first gray scale mean value array and the second gray scale mean value array and in combination with the light transmission weight of the transparent touch screen.

[0064] Optionally, the second generation unit includes:

[0065] Input a pure color screen with a gray scale of g into the transparent touch screen, and the transparent touch screen is bent with a preset curvature;

[0066] Perform backlight shading treatment on the transparent touch screen;

[0067] Use the acquisition camera to take a picture of the transparent touch screen and generate a front acquisition image;

[0068] Perform frontlight shading treatment on the transparent touch screen;

[0069] Use the acquisition camera to take a picture of the transparent touch screen and generate a back acquisition image, and determine the front acquisition image and the front acquisition image as the pure color display screen image.

[0070] Optionally, after the construction unit, the device further includes:

[0071] A sixth generation unit for generating a correction coefficient map according to the spatial position relationship of each pixel point and the perspective non-uniformity model;

[0072] A seventh generation unit for obtaining the image to be corrected, correcting the image to be corrected using the correction coefficient map, and generating a corrected image;

[0073] An eighth generation unit for calculating the gray mean value corresponding to each ROI region in the corrected image and performing centering normalization to obtain the centered and normalized gray mean value after correction;

[0074] An evaluation unit for evaluating the perspective non-uniformity model according to the relative brightness of the image to be corrected, the centered and normalized gray mean value, and the reference brightness.

[0075] Optionally, the alignment unit includes:

[0076] Input the centering calibration Pattern screen into the calibration screen;

[0077] Adjust the sampling camera so that the center of the image formed by the sampling camera is aligned with the center of the crosshair on the calibration screen.

[0078] The third aspect of the present application provides a system for correcting the perspective non-uniformity of a display screen image, including:

[0079] A processor, a memory, an input / output unit, and a bus;

[0080] The processor is connected to the memory, the input / output unit, and the bus;

[0081] The memory stores a program, and the processor calls the program to execute the methods in the first aspect and any optional methods of the first aspect.

[0082] The fourth aspect of the present application provides a computer-readable storage medium, on which a program is stored, and when the program is executed on a computer, it executes the methods in the first aspect and any optional methods of the first aspect.

[0083] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0084] In this application, first, align the center position of the sampling camera with the center position of the calibration screen, and use the center position as the reference position. Here, the calibration screen and the display screen to be corrected are of the same type. Input the calibration image for non-uniform viewing angle with gray level g into the calibration screen. There are several ROI regions distributed on the calibration image for non-uniform viewing angle. The distances from each ROI region to the center position are not exactly the same, and the orientations of the ROI regions are also different. Here, the ROI regions are determined by the colorimeters for implementation detection. Next, use the colorimeters to detect the ROI regions on the calibration screen and generate the corresponding luminance arrays. Then, input the pure color image with gray level g into the calibration screen and collect it through the sampling camera to generate the pure color display screen image. Calculate the gray level mean value of the pure color display screen image in the ROI regions to generate the gray level mean value array, that is, generate the corresponding array of gray level mean values and luminance values. Perform centering normalization processing on the luminance array and the gray level mean value array to generate the normalized data, and the normalized data includes the normalized luminance mean value data and the normalized gray level mean value data. Calculate the set of correction coefficients according to the normalized luminance mean value data and the normalized gray level mean value data. Generate a set of distance arrays according to the relationship between the position of each ROI region and the center position. Construct the non-uniform viewing angle model of the calibration screen at gray level g according to the distance arrays and the set of correction coefficients. The non-uniform viewing angle model of this calibration screen can perform non-uniform viewing angle correction on the pixel points at any position on other display screens to be corrected of the same type.

[0085] Accurately calibrate and measure the non-uniformity of the viewing angle under the display gray level of the display screen, and at the same time, perform mathematical modeling on the non-uniformity of the viewing angle under the display gray level to obtain a relatively unified mathematical coefficient model of the non-uniformity of the viewing angle. This mathematical coefficient model can correct the imaging system error caused by the non-uniformity of the viewing angle of the display screen during the photographing process, improve the correction effect of the non-uniformity of the viewing angle of the display screen, and further improve the optical measurement accuracy of the display screen Mura inspection and compensation system. Brief Description of the Drawings

[0086] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0087] Figure 1 It is a schematic diagram of an embodiment of the method for correcting the non-uniformity of the viewing angle of the display screen image of the present application;

[0088] Figure 2 It is a schematic diagram of an embodiment of the method for generating the luminance array of the present application;

[0089] Figure 3Schematic diagram of an embodiment of the method for generating a grayscale mean array for this application;

[0090] Figure 4 Schematic diagram of an embodiment of the method for generating a solid - color display screen image for this application;

[0091] Figure 5 Schematic diagram of an embodiment of the method for evaluating the perspective non - uniformity model for this application;

[0092] Figure 6 Schematic diagram of an embodiment of the method for aligning the center position of the sampling camera and the center position of the calibration screen for this application;

[0093] Figure 7 Schematic diagram of an embodiment of the device for correcting perspective non - uniformity of the display screen image for this application;

[0094] Figure 8 Schematic diagram of another embodiment of the device for correcting perspective non - uniformity of the display screen image for this application;

[0095] Figure 9 Schematic diagram of an embodiment of the 25 - gray - scale display screen image before correction for this application;

[0096] Figure 10 Schematic diagram of an embodiment of the 25 - gray - scale display screen image after correction for this application. Detailed implementation manners

[0097] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are presented to provide a thorough understanding of the embodiments of this application. However, those skilled in the art should clearly understand that this application can 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 avoid unnecessary details from interfering with the description of this application.

