Image quality compensation method for display screen, image quality compensation apparatus, and display apparatus

Abstract

An image quality compensation method for a display screen, an image quality compensation apparatus, and a display apparatus. The image quality compensation method for a display screen comprises: acquiring target image data to be displayed on a first display screen, wherein the target image data comprises original grayscales of pixel points in different color channels, and different color channels of one pixel point correspond to different sub-pixels of one pixel unit in the first display screen (S101); and on the basis of arrangement parameters corresponding to the first display screen and the target image data, acquiring a target sub-pixel to be compensated for from among a plurality of sub-pixels, and compensating for the original grayscale, in the target image data, corresponding to the target sub-pixel, wherein the arrangement parameters comprise a driving structure of the sub-pixels, and the driving structure represents a connection architecture between data lines and the sub-pixels in the first display screen (S102).

Description

Image quality compensation method, image quality compensation device and display device for display screen Technical Field

[0001] This disclosure relates to the field of display technology, and in particular to a method, apparatus, and device for image quality compensation of a display screen. Background Technology

[0002] With the development of screen display technology, various screen display technologies such as Liquid Crystal Display (LCD) and Organic Light-Emitting Diode (OLED) are widely used in various fields. Taking LCD as an example, an LCD consists of multiple sub-pixels, each containing liquid crystal. During display, an electric field is provided to the liquid crystal to deflect the liquid crystal molecules, thereby controlling the transmittance of the backlight emitted at the sub-pixel and thus displaying the desired image.

[0003] Overview

[0004] In a first aspect, a method for image quality compensation of a display screen is provided, wherein the method includes:

[0005] Acquire target image data to be displayed on the first display screen. The target image data includes the original grayscale of each pixel in different color channels. The different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen.

[0006] Based on the arrangement parameters corresponding to the first display screen and the target image data, the target sub-pixel to be compensated is obtained from the multiple sub-pixels, and the original grayscale corresponding to the target sub-pixel in the target image data is compensated.

[0007] The arrangement parameters include the driving structure of the sub-pixels, which characterizes the connection architecture between the data lines and each sub-pixel in the first display screen.

[0008] For example, the step of obtaining the target sub-pixel to be compensated from a plurality of sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data, and compensating the original grayscale corresponding to the target sub-pixel in the target image data, includes:

[0009] The original grayscale corresponding to the target sub-pixel in the target image data is compensated using the target model;

[0010] The target model is obtained by deep learning using multiple datasets as training samples. The datasets include first image data and second image data. The second image data is obtained by compensating the original grayscale of each pixel in the first image data in some or all color channels. The compensated color channels are obtained through the arrangement parameters.

[0011] For example, obtaining the target sub-pixel to be compensated from a plurality of sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data includes:

[0012] Based on the arrangement parameters, a first sub-pixel and a second sub-pixel that are connected to the same data line and are adjacent to each other are determined from a plurality of sub-pixels; the first sub-pixel and the second sub-pixel correspond to different colors;

[0013] Obtain the difference between the original gray levels corresponding to the first sub-pixel and the second sub-pixel in the target image data;

[0014] The target sub-pixel is determined from the first sub-pixel and the second sub-pixel based on the difference and the driving order of the first sub-pixel and the second sub-pixel.

[0015] For example, determining the target sub-pixel among the first sub-pixel and the second sub-pixel based on the difference and the driving order of the first sub-pixel and the second sub-pixel includes:

[0016] If the difference is greater than a preset difference, and the original grayscale of the sub-pixel with the later driving order is higher than that of the first sub-pixel and the second sub-pixel, then the sub-pixel with the later driving order is determined as the target sub-pixel.

[0017] If the difference is less than the preset difference, the sub-pixel that is driven later in the driving order is determined as the target sub-pixel;

[0018] For example, the data line connects multiple adjacent columns of sub-pixels, and the connected multiple columns of sub-pixels correspond to different colors; wherein, the first sub-pixel and the second sub-pixel are located in the same row.

[0019] For example, the step of compensating for the original grayscale corresponding to the target sub-pixel in the target image data using the target model includes:

[0020] Based on the arrangement parameters, a target model corresponding to the arrangement parameters is obtained, and different arrangement parameters correspond to different target models;

[0021] The original grayscale corresponding to the target sub-pixel in the target image data is compensated using the target model corresponding to the arrangement parameters.

[0022] For example, a data line connects a first pixel group and a second pixel group that are adjacent in position and different in color. Before compensating the original grayscale of the target sub-pixels in the target image data using the target model, the method further includes: preprocessing the target image data; wherein the preprocessing includes:

[0023] Based on the arrangement parameters, the grayscale data of each pixel in the target image data in multiple color channels are recombined to obtain at least one reconstructed image. The reconstructed image includes the original grayscale corresponding to at least two color channels, and the two adjacent color channels correspond to the first pixel group and the second pixel group, respectively.

[0024] The step of compensating for the original grayscale of the target sub-pixels in the target image data using a target model includes:

[0025] The original grayscale of a portion of the color channels in each of the reconstructed images is compensated using the target model, and the compensated color channels are the target sub-pixels.

[0026] The first image data is the image data after preprocessing.

[0027] For example, the step of recombining the grayscale data of each pixel in the target image data across multiple color channels based on the arrangement parameters to obtain at least one reconstructed image includes:

[0028] Based on the original gray levels of each pixel in the target image data in different color channels, multiple color component maps are obtained, and each color component map includes the original gray levels of each pixel in the same color channel.

[0029] Based on the arrangement parameters, the second pixel group is transferred from the color component map where the second pixel group is located to the color component map where the first pixel group is located, so that the grayscale data corresponding to the first pixel group and the second pixel group are located in the same color component map, thereby obtaining the reconstructed image.

[0030] For example, the image includes multiple reconstructed images, each of which corresponds to two different color channels, and the color channels corresponding to different reconstructed images are not exactly the same.

[0031] Exemplarily, it includes a first reconstructed image, a second reconstructed image, and a third reconstructed image;

[0032] The first reconstructed image corresponds to the blue and red channels, the second reconstructed image corresponds to the green and red channels, and the third reconstructed image corresponds to the blue and green channels.

[0033] For example, after compensating the original grayscale of a portion of the color channels in each of the reconstructed images using the target model, the method further includes:

[0034] Based on the arrangement parameters, the compensated reconstructed image is restored to obtain the restored color component map;

[0035] The target image data is displayed based on the restored color component map.

[0036] The restored color component map includes the compensated grayscale of each pixel in the same color channel.

[0037] For example, the process of obtaining the target model includes:

[0038] The first image data is input into a preset network, which is used to compensate the original grayscale of the pixels in the first image data in some or all color channels to output predicted image data.

[0039] The predicted image data output by the preset network is obtained, and the parameters of the preset network are updated based on the difference between the predicted image data and the second image data.

[0040] For example, the dataset includes a first dataset and a second dataset, the first dataset corresponds to the first display screen, the second dataset corresponds to a plurality of second display screens, the first display screen is one of the second display screens, or the first display screen does not belong to the second display screens;

[0041] The process of obtaining the target model includes:

[0042] Based on the second dataset, deep learning is performed on the preset network to obtain an initial model;

[0043] Based on the first dataset, deep learning is performed on the initial model to obtain the target model adapted to the first display screen.

[0044] For example, the process of acquiring the second image data includes:

[0045] Based on the first image data and the arrangement parameters of the second display screen, the third sub-pixel to be compensated is determined; wherein, the second display screen is different from the first display screen, or the first display screen and the second display screen are the same display screen;

[0046] The third sub-pixel is compensated at least once until the termination condition is met, and the compensated image obtained when the termination condition is met is used as the second image data.