[0098] It should be understood that when used in the specification and the appended claims of this application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0099] It should also be understood that the term "and / or" used in the specification and the appended claims of this application refers to any combination and all possible combinations of one or more of the related listed items, and includes these combinations.

[0100] As used in the specification of this application and the appended claims, the term "if" may be construed, depending on the context, as "when", "once", "in response to determining", or "in response to detecting". Similarly, the phrases "if determined" or "if [the described condition or event] is detected" may be construed, depending on the context, to mean "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]".

[0101] In addition, in the description of the specification of this application and the appended claims, the terms "first", "second", "third", etc. are used only for distinguishing descriptions and cannot be construed as indicating or implying relative importance.

[0102] References in the specification of this application to "one embodiment" or "some embodiments", etc. mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having", and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0103] In the prior art, manufacturers of optical inspection and compensation systems for specific display screens determine the correction model by manually trying different correction coefficients under specific shooting conditions and observing the compensation or inspection effects with the human eye. This method has many defects: First, the method of manual trial is time-consuming and laborious, and it is difficult to obtain an ideally theoretical correction effect. Second, the method of observing and confirming the effect with the human eye has too high a dependence on human subjective judgment, introducing a large subjective error into the system. Sometimes, for the optical Mura inspection and Mura compensation of a display screen, it is necessary to first model the light-emitting characteristics at different gray levels and then detect or compensate. However, the errors introduced by the manual trial and manual confirmation methods will be further transmitted to the subsequent optical modeling process, resulting in the model being greatly affected by the factor of viewing angle non-uniformity, and ultimately the inspection and compensation effects are poor, reducing the correction effect of the viewing angle non-uniformity of the display screen, and further reducing the optical measurement accuracy of the display screen Mura inspection and compensation system.

[0104] Based on this, this application discloses a method, device, and storage medium for correcting the viewing angle non-uniformity of a display screen image, which is used to improve the correction effect of the viewing angle non-uniformity of the display screen.

[0105] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0106] The method of the present application can be applied to a server, a device, a terminal, or other devices with logical processing capabilities. In this regard, the present application makes no limitation. For the convenience of description, the following will take the execution entity as a terminal as an example for description.

[0107] Please refer to Figure 1 , an embodiment of a method for correcting uneven viewing angles of a display screen image provided by the present application includes:

[0108] 101. Align the center position of the sampling camera with the center position of the calibration screen, and the calibration screen and the display screen to be corrected are of the same type;

[0109] In this embodiment, the display screen is mainly a flat screen, and the center of the display screen can be determined. For example, VR screens, flat screens, and curved screens, etc. The shapes of these display screens conform to symmetric shapes, and the center position is relatively easy to determine.

[0110] First, randomly select a calibration screen with the same specification model and the same type as the display screen to be corrected. Next, align the center position of the sampling camera with the center position of the calibration screen.

[0111] 102. Input a viewing angle unevenness calibration screen with a gray level of g into the calibration screen. There are several ROI regions distributed on the viewing angle unevenness calibration screen, and the ROI regions are determined by a chromaticity meter for implementation detection;

[0112] The terminal switches the dot screen image of the signal generator to make the screen display the viewing angle unevenness calibration Pattern screen (viewing angle unevenness calibration screen). Specifically, input a viewing angle unevenness calibration screen with a gray level of g into the calibration screen. There are several ROI regions distributed on the viewing angle unevenness calibration screen, and the ROI regions are determined by a chromaticity meter for implementation detection.

[0113] 103. Use a chromaticity meter to detect the ROI regions on the calibration screen and generate a brightness array;

[0114] The terminal uses a chromaticity meter to detect the ROI regions on the calibration screen and generate a brightness array. Specifically, use a dot chromaticity meter (such as: CA-VP410 or CA-VP427) to measure the display screen brightness values within each ROI region, and record them as a brightness array with a length of m×n: L g . Usually, the recording order is generally set from left to right and from top to bottom, which can be flexibly set.

[0115] 104. Input a pure color screen with a gray level of g into the calibration screen, and collect it through a sampling camera to generate a pure color display screen image;

[0116] The terminal inputs a pure color screen with a gray level of g into the calibration screen, and collects it through a sampling camera to generate a pure color display screen image. Specifically, the terminal makes the calibration screen display a pure color Pattern screen (without a calibration ROI frame) with a gray level of g by switching the dot screen image of the signal generator. Adjust the exposure time of the sampling camera to take a picture of the calibration screen, and obtain a pure color Pattern display screen image (pure color display screen image) with a gray level of g, denoted as image I g 。

[0117] So far, the data acquisition is completed, and the following are obtained: the true brightness array L of the display screen at gray level g g and the pure color display screen image I g 。

[0118] 105. Calculate the gray level mean value of the pure color display screen image in the ROI area to generate a gray level mean value array;

[0119] The terminal first obtains the image type type of the pure color display screen image I g . If type is not of double type, then convert the pure color display screen image I g to double type. Calculate the gray level mean value of the pure color display screen image in the ROI area to generate a gray level mean value array. Calculate the gray level mean value of the image I g in each calibration ROI area to obtain a gray level mean value array with a length of m×n, denoted as: G g , and the order of the recorded gray level mean values is consistent with the brightness data.

[0120] 106. Perform centering normalization processing on the brightness array and the gray level mean value array to generate normalized data, and the normalized data includes normalized brightness mean value data and normalized gray level mean value data;

[0121] Respectively perform centering normalization on the brightness array L g and the gray level mean value array G g , that is, first respectively obtain the values L g , G g at the center of the data length, which are the brightness data and the gray level mean value of the ROI area at the center of the image, and then divide each value in L gc by L gc , and divide each value in G g by G gc , G g by G gc , then the centered normalized brightness and gray level mean values are obtained, denoted as L g_Norm , G g_Norm, which is expressed by the formula as:

[0122]

[0123]

[0124] In the formula, L gc , G gc respectively represent the values at the center of the data lengths of the arrays L g , G g .

[0125] 107. Calculate a set of correction coefficients according to the normalized luminance mean data and the normalized grayscale mean data;

[0126] The terminal calculates the ideal correction coefficient Kg corresponding to each ROI region. Specifically, the terminal calculates a set of correction coefficients according to the normalized luminance mean data and the normalized grayscale mean data. The significance of the display screen viewing angle non-uniformity correction is to recover the ideal relative luminance from the image I g . And according to the aforementioned known true luminance of the ROI region, the data L g_Norm (which can be regarded as the ideal relative luminance) and the centered and normalized grayscale mean G g_Norm (which can be regarded as the true relative luminance), so the ideal correction coefficient K g_Norm corresponding to each ROI region can be calculated by using L g_Norm , G g , which is expressed by the formula as:

[0127]

[0128] 108. Generate a set of distance arrays according to the relationship between the position of each ROI region and the center position;

[0129] 109. Construct a viewing angle non-uniformity model of the calibrated screen at gray level g according to the distance arrays and the set of correction coefficients.

[0130] The terminal uses the spatial position of the ROI region on the calibrated screen relative to the image center and the ideal correction coefficient K g to perform viewing angle non-uniformity modeling. The viewing angle relationship can be approximately understood as the spatial position relationship of pixels relative to the image center when reflected in the camera-captured image. Therefore, here it is considered to use the spatial position of the ROI region relative to the image center and the ideal correction coefficient K g to perform viewing angle non-uniformity modeling. The specific modeling method is as follows:

[0131] Calculate the distance from the center of each ROI region to the center of the image to obtain a set of distance data, denoted as dist. Similarly, the order of the recorded spatial distances is consistent with the aforementioned luminance data. Taking the spatial distance dist as the independent variable and the ideal correction coefficient Kg as the dependent variable, and using the least squares method to perform polynomial fitting on Kg and dist, the viewing angle non-uniformity model of the display screen at gray level g can be obtained, denoted as f(dist).

[0132] It should be noted that the above viewing angle non-uniformity modeling method is only one specific embodiment listed. Similarly, it is also possible to separately perform fitting modeling in the horizontal and vertical directions or use methods such as surface fitting and machine learning for modeling.

[0133] In this embodiment, first, align the center position of the sampling camera with the center position of the calibration screen, and use the center position as the reference position, where the calibration screen and the display screen to be corrected are of the same type. Input the viewing angle non-uniformity calibration screen with gray level g into the calibration screen. There are several ROI regions distributed on the viewing angle non-uniformity calibration screen. The distance from each ROI region to the center position is not exactly the same, and the orientations of the ROI regions are also different. Among them, the ROI regions are determined by the chromometers for implementation detection. Next, use the chromometers to detect the ROI regions on the calibration screen to generate corresponding luminance arrays. Then input the pure color screen with gray level g into the calibration screen and collect it through the sampling camera to generate a pure color display screen image. Calculate the gray level mean value of the pure color display screen image in the ROI region to generate a gray level mean value array, that is, generate a corresponding array of gray level mean values and luminance values. Perform centering normalization processing on the luminance array and the gray level mean value array to generate normalized data, and the normalized data includes normalized luminance mean value data and normalized gray level mean value data. Calculate the correction coefficient set according to the normalized luminance mean value data and the normalized gray level mean value data. Generate a set of distance arrays according to the relationship between the position of each ROI region and the center position. Construct the viewing angle non-uniformity model of the calibration screen at gray level g according to the distance array and the correction coefficient set. The viewing angle non-uniformity model of this calibration screen can perform viewing angle non-uniformity correction on pixel points at any position on other display screens to be corrected of the same type.

[0134] Accurately calibrate and measure the viewing angle non-uniformity of the display screen at the displayed gray level, and at the same time perform mathematical modeling on the viewing angle non-uniformity at the displayed gray level to obtain a relatively unified mathematical coefficient model of the viewing angle non-uniformity. This mathematical coefficient model can correct the imaging system error caused by the viewing angle non-uniformity of the display screen during the photographing process, improve the correction effect of the viewing angle non-uniformity of the display screen, and further improve the optical measurement accuracy of the display screen Mura inspection and compensation system.