[0047] In each iteration of the compensation process, the grayscale of the third sub-pixel is increased or decreased according to a preset compensation value, and the first display brightness difference between the actual display brightness of the compensated third sub-pixel and the preset display brightness is determined. The preset display brightness is the ideal display brightness corresponding to the original grayscale of the third sub-pixel.

[0048] The termination condition includes: the first display brightness difference is less than the first preset brightness difference.

[0049] Exemplarily, the method further includes:

[0050] The second display brightness difference between the actual display brightness of each sub-pixel corresponding to the same color and gray level in the compensated first image data is determined.

[0051] The termination condition also includes: the second display brightness difference is less than the second preset brightness difference.

[0052] For example, the sub-pixels having the same color and grayscale include: the third sub-pixel and the uncompensated sub-pixels.

[0053] Exemplarily, the method further includes:

[0054] The first image data is divided into multiple image regions, and each image region corresponds to a sub-region in the display area of ​​the second display screen;

[0055] The step of performing at least one iteration of compensation on the third sub-pixel until the termination condition is met includes:

[0056] Perform at least one iteration compensation on the third sub-pixel within each sub-region until the termination condition is met;

[0057] The compensated image data of each of the sub-regions is stitched together when the termination condition is met to obtain the second image data.

[0058] For example, dividing the first image data into multiple image regions includes:

[0059] Obtain the resistivity variation curves of the signal lines on the second display screen at different regions of the display area;

[0060] Based on the resistivity change curve, the first image data is divided into multiple image regions; wherein, the signal lines include at least data lines.

[0061] Secondly, a screen compensation device is provided, the screen compensation device comprising:

[0062] The caching module is used to acquire target image data corresponding to the first display screen. The target image data includes the original grayscale of each pixel in different color channels. The different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen.

[0063] The compensation module is used to obtain the target sub-pixel to be compensated from multiple sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data, and to compensate the original grayscale corresponding to the target sub-pixel in the target image data.

[0064] The arrangement parameters include the driving structure of the sub-pixels, which characterizes the connection architecture between the data lines and each sub-pixel in the first display screen.

[0065] Thirdly, a display device is provided, comprising: a display panel, a data processing module, and a display driving module, wherein the data processing module is connected to the display driving module, and the display driving module is connected to the display panel; wherein...

[0066] The data processing module is used to execute the image quality compensation method for the display screen as described in any of the first aspects;

[0067] The display driving module is used to drive the display panel to display images based on the compensated target image data output by the data processing module.

[0068] For example, the display panel includes multiple gate lines and multiple data lines, and the sub-pixel is located in the area defined by the intersection of the gate lines and the data lines;

[0069] The display driving module includes a gate driving unit connected to the gate line and a data driving unit connected to the data line.

[0070] The data processing module includes a processing unit and a timing controller connected to the processing unit. The timing controller is connected to the gate driving unit and the data driving unit, respectively.

[0071] The processing unit is configured with a target model, which is used to execute the image quality compensation method for the display screen described in any of the first aspects.

[0072] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, specific embodiments of this disclosure are described below.

[0073] Brief description of the attached diagram

[0074] To more clearly illustrate the technical solutions in the embodiments or related technologies of this disclosure, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the scale in the drawings is for illustration only and does not represent the actual scale.

[0075] Figure 1 shows a schematic diagram of the planar structure of the array substrate;

[0076] Figure 2 shows the vertical stripes that appear under a certain pixel arrangement and scanning method;

[0077] Figures 3-6 show schematic diagrams of the pixel driving structure of several displays;

[0078] Figure 7 shows a flowchart of a method for image quality compensation of a display screen;

[0079] Figure 8 shows the correspondence between the pixels in the image represented by the target image data and the sub-pixels on the first display screen;

[0080] Figure 9 illustrates the process of determining the target sub-pixel;

[0081] Figure 10 shows a schematic diagram of the target model;

[0082] Figure 11 shows a schematic diagram of the image processing flow of the target model.

[0083] Figure 12 shows a schematic diagram of the process of obtaining the second image data;

[0084] Figure 13 shows a schematic diagram of the partition compensation process;

[0085] Figure 14 shows a schematic diagram of the regional division.

[0086] Figure 15 shows a schematic diagram of the preprocessing process for target image data;

[0087] Figure 16, using Figure 4 as an example, shows a schematic diagram of the target image data before and after preprocessing;

[0088] Figures 17-19 show schematic diagrams of the three target image data before and after preprocessing;

[0089] Figure 20 shows a schematic diagram of the image compensation device;

[0090] Figure 21 shows a schematic diagram of the display device.

[0091] Detailed description

[0092] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0093] In related technologies, displays include array substrates, such as liquid crystal displays and organic light-emitting diode displays. The array substrate includes multiple gate lines and multiple data lines. Please refer to Figure 1, which shows a schematic diagram of the planar structure of the array substrate. As shown in Figure 1, the gate lines and data lines intersect to define the area of ​​the sub-pixels, so that multiple sub-pixels can be arranged in multiple rows and columns. For example, the gate lines can connect multiple sub-pixels in the row direction, and the data lines can connect multiple sub-pixels in the column direction.

[0094] When driving the display, sub-pixels can be driven row by row to display the image. For example, as shown in Figure 1, multiple sub-pixels can be driven sequentially from top to bottom and left to right to display the image. Exemplarily, a gate drive signal can be input to the nth gate line, causing the thin-film transistor on the nth row of sub-pixels to be in a conducting state. Then, data voltages are input to multiple data lines, thereby applying voltage to each sub-pixel on the nth row, forming an electric field that drives the liquid crystal molecules to deflect, thus illuminating the nth row of sub-pixels. Afterward, a gate drive signal is sequentially input to the (n+1)th gate line to illuminate the (n+1)th row of sub-pixels.

[0095] In this way, multiple sub-pixels in the display screen are driven to light up sequentially.

[0096] However, fine lines often appear on displays. Taking LCD screens as an example, these lines are easily visible to the human eye, thus affecting image quality. Upon investigating the root causes of these fine lines, the inventors discovered the following fundamental reasons:

[0097] During pixel scanning, the charging rate of the previously scanned pixel affects the charging rate of the subsequently scanned pixel. For example, if two pixels in a previous scan need to display grayscale values ​​M and N (where M is much smaller than N), the screen pixel scan first illuminates the pixel with grayscale value M, and then illuminates the pixel with grayscale value N. Due to the influence of the previous pixel, the actual luminous brightness of the pixel unit corresponding to grayscale value N cannot reach the brightness that grayscale value N should originally reach, resulting in a difference in the brightness displayed by the subsequently scanned pixel, such as being lower than the required target brightness. If there are multiple pixels with grayscale value N in the same frame image, due to the influence of the previously scanned pixel unit, some pixels cannot reach the brightness that grayscale value N should originally reach, while other pixels can. This manifests as brightness and darkness variations in the image, appearing as fine lines.

[0098] In practice, to verify this phenomenon, the inventors used a solid color image for display and found that the fine lines were more obvious in the solid color image, as shown in Figure 2. Figure 2 shows a schematic diagram of the image quality under a solid color image using an LCD screen as the display. As shown in Figure 2, fine lines with varying brightness and darkness appear on the screen. This fully demonstrates that the charging degree of each pixel is affected by the previous scanned pixel. If the subsequent pixel is not charged enough, the brightness difference of the same sub-pixel of the same color and the same gray level will be large, thus resulting in fine lines with varying brightness and darkness.

[0099] It should be noted that Figure 2 shows vertical stripes (also known as longitudinal stripes) that appear under one pixel arrangement and scanning method. However, in some other displays, due to differences in their pixel arrangement and scanning methods, horizontal stripes may also appear.