[0135] Please refer to Figure 2, the calibration screen of this application is a bendable transparent touch screen. The colorimeter includes a convex colorimeter and a concave colorimeter. An embodiment of a method for generating a luminance array provided by this application includes:

[0136] 201. Bend the bendable transparent touch screen to a preset curvature so that the front side of the transparent touch screen is located at the inner bent part;

[0137] In this embodiment, in addition to constructing a model for the use of a flat screen, a model for correcting the viewing angle non-uniformity can also be constructed for special types of display screens.

[0138] Specifically, the display screen targeted in this embodiment is a bendable transparent touch screen. There are many differences between the bendable transparent touch screen and the flat screen. First, the bendable transparent touch screen needs to be bent to different degrees during use. At each pixel position, the bending degree is different. Therefore, during the detection process, in links such as luminance detection and average value detection, the detection effect is very different from that of the flat screen. The flat screen displays for users through a flat surface, and after the flat screen is bent, its display effect will change. Second, in addition to the bending factor, the light transmissibility of the transparent display screen will also have a great impact on the correction of viewing angle non-uniformity. Because when there are abnormalities, defects, or deviations in the pixels of the transparent touch screen, a part of the light displayed on the front will be transmitted to the back. Even for qualified pixels, there may be slight deviations, and a part of the light displayed on the front will also be transmitted to the back, resulting in the actual display gray level and real brightness not meeting the requirements of the preset input image. During the detection process, the detected gray level and brightness also differ from the preset input image.

[0139] Moreover, the two characteristics of bending and transparency make it more difficult to correct the viewing angle non-uniformity of this type of bendable transparent touch screen (especially when there are abnormalities, defects, or deviations in the pixels). Because this type of bendable transparent touch screen usually needs to be bent to different degrees during use, at any gray level and any bending degree, the display situation of the transparent touch screen will be different. Because the light fiber displayed on the front is transmitted to the back, which is not a flat surface, but a concave or convex surface, this makes the reflected light more complex, and makes the difference between the actual gray level average value and brightness and the input image more complex. If the traditional detection methods for the gray level average value and brightness of the flat screen are applied to the bendable transparent touch screen, the accuracy of the detection results will be reduced.

[0140] Secondly, this embodiment provides a set of calibration calculation processes for correcting the imaging system error caused by the viewing angle non-uniformity of the LCD display screen during the photographing process, which can be used to improve the optical measurement accuracy of the LCD display screen Mura inspection and compensation system, and further achieve a better inspection and compensation effect.

[0141] Secondly, this embodiment provides a mathematical model for correcting the viewing angle non-uniformity of an LCD display screen. The non-uniformity of the viewing angle of the display screen can be corrected with optimal coefficients through the model, improving the calculation accuracy of the Mura inspection and compensation system.

[0142] Secondly, the mathematical modeling method for viewing angle non-uniformity given in this embodiment has high calculation efficiency and can quickly calculate the non-uniformity correction coefficients of a single image.

[0143] In this embodiment, a detection method specifically for flexible transparent touch screens is used to obtain data and improve the accuracy of correcting the viewing angle non-uniformity of flexible transparent touch screens.

[0144] First, the terminal bends the flexible transparent touch screen to a preset curvature so that the front side of the transparent touch screen is located in the inner bent part, that is, the flexible transparent touch screen is first adjusted to a preset bending degree.

[0145] It should be noted that in this embodiment, the front side of the transparent touch screen is located in the inner bent part. In other cases, it can also be that the front side is located in the outer bent part, that is, the front side of the display screen can be concave or convex, and the calculation methods are similar.

[0146] 202. Use a convex colorimeter to detect the brightness of the ROI area on the front side of the transparent touch screen to generate a first initial brightness array;

[0147] When the front side of the display screen is concave, use a convex colorimeter to detect the brightness of the ROI area on the front side of the transparent touch screen to generate a first initial brightness array. At this time, the detection port of the convex colorimeter and the bent transparent touch screen fit each other, which can improve the accuracy of brightness detection.

[0148] If the front side of the display screen is convex, then use a concave colorimeter to detect the brightness of the ROI area on the front side of the transparent touch screen.

[0149] 203. Use a concave colorimeter to detect the brightness of the ROI area on the back side of the transparent touch screen to generate a second initial brightness array;

[0150] Next, the terminal uses a concave colorimeter to detect the brightness of the ROI area on the back side of the transparent touch screen to generate a second initial brightness array.

[0151] 204. Generate a final brightness array according to the first initial brightness array, the second initial brightness array and in combination with the light transmission parameters of the transparent touch screen.

[0152] When the transparent touch screen displays an image of gray scale g, some pixel points with slight defects will cause some light to not be fully displayed from the front and will be displayed on the back. And there is a secondary reflection on the back, making the gray scale displayed on the front more complex. The terminal needs to generate a final brightness array based on the first initial brightness array, the second initial brightness array and in combination with the light transmission parameters of the transparent touch screen.

[0153] The light transmission parameter refers to a small reference screen (a bendable transparent touch screen with all the pixel points installed on the main display inside meeting the standards). At a preset bending degree, a flexible screen (light-impermeable) is used to display a pure color image of gray scale g and is attached to the back of the bendable transparent touch screen. The light transmittance is detected from the front to obtain the first light transmission parameter T g1 (by dividing the brightness detected from the flexible screen itself by the brightness detected from the front of the bendable transparent touch screen), and then it is attached to the front of the bendable transparent touch screen, and the light transmittance is detected from the back to obtain the second light transmission parameter T g2 .

[0154] The terminal generates a final brightness array based on the first initial brightness array, the second initial brightness array and in combination with the light transmission parameters of the transparent touch screen. Specifically, the first light transmission parameter and the second light transmission parameter are used to supplement and calculate the brightness values corresponding to the first initial brightness array and the second initial brightness array. The formula is as follows:

[0155] L * =T g1 *L 1i + (1 - T g2 ) *L 2i

[0156] L * is the final brightness data, the first light transmission parameter T g1 , the second light transmission parameter T g2 , L 1i is the brightness data of the i-th ROI area in the first initial brightness array, and L 2i is the brightness data of the i-th ROI area in the second initial brightness array.

[0157] In the above method, a convex colorimeter and a concave colorimeter are combined. By calculating the light transmission parameters, the problems caused by the combination of the bending and transparency characteristics of the bendable transparent touch screen are effectively eliminated. During the process of correcting the viewing angle non-uniformity, highly accurate brightness data are effectively obtained, improving the accuracy of subsequent correction.

[0158] Please refer to Figure 3 , in this application, the pure color display screen image includes a front acquisition image and a back acquisition image. This application provides an embodiment of a method for generating a gray scale mean array, including:

[0159] 301. Flatten and correct the pixel points on the front captured image and the back captured image in the pure color display screen image;

[0160] 302. Calculate the gray level mean value in the ROI area of the flattened front captured image to generate the first gray level mean value array;

[0161] 303. Calculate the gray level mean value in the ROI area of the flattened back captured image to generate the second gray level mean value array;

[0162] 304. Calculate the final gray level mean value array according to the first gray level mean value array and the second gray level mean value array and in combination with the light transmission weight of the transparent touch screen.

[0163] In this embodiment, due to the two characteristics of bending and transparency, the gray level will also be affected, but this kind of influence is not as large as that on the brightness. The terminal needs to first flatten and correct the pixel points on the front captured image and the back captured image in the pure color display screen image, so that the pixel points of the two images can correspond one by one. Flattening correction is a way to flatten the pixel point coordinates into the shape of a flat screen, and then combined with the interpolation method to complement the pixel points, and two images can be obtained. The terminal calculates the gray level mean value in the ROI area of the flattened front captured image to generate the first gray level mean value array, and then calculates the gray level mean value in the ROI area of the flattened back captured image to generate the second gray level mean value array. Finally, the true real gray level mean value is calculated through the light transmission weight.

[0164] U * = Ƞ 1 * U 1i + Ƞ 2 * U 2i

[0165] U * is the final gray level mean value array, U 1i is the gray level mean value data of the i-th ROI area in the first initial brightness array, U 2i is the gray level mean value data of the i-th ROI area in the second initial brightness array. Ƞ 1 and Ƞ 2Let \( \xi_{f} \) and \( \xi_{b} \) be the grayscale light transmittance weights on the front and back of the transparent touch screen. The calculation method is as follows: Use a flexible screen (opaque) to display a pure color image of grayscale \( g \), attach it to the back of the bendable transparent touch screen, take an image from the front, detect the first real-time grayscale average value, divide the first grayscale \( g \) by the real-time grayscale average value to obtain the first parameter, then attach it to the front of the bendable transparent touch screen, take an image from the back, detect the second real-time grayscale average value, and divide the first grayscale \( g \) by the real-time grayscale average value to obtain the second parameter. Normalize the first parameter and the second parameter to generate two grayscale light transmittance weights that add up to 1, which are the grayscale light transmittance weight \( \xi_{f} \) of the transparent touch screen on the front and the grayscale light transmittance weight \( \xi_{b} \) on the back, respectively. 1 and the grayscale light transmittance weight \( \xi_{b} \) on the back. 2 This method can effectively reduce the influence on the grayscale average value due to the two characteristics of bending and transparency. By using a flexible screen to cooperate with the display of the grayscale \( g \) image, the transmittance weights of the grayscale average values on the front and back are calculated, effectively reducing errors. During the process of correcting the viewing angle non-uniformity, highly accurate grayscale average value data is effectively obtained, improving the accuracy of subsequent correction.

[0166] Please refer to Figure 4 , an embodiment of a method for generating a pure color display screen image provided by the present application includes:

[0167] 401. Input a pure color image with grayscale \( g \) into the transparent touch screen, and the transparent touch screen is bent with a preset curvature;

[0168] 402. Perform backlight shading treatment on the transparent touch screen;

[0169] 403. Use a capture camera to capture the transparent touch screen to generate a front capture image;

[0170] 404. Perform frontlight shading treatment on the transparent touch screen;

[0171] 405. Use a capture camera to capture the transparent touch screen to generate a back capture image, and determine the front capture image and the back capture image as the pure color display screen image.

[0172] The terminal first inputs a pure-color screen with a gray scale of g into the transparent touch screen, and at this time, the transparent touch screen is bent with a preset curvature. Next, the back of the transparent touch screen is first shaded to reduce the situation where the light projected by the reduced pixel points is reflected again. Then, the terminal takes a picture of the transparent touch screen by using a capture camera to generate a front capture image. Then, the front of the transparent touch screen is shaded, and the capture camera is used to take a picture of the transparent touch screen to generate a back capture image, and the front capture image and the front capture image are determined as the pure-color display screen image. In the traditional method, only a darkroom environment is used for shooting. In this embodiment, not only a darkroom environment is adopted, but also shading treatments for the front and the back are required, further reducing the occurrence of complex situations such as light refraction and reflection, and improving the accuracy for subsequent calculations.