[0100] In view of this, embodiments of this disclosure propose a display screen image quality compensation method, an image quality compensation device, and a display device. The image quality compensation method can compensate for the original grayscale corresponding to the affected target sub-pixels in the target image data based on the target image data to be displayed and the arrangement parameters of the first display screen to be displayed. The arrangement parameters may include the arrangement of sub-pixels of different colors and the driving structure of multiple sub-pixels. The target sub-pixel is a sub-pixel whose charging rate is affected by the previously lit sub-pixel. By compensating for the original grayscale, the display brightness difference caused by the influence of the charging rate of the previously lit sub-pixel on the target sub-pixel can be reduced, thereby improving the phenomenon of fine lines.

[0101] Furthermore, to improve the efficiency of fine line correction on various displays, the inventors proposed a deep learning-based image quality compensation scheme. This scheme uses a target model to perform end-to-end enhancement on the input target image data to compensate for fine lines, achieving a visually smooth, line-free effect. This allows for accurate compensation for different screens, screen areas, brightness levels, and colors. The target model not only improves fine line correction efficiency but also offers high versatility, adapting to fine line correction on a wide range of displays, thus expanding the application scope of the image quality compensation method in this embodiment.

[0102] The following description, using an LCD display screen as an example and in conjunction with the accompanying drawings, will describe a display screen image quality compensation method, image quality compensation device, and display device according to an embodiment of the present disclosure.

[0103] It should be noted that, in the embodiments, the red sub-pixels refer to the sub-pixels marked with R in the accompanying drawings, the green sub-pixels refer to the sub-pixels marked with G in the accompanying drawings, and the blue sub-pixels refer to the sub-pixels marked with B in the accompanying drawings.

[0104] Referring to Figures 3-5, Figures 3-6 show schematic diagrams of the pixel driving structure of several displays, and Figure 7 shows a flowchart of a display image quality compensation method. As shown in Figures 3-7, the display image quality compensation method in this embodiment mainly includes the following steps:

[0105] Step S101: Obtain the target image data to be displayed on the first display screen;

[0106] The target image data includes the original grayscale of each pixel in different color channels, and the different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen.

[0107] Step S102: Based on the arrangement parameters of each sub-pixel in the first display screen, obtain the target sub-pixel to be compensated from multiple sub-pixels, and compensate the original grayscale corresponding to the target sub-pixel in the target image data.

[0108] The arrangement parameters include the driving structure of the sub-pixels, which characterizes the connection architecture between the data lines and each sub-pixel in the first display screen.

[0109] In this embodiment, the arrangement parameters proposed in this embodiment will be explained first.

[0110] The arrangement parameters include the driving structure of the sub-pixels.

[0111] The driving structure is a connection architecture between the data lines and each sub-pixel in the first display screen. Currently, the driving structure can include a triple gate (TG), a single gate, and a double gate.

[0112] Figure 3 illustrates the arrangement of sub-pixels in a three-gate structure. In this driving structure, sub-pixels of the same color are arranged in a column, and the colors of multiple sub-pixels in the row direction are interleaved. One pixel unit corresponds to two data lines, one of which connects two adjacent columns of sub-pixels, and the other connects to the remaining column of sub-pixels in the pixel unit. As shown in Figure 3, one pixel unit corresponds to data lines S1 and S2. Data line S1 connects the blue and green sub-pixels in the pixel unit, and data line S2 connects the red sub-pixels in the pixel unit. Each row of sub-pixels corresponds to three gate lines, and each gate line connects sub-pixels of the same color in that row.

[0113] In the display shown in Figure 3, the driving process can be as follows: gate driving signals are sequentially input to multiple gate lines. After each gate driving signal is input to a gate line, data signals are input to each data line to illuminate multiple sub-pixels connected to that gate line. Since two sub-pixels connected to the same data line are connected to different gate lines, when two adjacent gate lines are driven, the same data line will input signals in the illumination cycles of two adjacent sub-pixels in the column direction. For example, when a gate driving signal is input to gate line G1, data line S1 inputs a data signal to the blue sub-pixel to illuminate the blue sub-pixel. When a gate driving signal is input to gate line G2, data line S1 inputs a data signal to the green sub-pixel to illuminate the green sub-pixel. In this way, the charging rate of the green sub-pixel is affected by the previously illuminated blue sub-pixel. Therefore, among two adjacent sub-pixels connected to the same data line, the previously illuminated sub-pixel will affect the charging rate of the subsequently illuminated sub-pixel.

[0114] Figure 4 shows a schematic diagram of the sub-pixel arrangement in a dual-gate structure. In this driving structure, sub-pixels of the same color are arranged in a column, and the colors of multiple sub-pixels in the row direction are staggered. Adjacent columns of sub-pixels are connected to the same data line. For example, data line S1 connects a column of blue sub-pixels and a column of green sub-pixels. Each row of sub-pixels corresponds to two gate lines. Two sub-pixels connected to the same data line are connected to different gate lines. For example, the blue and green sub-pixels connected to data line S1 are connected to gate lines G1 and G2, respectively. In this way, the gate lines connect sub-pixels of different colors spaced apart in the row direction.

[0115] In a dual-gate display, the driving process can be as follows: gate driving signals are sequentially input to multiple gate lines. After inputting a gate driving signal to each gate line, data signals are input to each data line to illuminate multiple sub-pixels connected to that gate line. Since two sub-pixels connected to the same data line are connected to different gate lines, when two adjacent gate lines are driven, the same data line will input signals in the illumination cycles of two adjacent sub-pixels in the column direction. As shown in Figure 4, when a gate driving signal is input to gate line G1, data line S1 inputs a data signal to the blue sub-pixel to illuminate the blue sub-pixel. When a gate driving signal is input to gate line G2, data line S1 inputs a data signal to the green sub-pixel to illuminate the green sub-pixel. Thus, the charging rate of the green sub-pixel is affected by the previously illuminated blue sub-pixel. Therefore, among two adjacent sub-pixels connected to the same data line, the previously illuminated sub-pixel will affect the charging rate of the subsequently illuminated sub-pixel.

[0116] Figure 5 shows a schematic diagram of the sub-pixel arrangement in a three-gate structure. In this driving structure, sub-pixels of the same color are arranged in a column, and the colors of multiple sub-pixels in the row direction are staggered. Adjacent columns of sub-pixels are connected to the same data line, and one data line connects three adjacent columns of sub-pixels. A row of sub-pixels corresponds to three gate lines, and each gate line connects a sub-pixel of one color in the row. For example, data line S1 connects adjacent blue, red, and green sub-pixels; gate line G1 connects the blue sub-pixel in a row, gate line G2 connects the green sub-pixel in a row, and gate line G3 connects the red sub-pixel in a row.

[0117] In a three-gate display, the driving process can be referenced from that of a two-gate structure. As shown in Figure 5, when a gate drive signal is input to gate line G1, data line S1 inputs a data signal to the blue sub-pixel to illuminate it; when a gate drive signal is input to gate line G2, data line S1 inputs a data signal to the green sub-pixel to illuminate it; and when a gate drive signal is input to gate line G3, data line S1 inputs a data signal to the red sub-pixel to illuminate it. Thus, the charging rate of the red sub-pixel is affected by the previously illuminated green sub-pixel, and the charging rate of the green sub-pixel is affected by the previously illuminated blue sub-pixel. Therefore, among three adjacent sub-pixels connected to the same data line, the sub-pixel illuminated first will affect the charging rate of the sub-pixel illuminated later.

[0118] Figure 6 shows a schematic diagram of the arrangement of sub-pixels under a single-gate structure. In this driving structure, sub-pixels of the same color are arranged in a row, and sub-pixels of different colors are arranged alternately in the column direction. A data line connects multiple sub-pixels in a column, and a row of sub-pixels corresponds to a gate line.

[0119] In a single-gate display, the driving process is as follows: when a gate driving signal is input to the i-th gate line, all sub-pixels in the i-th row are input to the gate driving signal. Then, data signals are input to multiple data lines, causing all sub-pixels in the i-th row to be lit. After that, after inputting a gate driving signal to the (i+1)-th gate line, data signals are again input to multiple data lines, causing all sub-pixels in the (i+1)-th row to be lit. As shown in Figure 6, when gate driving signals are input to gate lines G1, G2, and G3, the sub-pixels in the i-th column are always provided with signals by data line S1. Therefore, among two adjacent sub-pixels in the i-th column, the sub-pixel that is lit earlier will affect the sub-pixel that is lit later. For example, the blue pixel in the i-th column will affect the charging rate of the green pixel, and the charging rate of the green pixel may affect the charging rate of the red pixel.

[0120] The image quality compensation method in this embodiment can be adapted to any of the above-mentioned driving structures.

[0121] As shown in Figures 3-6 above, the driving structure defines the connection method between the data lines and the sub-pixels, thereby defining the arrangement of sub-pixels of different colors in the display area. This includes sub-pixels of the same color arranged in a row, and sub-pixels of different colors arranged alternately in the column direction. As shown in Figure 6, sub-pixels of the same color are arranged in a row, and the colors of multiple sub-pixels in the column direction are alternated, so that every two adjacent rows of sub-pixels correspond to different colors.

[0122] In one example, there is also a case where subpixels of the same color are arranged in a column, and the colors of multiple subpixels in the row direction are interleaved, so that each pair of adjacent columns of subpixels correspond to different colors, as shown in Figures 3 and 5.

[0123] When the first display screen needs to display an image, it can acquire the target image data to be displayed on the first display screen. The target image data can be image data of a solid color image or image data of a color image, without limitation.

[0124] The target image data may include the grayscale values ​​of each pixel in different color channels. As shown in Figure 8, the target image data represents the correspondence between pixel PA in the image and each sub-pixel P on the first display screen. As shown in Figure 8, a pixel PA in the image corresponds to a pixel unit PC on the first display screen. The target image data stores the grayscale values ​​of the pixel in different color channels. Thus, one color channel of a pixel PA corresponds to a sub-pixel P in the pixel unit.

[0125] For example, if a pixel in the image includes a blue channel, a green channel, and a red channel, then the blue channel, green channel, and red channel of the pixel correspond to the blue sub-pixel, green sub-pixel, and red sub-pixel of the pixel unit at the corresponding position in the first display screen, respectively.

[0126] When driving the display screen, the pixel value of a pixel is split into grayscale values ​​in the blue channel, green channel, and red channel. This drives the blue, green, and red subpixels of the corresponding pixel unit to display the corresponding brightness at the corresponding grayscale values, thereby achieving the color display of the pixel.

[0127] In step S102, based on the arrangement of sub-pixels of different colors in the first display screen and the driving structure of the sub-pixels, multiple sub-pixels whose charging rates are mutually affected during driving can be determined. It should be noted that if multiple mutually affected sub-pixels are connected to the same data line, then multiple sets of mutually affected sub-pixels can be determined in the first display screen.

[0128] Among them, multiple sub-pixels that affect each other can be located in N adjacent columns of sub-pixels or in N adjacent rows of sub-pixels, where N can be a positive integer greater than or equal to 2.

[0129] For example, as shown in FIG3, the blue sub-pixel and the green sub-pixel connected to the data line S1 are a group of mutually influential sub-pixels;

[0130] As shown in Figure 4, the blue and green sub-pixels connected to data line S1 form a group of mutually influential sub-pixels, and the red and blue sub-pixels connected to data line S2 form a group of mutually influential sub-pixels.

[0131] As shown in Figure 5, the three columns of sub-pixels connected to the data line S1 are a group of mutually influential sub-pixels, that is, the blue sub-pixels, green sub-pixels and red sub-pixels adjacent in the row direction are mutually influential sub-pixels.

[0132] As shown in Figure 6, in a column of sub-pixels connected to data line S1, two adjacent sub-pixels are mutually influential sub-pixels. For example, the blue sub-pixel connected to gate line G1 and the green sub-pixel connected to gate line G2 are a group of mutually influential sub-pixels.

[0133] Next, the difference between the original grayscale values ​​of each group of mutually influencing sub-pixels can be obtained from the target image data. Based on this difference, it can be determined whether some sub-pixels of the multiple sub-pixels need to be compensated. If compensation is required, the target sub-pixel to be compensated can be determined from the multiple mutually influencing sub-pixels.

[0134] In an exemplary embodiment, the target sub-pixel can be a sub-pixel whose charging rate is affected among a plurality of mutually influential sub-pixels. Taking FIG3 as an example, the blue sub-pixel and the green sub-pixel connected to the data line S1 are mutually influential sub-pixels. The charging rate of the green sub-pixel is affected by the blue sub-pixel, so the green sub-pixel can be determined as the sub-pixel to be compensated.

[0135] In compensating for target sub-pixels, the original grayscale of the target sub-pixel in the target image data can be compensated, such as by increasing or decreasing the original grayscale of the target sub-pixel. The amount of compensation can be determined based on the difference between the original grayscale of the target sub-pixel and the original grayscale of other sub-pixels that affect the charging rate of the target sub-pixel. For example, if the difference is large and the impact on the charging rate of the target sub-pixel is high, the compensation amount can be large; if the difference is small and the impact on the charging rate of the target sub-pixel is low, the compensation amount can be small.

[0136] Taking Figure 4 as an example, assuming that the target image data needs to display a solid color image, then different pixels have the same original grayscale value in the same color channel. For example, the original grayscale value of pixel A in the blue channel is the same as that of pixel B in the blue channel. In this way, in the first display screen, sub-pixels of the same color need to correspond to the same display brightness, such as green sub-pixels corresponding to the same display brightness.

[0137] As shown in Figure 4, because the charging rate of some green sub-pixels is affected by the blue sub-pixels, the actual display brightness of the green sub-pixels differs from the ideal display brightness corresponding to the original grayscale. For example, if the actual display brightness of the green sub-pixels is darker than the ideal display brightness, fine lines will appear on the first display screen. The compensation method of this embodiment can compensate for the original grayscale of the green sub-pixels. This compensation can offset the influence of the charging rate of the blue sub-pixels, thus presenting a brightness that is close to or even consistent with the ideal display brightness, thereby improving the fine lines in that area.

[0138] It should be noted that if, based on the difference between the original grayscale values ​​of a group of mutually influencing sub-pixels, it is determined that no compensation is needed for that group of mutually influencing sub-pixels, then no compensation is required.

[0139] In this embodiment, based on the arrangement parameters of the first display screen and the target image data, the target sub-pixel to be compensated can be determined among multiple sub-pixels whose charging rates affect each other. The original grayscale of the target sub-pixel is then compensated. This allows for compensation of the original grayscale of the target sub-pixel in the target image data, offsetting the influence of charging rates among multiple sub-pixels, for example, offsetting the influence of other sub-pixels on the charging rate of the target sub-pixel. Consequently, the difference between the displayed brightness of each sub-pixel and the required actual brightness can be reduced, avoiding uneven screen brightness and the resulting fine lines.

[0140] In some embodiments, when determining the target sub-pixel to be compensated based on the arrangement parameters and target image data, it can be based on selecting from two sub-pixels connected to the same data line and adjacent in the driving sequence. As shown in Figure 5, a data line connects three columns of sub-pixels, so the target sub-pixel to be compensated can be selected from two adjacent columns of sub-pixels. As shown in Figure 5, the target sub-pixel to be compensated is selected from the blue sub-pixel and the green sub-pixel.

[0141] In one example, the target sub-pixel to be compensated can be determined from the two sub-pixels connected to the same data line and driven later in the driving order, as shown in Figure 5. A data line connects three columns of sub-pixels, and the red and green sub-pixels are driven later in the driving order, so the target sub-pixel can be determined from the red and green sub-pixels.