[0173] Please refer to Figure 5 , an embodiment of a method for evaluating a perspective unevenness model provided by this application includes:

[0174] 501. Generate a correction coefficient map according to the spatial position relationship of each pixel point and the perspective unevenness model;

[0175] 502. Obtain the image to be corrected, and use the correction coefficient map to correct the image to be corrected to generate a corrected image;

[0176] 503. Calculate the gray-scale mean value corresponding to each ROI region in the corrected image and perform centering normalization to obtain the centered and normalized gray-scale mean value after correction;

[0177] 504. Evaluate the perspective unevenness model according to the relative brightness of the image to be corrected, the centered and normalized gray-scale mean value, and the reference brightness.

[0178] In this embodiment, the terminal first calculates the correction coefficient map. A correction coefficient map is generated according to the spatial position relationship of each pixel point and the perspective unevenness model. Specifically, a correction coefficient map, denoted as I, is calculated according to the spatial position relationship of each pixel point on the image and the perspective unevenness model obtained by calibration and modeling. COE . For example, according to one of the above embodiments: First, the distance dist_i of each pixel point on the image relative to the center of the image can be calculated, and then the corresponding correction coefficient can be calculated according to the model f(dist). In this way, the correction coefficient map I can be obtained by using parallel computing and multi-core optimization. COE .

[0179] Next, for perspective unevenness correction, the terminal obtains the image to be corrected, and uses the correction coefficient map to correct the image to be corrected to generate a corrected image. The image to be corrected I g is multiplied with the correction coefficient map I COE to obtain the corrected image I correct .

[0180] Please refer to Figure 9 and Figure 10 , Figure 9 and Figure 10 Images before and after 25 - gray - scale correction in one of the embodiments. It can be seen that the corrected image has a better effect than the image before correction, and the correction model obtained using the calibration screen has achieved a good correction effect.

[0181] Next, for the evaluation of the correction effect, the terminal calculates the gray - scale mean value corresponding to each ROI region in the corrected image and performs centering normalization to obtain the centered and normalized gray - scale mean value after correction. Specifically, the terminal calculates the gray - scale mean value corresponding to each calibrated ROI region in the corrected image Icorrect and performs centering normalization to obtain the centered and normalized gray - scale mean value after correction (which can be regarded as the relative brightness after correction), denoted as G correct_Norm . The non - uniform viewing - angle model is evaluated according to the relative brightness of the image to be corrected, the centered and normalized gray - scale mean value, and the reference brightness. Specifically, the relative brightness G g_Norm before correction, which is known, and the ideal relative brightness L g_Norm are used. Therefore, the similarity between the relative brightness before and after correction and the ideal relative brightness can be calculated respectively to evaluate the correction effect. The similarity measurement method can adopt methods such as Euclidean distance method, Manhattan distance, Chebyshev distance, cosine similarity, and Hamming similarity.

[0182] Please refer to Figure 6 , this application provides an embodiment of a method for aligning the center position of the sampling camera and the center position of the calibration screen, including:

[0183] 601. Input the centered calibration Pattern screen into the calibration screen;

[0184] 602. Adjust the sampling camera so that the center of the image formed by the sampling camera is aligned with the center of the crosshair on the calibration screen.

[0185] First, the terminal randomly selects a calibration screen with the same specification model and the same type as the display screen to be corrected. Next, the center position of the sampling camera is aligned with the center position of the calibration screen. Specifically, the terminal makes a cross - centered calibration Pattern diagram with a gray - scale of g (centered calibration Pattern screen), imports the cross - centered calibration Pattern diagram into the signal generator, lights up the calibration screen and makes it display the centered Pattern screen. Next, the terminal adjusts the camera so that the center of the image formed by the sampling camera is aligned with the center of the crosshair on the display screen.

[0186] Please refer to Figure 7 , this application provides an embodiment of a device for correcting non - uniform viewing - angle of display - screen images, including:

[0187] An alignment unit 701 for aligning the center position of the sampling camera with the center position of the calibration screen, where the calibration screen and the display screen to be corrected are of the same type;

[0188] Optionally, the alignment unit 701 includes:

[0189] Input the calibration Pattern screen into the calibration screen;

[0190] Adjust the sampling camera so that the center of the image formed by the camera of the sampling camera is aligned with the center of the crosshair of the calibration screen.

[0191] An input unit 702 for inputting a non-uniform viewing angle calibration screen with a gray level of g into the calibration screen, where several ROI regions are distributed on the non-uniform viewing angle calibration screen, and the ROI regions are determined by the colorimeters for implementation detection;

[0192] A first generation unit 703 for detecting the ROI regions on the calibration screen using a colorimeter and generating a brightness array;

[0193] Optionally, the calibration screen is a flexible transparent touch screen, and the colorimeter includes a convex colorimeter and a concave colorimeter;

[0194] The first generation unit 703 includes:

[0195] Bend the flexible transparent touch screen with a preset curvature so that the front surface of the transparent touch screen is located in the inner bent part;

[0196] Use the convex colorimeter to detect the brightness of the ROI regions on the front surface of the transparent touch screen and generate a first initial brightness array;

[0197] Use the concave colorimeter to detect the brightness of the ROI regions on the back surface of the transparent touch screen and generate a second initial brightness array;

[0198] Generate a final brightness array based on the first initial brightness array, the second initial brightness array, and in combination with the light transmittance parameters of the transparent touch screen.