[0142] For example, please refer to Figure 9, which illustrates the process of determining the target sub-pixel. As shown in Figure 9, the target sub-pixel can be determined according to the following steps:

[0143] Step S21: Based on the arrangement parameters, determine the first and second sub-pixels that are connected to the same data line and are adjacent from multiple sub-pixels;

[0144] Step S22: Obtain the difference between the original gray levels corresponding to the first sub-pixel and the second sub-pixel in the target image data;

[0145] Step S23: Based on the difference and the driving order of the first and second sub-pixels, determine the target sub-pixel in the first and second sub-pixels.

[0146] In step S21, based on the arrangement of multiple sub-pixels of different colors on the first display screen and the driving structure of the pixels, the first sub-pixel and the second sub-pixel that are connected to the same data line and are adjacent can be obtained.

[0147] As shown in Figures 3-6, according to the driving structure, the first and second sub-pixels that are connected to the same data line and are adjacent can be two sub-pixels with different colors and adjacent driving order.

[0148] In this context, the first sub-pixel and the second sub-pixel can belong to the same pixel unit. Thus, the first sub-pixel and the second sub-pixel can correspond to the same pixel in the target image data. Therefore, the first sub-pixel and the second sub-pixel can be different color channels of a pixel in the target image data.

[0149] For example, as shown in FIG3, if the blue sub-pixel (first sub-pixel) and the green sub-pixel (second sub-pixel) connected to the data line S1 are located in the same pixel unit, then they correspond to the blue channel and green channel of the same pixel point in the target image data.

[0150] The first sub-pixel and the second sub-pixel can also belong to different pixel units, so that the first sub-pixel and the second sub-pixel can correspond to different color channels of two pixels in the target image data.

[0151] For example, as shown in FIG3, the red sub-pixel (first sub-pixel) and the blue sub-pixel (second sub-pixel) connected to the data line S2 are located in different pixel units, which correspond to the blue channel and red channel of two adjacent pixels in the target image data.

[0152] The first display screen includes multiple groups of first sub-pixels and second sub-pixels. In some groups, the first sub-pixels and second sub-pixels belong to the same pixel unit, while in other groups, the first sub-pixels and second sub-pixels belong to different pixel units.

[0153] In this context, the first and second sub-pixels can be the later-driving sub-pixel in a group of sub-pixels connected to the same data line, located adjacently, and with different colors. This reduces the difficulty of grayscale compensation for sub-pixels.

[0154] In an exemplary embodiment, the first sub-pixel and the second sub-pixel can be connected to the same data line, and the first sub-pixel and the second sub-pixel can be adjacent sub-pixels located in the same column or adjacent sub-pixels located in the same row.

[0155] As shown in Figures 3-5, in the driving structures of the dual-gate and triple-gate structures, a data line can connect at least two adjacent columns of sub-pixels. Thus, the first sub-pixel and the second sub-pixel can be adjacent sub-pixels located in the same row, or they can be adjacent sub-pixels located in different rows.

[0156] For example, as shown in FIG3, data line S1 connects the blue sub-pixel in column i and the green sub-pixel in column i+1. The first sub-pixel and the second sub-pixel are driven in adjacent order and are located adjacently. Then the first sub-pixel and the second sub-pixel can be the green sub-pixel in row n and column i+1 and the blue sub-pixel in row n+1 and column i, that is, two sub-pixels located on the diagonal.

[0157] For example, as shown in Figure 3, data line S1 connects the blue sub-pixel in column i and the green sub-pixel in column i+1. The first sub-pixel and the second sub-pixel are driven sequentially and are located adjacently. Therefore, the first sub-pixel and the second sub-pixel can be the blue sub-pixel in row n and column i and the green sub-pixel in row n and column i+1. This simplifies the subsequent calculations and improves the compensation efficiency.

[0158] As shown in Figure 6, in the driving structure of a single-gate structure, a data line connects sub-pixels in the same column, and different data lines connect sub-pixels in different columns. In this case, the first sub-pixel and the second sub-pixel are both located in the same column.

[0159] In step S22, the original grayscale of the first sub-pixel and the original grayscale of the second sub-pixel can be determined in the target image data. As described above, when the first sub-pixel and the second sub-pixel belong to the same pixel, the original grayscale of the corresponding pixel in the color channel corresponding to the first sub-pixel and the second sub-pixel can be determined in the target data. When the first sub-pixel and the second sub-pixel belong to different pixel, the original grayscale of the two corresponding pixels in the color channel corresponding to the first sub-pixel and the second sub-pixel can be determined in the target data.

[0160] For example, as shown in Figure 3, the red sub-pixel and the blue sub-pixel connected to the data line S2 are located in different pixel units. Assuming they are located in pixel unit A and pixel unit B, the original grayscale of the red channel of pixel A corresponding to pixel unit A and the original grayscale of the blue channel of pixel B corresponding to pixel unit B can be obtained in the target image data.

[0161] In step S23, the difference between the original grayscale of the first sub-pixel and the original grayscale of the second sub-pixel can be determined, and based on the difference and the driving order of the first and second sub-pixels, the target sub-pixel to be compensated is determined in the first and second sub-pixels.

[0162] The driving sequence refers to the same data line. For example, in the case of the first sub-pixel and the second sub-pixel connected to the same data line, the second sub-pixel lights up after the first sub-pixel lights up; or the first sub-pixel lights up after the second sub-pixel lights up.

[0163] In an exemplary embodiment, the need for compensation can be determined based on the magnitude of the difference. For example, if the difference indicates that the original grayscale of the later-driving sub-pixel is lower than that of the earlier-driving sub-pixel, then because the original grayscale of the later sub-pixel is very low, no compensation is needed for the first and second sub-pixels. If the difference is greater than a preset difference, and the original grayscale of the later-driving sub-pixel is higher than that of the first and second sub-pixels, then the later-driving sub-pixel is determined as the target sub-pixel; if the difference is less than a preset difference, then the later-driving sub-pixel is determined as the target sub-pixel.

[0164] The preset difference can be set according to requirements. For example, it can be set to a value of 10 to 20. When the difference is greater than the preset difference, it indicates that the driving signal difference between the first sub-pixel and the second sub-pixel is large. As a result, the sub-pixel with the earlier driving order has a greater impact on the charging rate of the sub-pixel with the later driving order. Therefore, the sub-pixel with the later driving order can be compensated, that is, the sub-pixel with the later driving order is used as the target sub-pixel.

[0165] For example, as shown in FIG4, among the blue sub-pixels and green sub-pixels located in the same row connected to the data line S1, the driving order of the blue sub-pixels is before that of the green sub-pixels. Therefore, during compensation, the green sub-pixels can be compensated.

[0166] Thus, in dual-gate and triple-gate structures, some sub-pixels connected to the same data line and located in the same column may be compensated, while others may not. In a single-gate structure, some sub-pixels connected to the same gate line and located in the same row may be compensated, while others may not.

[0167] In an exemplary embodiment, to improve the efficiency of image quality compensation and expand its application scope, a target model can be used to perform end-to-end compensation on the input target image data to improve fine lines. This target model can be obtained by performing deep learning on a preset network using multiple datasets as training samples. Specifically, the target model can compensate for the original grayscale of each pixel in the target image data across some or all color channels, where the compensated color channels are determined based on arrangement parameters.

[0168] Please refer to Figures 10 and 11. Figure 10 shows a schematic diagram of the target model's structure, and Figure 11 shows a schematic diagram of the image processing flow of the target model. The target model may include a model using the U-Net paradigm, comprising an encoder and a decoder. The encoder may include a first feature extraction module, a second feature extraction module, a third feature extraction module, a fourth feature extraction module, and a fifth feature extraction module connected in series. The decoder may include a first decoding module, a second decoding module, a third decoding module, a fourth decoding module, and a fifth decoding module connected in series. The first feature extraction module is connected to the fifth decoding module, the second feature extraction module is connected to the fourth decoding module, the third feature extraction module is connected to the third decoding module, the fourth feature extraction module is connected to the second decoding module, and the fifth feature extraction module is connected to the first decoding module.

[0169] In this embodiment, each feature extraction module in the encoder can consist of a 3*3 convolutional layer, a ReLU activation function, and downsampling to learn and extract image features from the target image data. Each decoding module in the decoder can consist of upsampling, a 3*3 convolutional layer, and a ReLU activation function. As shown in Figure 11, skip connections are established between the encoder and decoder layers to combine semantic information from lower and higher layers, reducing the loss of detailed information.

[0170] In this embodiment, the feature extraction module in the encoder is used to extract the original grayscale of each pixel in each color channel in the target image data. The feature extraction module can determine the grayscale difference of each pixel in each color channel, i.e. the color brightness difference, based on the extracted feature map. Then, the original grayscale of some pixels in some color channels can be compensated based on the color brightness difference, and then the decoder decodes and outputs the result.

[0171] In practical use, the target image data can be input into the target model, and then the target model can extract and compensate the features of the target image data, and then output the compensated target image data.

[0172] The dataset used to train the target model includes first image data and second image data. The driving structure of the display screen targeted by the first image data and the target image data can be the same. The first image data can be called the original image data. The second image data can be obtained by compensating the original grayscale of some sub-pixels in the first image data according to the driving structure. The compensation process of the compensated sub-pixels can refer to the process of step S102 above.

[0173] In this way, the process of training the target model can be carried out by using the second image data as supervision, constructing a loss function, and continuously updating the parameters of the preset network to obtain the target model.

[0174] In an exemplary embodiment, the process of obtaining the target model can be as follows:

[0175] First, the first image data is input into a preset network. The preset network is used to compensate the original grayscale of the pixels in some or all color channels in the first image data in order to output predicted image data.

[0176] Next, the predicted image data output by the preset network is obtained, and the parameters of the preset network are updated based on the difference between the predicted image data and the second image data.

[0177] Specifically, when updating the parameters of the preset network based on the difference between the predicted image and the second image data, the difference between the grayscale value of each color channel of each pixel in the predicted image and the grayscale value in the second image data can be determined. A loss function is constructed based on this difference, and then the loss value is obtained. The parameters of the preset network are updated based on the loss value. After multiple updates, if the loss value is less than the preset loss value, or if the preset network begins to converge, training can be stopped, and the preset network at the point of training stoppage can be used as the target model.

[0178] The technical solution of this embodiment, by compensating the target image data through a target model, can improve compensation efficiency. In practice, it only needs to be configured on the display screen for use. Furthermore, since the target model is based on deep learning, it can be applied to screens of different sizes, as well as different areas of the same screen, and images of different colors and brightness levels, thus improving the accuracy of fine line compensation and expanding its applicability.

[0179] In an exemplary embodiment, since display screens with different driving structures have different pixel arrangements, data lines and connection methods of multiple sub-pixels, the target sub-pixels to be compensated may be different. Therefore, in order to improve the accuracy of image quality compensation, independent target models can be constructed for display screens with different driving structures.

[0180] In this way, when compensating for target sub-pixels in target image data based on the target model, the target model corresponding to the arrangement parameters can be obtained from multiple models based on the arrangement parameters, and the target model can be used to compensate for the target sub-pixels in the target image data.

[0181] For example, as shown in Figures 3-6, a targeted dataset can be prepared for the dual-grid display shown in Figure 3, and a target model 1 can be trained; a targeted dataset can be prepared for the dual-grid display shown in Figure 4, and a target model 2 can be trained; a targeted dataset can be prepared for the triple-grid display shown in Figure 5, and a target model 3 can be trained; and a targeted dataset can be prepared for the single-grid display shown in Figure 6, and a target model 4 can be trained.

[0182] When the first display screen is arranged in a double-grid structure as shown in Figure 4, the target image data can be compensated using target model 2. When the first display screen is arranged in a single-grid structure as shown in Figure 6, the target image data can be compensated using target model 4.

[0183] This allows displays with different driving structures to have a suitable target model, thereby improving the accuracy of image quality compensation.

[0184] To further improve the accuracy of image quality compensation for individual display screens, the model training process can include a pre-training phase using a public dataset and a training phase using a screen tone supplement dataset. Referring to Figure 19, which illustrates the process of obtaining the target model, during the public dataset pre-training phase, a second dataset of displays of different sizes and types with the same driving structure can be collected, and an initial model can be obtained using this second dataset. Furthermore, during the screen tone supplement dataset phase, a first dataset of the display screen to be used can be collected, and the initial model can be updated using this first dataset to obtain a target model suitable for the current display screen.

[0185] Specifically, the first dataset corresponds to the first display screen, the second dataset corresponds to multiple second display screens, and the first display screen can be one of the second display screens, or the first display screen may not belong to any of the second display screens.

[0186] For example, assuming the second dataset targets multiple second displays with a single-grid structure, the size, shape, resolution, and pixel characteristics of these second displays can vary. Therefore, the resolution and size of the first image data can also vary. The first dataset can be collected separately for one of the multiple second displays, or it can be collected for a new display that also has a single-grid structure but is not part of the second displays.

[0187] The first dataset includes first image data corresponding to the first display screen and second image data corresponding to the first image data. The second dataset also includes first image data corresponding to the second display screen and second image data corresponding to the first image data.

[0188] Using the technical solution of this embodiment, the initial model is trained based on the second dataset. The initial model can compensate for multiple second displays and achieve a good compensation state in most areas of most second displays. The parameters of the initial model can achieve good compensation in most areas of most screens. After one training, it can be used for all second displays.

[0189] In the screen patch supplementation dataset stage, the first dataset can be based on the second dataset. During this process, training can be performed based on the first dataset, which can quickly achieve the convergence of the initial model. The initial model after this training can achieve good compensation effect in all areas of the first display screen.

[0190] The following explains how to obtain the second image data.

[0191] In one exemplary embodiment, please refer to FIG12, which illustrates a schematic diagram of the process of acquiring the second image data. As shown in FIG12, the process may specifically include the following steps:

[0192] Step S11: Based on the first image data and the arrangement parameters of the second display screen, determine the third sub-pixel to be compensated;

[0193] Step S12: Perform at least one iteration of compensation on the third sub-pixel until the termination condition is met, and use the compensated image obtained when the termination condition is met as the second image data;

[0194] In each iteration of compensation, the grayscale of the third sub-pixel is increased or decreased according to the preset compensation value, and the first display brightness difference between the actual display brightness of the compensated third sub-pixel and the preset display brightness is determined.

[0195] The termination condition includes: the first display brightness difference is less than the first preset brightness difference.

[0196] As shown in Figure 12, when preparing the second image data, it can be done based on the second display screen. That is, the first image data input to the second display screen is used as the original data. The first image data is compensated according to the arrangement parameters of the second display screen, and the compensated first image data is displayed on the second display screen. Based on the difference between the actual display brightness of the third sub-pixel and the preset display brightness when the compensated first image data is displayed, it can be determined whether the third sub-pixel has received brightness compensation.

[0197] Specifically, it can be determined whether the third sub-pixel has received brightness compensation by judging the relationship between the first display brightness difference and the first preset brightness difference. If the first display brightness difference is less than the first preset brightness difference, it is determined that the third sub-pixel has received brightness compensation; otherwise, it is determined that the third sub-pixel has not received brightness compensation and grayscale compensation needs to be performed on the third sub-pixel.