[0199] A second generation unit 704 for inputting a pure color screen with a gray level of g into the calibration screen and collecting it through a sampling camera to generate a pure color display screen image;

[0200] Optionally, the second generation unit 704 includes:

[0201] Input a pure color screen with a gray level of g into the transparent touch screen, and the transparent touch screen has been bent with a preset curvature;

[0202] Perform backlight shading treatment on the transparent touch screen;

[0203] Use the acquisition camera to take a picture of the transparent touch screen to generate a front acquisition image;

[0204] Perform front shading treatment on the transparent touch screen;

[0205] Use a collection camera to take pictures of the transparent touch screen, generate a back collection image, and determine the front collection image and the front collection image as a solid color display screen image.

[0206] The third generation unit 705 is used to calculate the gray scale mean value of the solid color display screen image in the ROI area and generate a gray scale mean value array;

[0207] Optionally, the solid color display screen image includes a front collection image and a back collection image;

[0208] The third generation unit 705 includes:

[0209] Perform flattening and correction processing on the pixel points on the front collection image and the back collection image in the solid color display screen image;

[0210] Calculate the gray scale mean value of the flattened front collection image in the ROI area and generate a first gray scale mean value array;

[0211] Calculate the gray scale mean value of the flattened back collection image in the ROI area and generate a second gray scale mean value array;

[0212] Calculate the final gray scale mean value array according to the first gray scale mean value array and the second gray scale mean value array and in combination with the light transmission weight of the transparent touch screen.

[0213] The fourth generation unit 706 is used to perform centering normalization processing on the brightness array and the gray scale mean value array to generate normalized data, and the normalized data includes normalized brightness mean value data and normalized gray scale mean value data;

[0214] The calculation unit 707 is used to calculate a set of correction coefficients according to the normalized brightness mean value data and the normalized gray scale mean value data;

[0215] The fifth generation unit 708 is used to generate a set of distance arrays according to the relationship between the position of each ROI area and the center position;

[0216] The construction unit 709 is used to construct a viewing angle non-uniformity model of the calibrated screen under gray scale g according to the distance array and the set of correction coefficients;

[0217] The sixth generation unit 710 is used to generate a correction coefficient map according to the spatial position relationship of each pixel point and the viewing angle non-uniformity model;

[0218] The seventh generation unit 711 is used to obtain the image to be corrected, and use the correction coefficient map to correct the image to be corrected to generate a corrected image;

[0219] An eighth generation unit 712, configured to calculate the average gray value corresponding to each ROI region in the corrected image and perform centering normalization to obtain the corrected centering-normalized average gray value;

[0220] An evaluation unit 713, configured to evaluate the viewing angle non-uniformity model according to the relative brightness of the image to be corrected, the centering-normalized average gray value, and the reference brightness.

[0221] Please refer to Figure 8 , this application provides a device for correcting the viewing angle non-uniformity of a display screen image, including:

[0222] A processor 801, a memory 802, an input / output unit 803, and a bus 804.

[0223] The processor 801 is connected to the memory 802, the input / output unit 803, and the bus 804.

[0224] The memory 802 stores a program, and the processor 801 calls the program to execute the methods such as Figure 1 , Figure 2 and Figure 3 , Figure 4 , Figure 5 and Figure 6 in the method.

[0225] This application provides a computer-readable storage medium, on which a program is stored, and when the program is executed on a computer, it executes the methods such as Figure 1 , Figure 2 and Figure 3 , Figure 4 , Figure 5 and Figure 6 in the method.

[0226] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0227] In several embodiments provided by this 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 merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, 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 displayed or discussed mutual coupling or direct coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0228] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0229] In addition, each functional unit in various embodiments of the present application may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0230] 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 such an understanding, the technical solution of the present application, in essence, 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 and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, read-only memory), random access memories (RAM, random access memory), magnetic disks, or optical discs that can store program codes.

Claims

1. A method for correcting uneven viewing angle of an image on a display screen, characterized in that: include: Align the center position of the sampling camera with the center position of the calibration screen, where the calibration screen is of the same type as the display screen to be calibrated; Inputting a viewing angle unevenness calibration picture with a gray scale of g into the calibration screen, wherein a plurality of ROI areas are distributed on the viewing angle unevenness calibration picture, and the ROI areas are determined by a colorimeter for performing detection; Using the colorimeter to detect the ROI area on the calibration screen to generate a brightness array; The colorimeter is used to detect the ROI area on the calibration screen to generate a brightness array, including: bending the bendable transparent touch screen with a preset curvature so that the front of the transparent touch screen is located at the inner curved part, the calibration screen is a bendable transparent touch screen, and the colorimeter includes a convex colorimeter and a concave colorimeter; using the convex colorimeter to detect the brightness of the ROI area on the front of the transparent touch screen to generate a first initial brightness array; using the concave colorimeter to detect the brightness of the ROI area on the back of the transparent touch screen to generate a second initial brightness array; generating a final brightness array according to the first initial brightness array, the second initial brightness array and in combination with the light transmittance parameter of the transparent touch screen; Inputting a pure color picture with a gray scale of g into the calibration screen, collecting it through the sampling camera, and generating a pure color display screen image; Calculate the grayscale mean of the pure color display screen image in the ROI area to generate a grayscale mean array; Performing centering normalization processing on the brightness array and the grayscale mean array to generate normalized data, wherein the normalized data includes normalized brightness mean data and normalized grayscale mean data; Calculating a correction coefficient set according to the normalized brightness mean data and the normalized grayscale mean data; Generate a set of distance arrays based on the relationship between the position of each ROI area and the center position; A viewing angle non-uniformity model of the calibration screen at gray scale g is constructed according to the distance array and the correction coefficient set.