[0198] In cases where the first display brightness difference is greater than or equal to the first preset brightness difference, the gray level of the third sub-pixel can be increased or decreased according to the preset compensation value. For example, if the original gray level of the third sub-pixel is increased according to the preset compensation value during the first compensation, then the preset compensation value can be added to the gray level of the third sub-pixel obtained during the second compensation, and so on, until the termination condition is met.

[0199] Of course, if the original gray level of the third sub-pixel is reduced by the preset compensation value during the first compensation, then during the second compensation, the preset compensation value can be reduced on the gray level of the third sub-pixel obtained in the first compensation, and so on, until the termination condition is met.

[0200] The preset compensation value can be set to a relatively small value, such as 1 to 10, which can improve the accuracy of the compensation.

[0201] Of course, in some other examples, the preset compensation value used for each compensation may not be the same. For example, the preset compensation value for the first k compensations may be greater than the preset compensation value for the (k+1)th and subsequent compensations. This allows for increased compensation efficiency in the early stages of debugging and improved compensation precision in the later stages of debugging.

[0202] In another example, a second display brightness difference can also be determined between the actual display brightness of sub-pixels corresponding to the same color and gray level in the compensated first image data. Then, the termination condition can also include the second display brightness difference being less than a second preset brightness difference.

[0203] In this embodiment, the uniformity of the brightness of the compensated first image data can be comprehensively evaluated by the second display brightness difference, so that when compensating for the mutually influencing sub-pixels, the brightness of sub-pixels with consistent color and grayscale is ensured to be uniform throughout the entire image.

[0204] Therefore, by continuously adjusting the compensation value of the third sub-pixel, the phenomenon of brightness difference caused by the charging rate of the previous sub-pixel to the next sub-pixel can be eliminated, while ensuring the uniformity of brightness of sub-pixels of the same color and gray level in the image.

[0205] When the above termination condition is met, the first image data that meets the termination condition can be used as the second image data.

[0206] The second preset brightness difference can be the same as the first preset brightness difference, or the second preset brightness difference can be slightly greater than the first preset brightness difference, or the second preset brightness difference can be slightly less than the first preset brightness difference.

[0207] In this embodiment, sub-pixels corresponding to the same color and having the same grayscale can refer to pixels in the compensated first image data that have the same grayscale in the same color channel.

[0208] In an exemplary embodiment, sub-pixels of the same color and with the same grayscale can be sub-pixels of the same color and with the same grayscale among a plurality of third sub-pixels being compensated. For example, as shown in FIG3, if the third sub-pixel being compensated is the green sub-pixel in the (i+1)th column, then sub-pixels of the same color and with the same grayscale can be sub-pixels of the green sub-pixel in the (i+1)th column that have the same grayscale after compensation. For example, if the green sub-pixel in the nth row and the (i+1)th column and the green sub-pixel in the (n+2)th row and the (i+1)th column have the same grayscale after compensation, then the difference in the second display brightness of the two green sub-pixels can be compared.

[0209] In one exemplary embodiment, sub-pixels of the same color and grayscale can include sub-pixels of the same color and grayscale from a plurality of compensated third sub-pixels, and sub-pixels of the same color and grayscale from a plurality of uncompensated sub-pixels. Since compensation can be determined based on the difference in original grayscale between two mutually affecting sub-pixels when compensating for the original grayscale of each sub-pixel in the first image data according to the arrangement parameters and the first image data, in some cases, compensation for the affected sub-pixels may not be necessary. For example, if the original grayscale of the first illuminated sub-pixel is higher than that of the second illuminated sub-pixel in two mutually affecting sub-pixels, then the second illuminated sub-pixel may not need to be compensated.

[0210] In this way, some sub-pixels in the same column of sub-pixels connected to the same data line are compensated, while some sub-pixels may not be compensated. Therefore, not all sub-pixels of the same color are compensated; thus, sub-pixels of the same color may include compensated third sub-pixels as well as uncompensated sub-pixels.

[0211] For example, as shown in Figure 3, assuming the compensated third sub-pixel is the green sub-pixel in row n and column i+1, and the uncompensated sub-pixel is the green sub-pixel in row n+1 and column i+1, then in reality, there are multiple uncompensated green sub-pixels and multiple compensated green sub-pixels in the second display screen. Therefore, when determining the second display brightness difference, the actual display brightness difference between compensated and uncompensated green sub-pixels with the same grayscale can be determined. In this case, "same original grayscale" means that the original grayscale of the third sub-pixel before compensation is the same as the original grayscale of the uncompensated sub-pixel. As shown in Figure 3, the original grayscale of the green sub-pixel in row n and column i+1 is the same as the original grayscale of the green sub-pixel in row n+1 and column i+1, but the green sub-pixel in row n+1 and column i+1 is not compensated. Therefore, the actual display brightness difference between the green sub-pixel in row n and column i+1 and the green sub-pixel in row n+1 and column i+1 will be calculated, which is the second display brightness difference.

[0212] In this way, based on the difference in the second display brightness, and with the display brightness of the third sub-pixel to be compensated as a reference, it can be determined whether the compensation result for the third sub-pixel has eliminated the influence of the previously lit sub-pixel on its charging rate.

[0213] In this embodiment, the third sub-pixel that is compensated in the first image data is determined based on the arrangement parameters of the second display screen and the original grayscale corresponding to each sub-pixel in the first image data. The specific process can be referred to the description of step 102 above.

[0214] In this embodiment, the fact that the first display brightness difference is less than the first preset brightness difference represents the compensation result for the third sub-pixel.

[0215] In one exemplary embodiment, when determining the display brightness difference, an image can be acquired from the second display screen, thereby determining the display brightness difference between sub-pixels of the same color and grayscale based on the acquired image. For example, an image acquisition device can be used to acquire a brightness image of the second display screen when displaying compensated first image data; then, based on the brightness image, the display brightness difference between sub-pixels of the same color and grayscale can be determined.

[0216] The image acquisition device can be a point-and-shoot luminance meter or an imaging luminance meter, which can acquire the display brightness of each sub-pixel in the second display screen. Alternatively, the image acquisition device can be a high-resolution camera.

[0217] In an exemplary embodiment, in order to improve the calibration efficiency of the second image data and the accuracy of brightness acquisition, the first image data can be divided into multiple image regions, each image region corresponding to a sub-region in the display area of ​​the second display screen; thus, when iteratively compensating the third sub-pixel, each sub-region can be compensated separately, such as performing at least one iterative compensation on the third sub-pixel in each sub-region until the termination condition is met; finally, the compensated image data corresponding to each sub-region when the termination condition is met are stitched together to obtain the second image data.

[0218] In this embodiment, each image area corresponds to a sub-area of ​​the display area of ​​the second display screen. That is, when displaying, the image of the image area will be displayed in the corresponding sub-area.

[0219] In this embodiment, the second display screen is the display screen used when preparing the second dataset and the first dataset. The second display screen can be a display screen different from the first display screen, or it can be the first display screen.

[0220] In this embodiment, the division of the second image data can be based on the size of the second display screen. If the size is large, in order to improve the acquisition accuracy, it can be divided into more image regions, so that the size of each image region is within the acquisition range of the image acquisition device.

[0221] As shown in Figures 13 and 14, Figure 13 shows a schematic diagram of the zoning compensation process, and Figure 14 shows a schematic diagram of the region division.

[0222] As shown in Figure 13, when dividing the second image data, it can be divided into multiple rectangular image regions, such as a nine-grid division. In an exemplary embodiment, the size of each image region can be the same, such as multiple sub-pixels of each image region being an N*M matrix, as shown in (1) and (3) of Figure 14. In an exemplary embodiment, the sizes of the multiple image regions can not be completely the same, as shown in (2) of Figure 14.