2. The method according to claim 1, characterized in that The pure color display screen image includes a front-side captured image and a back-side captured image; Calculating the grayscale mean of the pure color display screen image in the ROI area to generate a grayscale mean array, including: Performing flattening correction processing on pixel points on the front collected image and the back collected image in the pure color display screen image; Calculate the grayscale mean of the flattened frontal captured image in the ROI area to generate a first grayscale mean array; Calculate the grayscale mean of the flattened back side captured image in the ROI area to generate a second grayscale mean array; A final grayscale mean array is calculated according to the first grayscale mean array and the second grayscale mean array in combination with the light transmittance weight of the transparent touch screen.

3. The method according to claim 2, characterized in that Inputting a pure color picture with gray scale g into the calibration screen, collecting it through the sampling camera, and generating a pure color display screen image, including: A pure color picture with a gray scale of g is inputted into the transparent touch screen, and the transparent touch screen is bent with a preset curvature; Performing back shading processing on the transparent touch screen; Using the sampling camera to photograph the transparent touch screen to generate a front captured image; Performing front shading processing on the transparent touch screen; The transparent touch screen is photographed using the sampling camera to generate a back-side captured image, and the front-side captured image and the front-side captured image are determined as pure color display screen images.

4. The method according to any one of claims 1 to 3, characterized in that After constructing a viewing angle unevenness model of the calibration screen at gray scale g according to the distance array and the correction coefficient set, the method further includes: Generate a correction coefficient map according to the spatial position relationship of each pixel point and the viewing angle unevenness model; Acquire an image to be corrected, and use the correction coefficient map to correct the image to be corrected to generate a corrected image; Calculate the grayscale mean corresponding to each ROI area in the corrected image and perform center normalization to obtain the corrected center normalized grayscale mean; The viewing angle non-uniformity model is evaluated according to the relative brightness of the image to be corrected, the centered normalized grayscale mean value and the reference brightness.

5. The method according to any one of claims 1 to 3, characterized in that Align the center position of the sampling camera with the center position of the calibration screen, where the calibration screen is of the same type as the display screen to be calibrated, including: Input the centering calibration pattern image into the calibration screen; The sampling camera is adjusted so that the center of the image formed by the sampling camera is aligned with the center of the crosshairs of the calibration screen.

6. A device for correcting uneven viewing angle of a display screen image, characterized in that: include: An alignment unit, used to align the center position of the sampling camera with the center position of the calibration screen, wherein the calibration screen is of the same type as the display screen to be calibrated; An input unit, used for inputting a viewing angle unevenness calibration picture with a gray scale of g into the calibration screen, wherein a plurality of ROI areas are distributed on the viewing angle unevenness calibration picture, and the ROI areas are determined by a colorimeter for performing detection; A first generating unit, configured to detect the ROI area on the calibration screen using the colorimeter to generate a brightness array; The first generation unit comprises: The bendable transparent touch screen is bent with a preset curvature so that the front of the transparent touch screen is located at the inner curved part, the calibration screen is a bendable transparent touch screen, and the colorimeter includes a convex colorimeter and a concave colorimeter; the convex colorimeter is used to perform brightness detection on the ROI area on the front of the transparent touch screen to generate a first initial brightness array; the concave colorimeter is used to perform brightness detection on the ROI area on the back of the transparent touch screen to generate a second initial brightness array; the final brightness array is generated according to the first initial brightness array, the second initial brightness array and in combination with the light transmittance parameter of the transparent touch screen; A second generating unit is used to input a pure color picture with a gray scale of g into the calibration screen, collect it through the sampling camera, and generate a pure color display screen image; A third generating unit is used to calculate the grayscale mean of the pure color display screen image in the ROI area to generate a grayscale mean array; a fourth generating unit, configured to perform centering normalization processing on the brightness array and the grayscale mean value array to generate normalized data, wherein the normalized data includes normalized brightness mean value data and normalized grayscale mean value data; A calculation unit, configured to calculate a correction coefficient set according to the normalized brightness mean data and the normalized grayscale mean data; A fifth generating unit, used for generating a set of distance arrays according to the relationship between the position of each ROI area and the center position; A construction unit is used to construct a viewing angle unevenness model of the calibration screen at gray scale g according to the distance array and the correction coefficient set.

7. A system for correcting uneven viewing angle of a display screen image, characterized in that: include: A processor, a memory, an input-output unit and a bus, wherein the processor is connected to the memory, the input-output unit and the bus, the memory stores a program, and the processor calls the program to execute the method as claimed in any one of claims 1 to 5.

8. A computer-readable storage medium having a program stored thereon, wherein the program, when executed on a computer, performs the method according to any one of claims 1 to 5.

Images