[0223] In an exemplary embodiment, when dividing the second image data, the division can be based on the resistivity change curve of the signal line on the second display screen at different areas of the display area, dividing the first image data into multiple image areas; and / or, according to the resolution of the second display screen and the acquisition resolution of the image acquisition device, dividing the first image data into multiple image areas.

[0224] Signal lines include at least data lines.

[0225] In this embodiment, the second display screen may include multiple signal lines, which may include data lines, gate lines, and common electrode lines. A resistivity change curve can be plotted based on the resistivity of any one of the signal lines in different areas of the display area. Based on the resistivity change curve, the signal voltage drop in different areas of the display area can be determined, thereby determining the signal attenuation degree in each different area. Areas with approximately the same signal attenuation degree can be divided into the same sub-area.

[0226] For example, as shown in Figure 14(2), the signal voltage drop is generally larger in the middle area of ​​the display area. Therefore, the middle area can be used as a sub-area when dividing the area. Thus, the display area can be divided into multiple sub-areas from the inside to the outside. The middle area is a rectangular or circular area, and multiple ring-shaped sub-areas can be transitioned from the middle area to the non-display area.

[0227] In this way, when compensating for the first image data, the signal attenuation of the signal line can be taken into account, so that the image quality compensation not only compensates for the influence of the charging rate of the previous sub-pixel on the next sub-pixel, but also compensates for the influence of signal attenuation on the display brightness of the sub-pixel.

[0228] In an exemplary embodiment, the display area can be divided according to the resistivity change curve of the gate line, or according to the resistivity change curve of the data line, or according to the resistivity change curve of the common electrode line, or according to the resistivity change curves of any two of the gate line, data line and common electrode line; or, the display area can be divided according to the resistivity change curves of the gate line, data line and common electrode line.

[0229] In this exemplary embodiment, the first image data can also be divided into multiple image regions based on the resolution of the second display screen and the acquisition resolution of the image acquisition device.

[0230] For example, if the resolution of the second display screen is high and the acquisition resolution of the image acquisition device is low, the first image data can be divided into K1 image regions. If the resolution of the second display screen is high and the acquisition resolution of the image acquisition device is high, the first image data can be divided into K2 image regions, where K2 can be less than K1.

[0231] In practice, the higher the resolution of an image acquisition device, the more refined its acquisition, and the more brightness of each pixel can be captured; thus, the size of the sub-regions can be larger.

[0232] In another example of this embodiment, the compensation process for the first image data described above can be coded and burned into an electronic device to complete automated data acquisition. For example, the process shown in Figure 12, i.e., the debugging process of the first image data, can be written into a computer-readable program. This computer-readable program can be written into the image acquisition device, the SOC (System-on-a-Chip) of the second display screen, and the computer device. Through the program burned into the SOC, the third sub-pixel in the first image data can be compensated, and the gate driver, source driver, and timing controller on the second display screen can be used to complete the display of the compensated first image data.

[0233] Next, the program burned into the image acquisition device can call the micro camera and the second display screen to play the image, complete the brightness acquisition and upload;

[0234] Next, the computer device runs the burned program, receives the brightness information collected by the micro camera, completes brightness statistics, brightness difference calculation, etc., and controls the SOC to continue compensation or stop compensation based on the brightness difference results. This process is repeated to achieve automated debugging of the dataset.

[0235] In an exemplary embodiment, during the training of the target model, in order to improve the training speed of the model and the processing efficiency of the target model in inference applications, the target image data input to the target model can be preprocessed to make the target image data consistent with the arrangement parameters of the first display screen.

[0236] For example, referring to Figures 15 and 16, Figure 15 shows a schematic diagram of the preprocessing of target image data, and Figure 16, using Figure 4 as an example, shows a schematic diagram of the target image data before and after preprocessing. As shown in Figures 15 and 16, taking a first display screen with a dual-grid structure and a triple-grid structure as an example, the first display screen includes data lines extending along the column direction. One data line connects two adjacent columns of first and second pixel groups with different colors. When preprocessing the target image data, the grayscale data in multiple color channels of the target image data can be recombined based on the arrangement parameters to obtain at least one reconstructed image.

[0237] The reconstructed image includes the original grayscale corresponding to at least two color channels, and the two adjacent color channels correspond to the first pixel group and the second pixel group, respectively.

[0238] Accordingly, when compensating for the original grayscale of the target sub-pixels in the target image data using the target model, the original grayscale of the target sub-pixels in each reconstructed image can be compensated using the target model; wherein, the first image data is also the image data after the above preprocessing.

[0239] Recombination can be understood as processing the color component map of the target image data into an image whose color channel arrangement is consistent with the arrangement of sub-pixels of different colors on the first display screen.

[0240] The grayscale data of the color channels can include the original grayscale values ​​of the color channels.

[0241] In this embodiment, in the first display screen with a single-gate structure, as shown in FIG6, each sub-pixel in the first pixel group is located in the same row, and each sub-pixel in the second pixel group is located in the same row; in the first display screen with a dual-gate structure, as shown in FIG3-FIG4, each sub-pixel in the first pixel group is located in the same column, and each sub-pixel in the second pixel group is located in the same column.

[0242] In one exemplary embodiment, as shown in FIG16, the target image data generally includes three color channels: a red channel, a blue channel, and a green channel. Each color channel corresponds to a color component map, and the original grayscale of a pixel in the three color channels is distributed in the three color component maps respectively. Thus, each color component map includes the original grayscale of each pixel in that color channel.

[0243] When recombining color channels based on the arrangement parameters of the first display screen, the color channels of mutually influencing sub-pixels are arranged into the same image, that is, arranged into a color component map. For example, as shown in Figure 15, this may include:

[0244] Step S21: Based on the original gray levels of each pixel in different color channels in the target image data, obtain multiple color component maps, each color component map including the original gray levels of each pixel in the same color channel.

[0245] Step S22: Based on the arrangement parameters, the second pixel group is transferred from the color component map where the second pixel group is located to the color component map where the first pixel group is located, so that the grayscale data corresponding to the first pixel group and the second pixel group are located in the same color component map, thus obtaining the reconstructed image.

[0246] In this embodiment, the second pixel group can be transferred from the color component map where the second pixel group is located to the color component map where the first pixel group is located, thereby obtaining a reconstructed image; or, the first pixel group can be transferred from the color component map where the first pixel group is located to the color component map where the second pixel group is located, thereby obtaining a reconstructed image.

[0247] As shown in Figures 4 and 16, the blue pixel column and the green pixel column are connected to the same data line S1. The blue sub-pixels in the blue pixel column and the green sub-pixels in the green pixel column are two sub-pixels that influence each other.

[0248] In this image, the blue pixel column and the green pixel column are the first pixel group and the second pixel group, respectively. In the target image data, the original gray levels of each sub-pixel in the blue pixel column are included in the blue component image, and the original gray levels of each sub-pixel in the green pixel column are included in the green component image. During reconstruction, the original gray levels of each sub-pixel in the blue pixel column and the original gray levels of each sub-pixel in the green pixel column need to be reconstructed into a component image so that the target model can perform feature extraction and gray level comparison.

[0249] As shown in Figure 16, in an exemplary embodiment, all mutually influencing pixel columns can be integrated into the same channel according to the arrangement parameters of the first display screen. For example, the three color component images can be arranged into one image according to the arrangement of each sub-pixel in the first display screen, thereby obtaining a reconstructed image CM. The total number of color channels included in the reconstructed image is consistent with the number of sub-pixels included in the first display screen, and the arrangement of all color channels is also the same as the arrangement of each color sub-pixel in the first display screen.

[0250] For example, as shown in Figure 16, the reconstructed image in Figure 16 has the same sub-pixel arrangement as that in Figure 4. In this way, the target model can perform feature extraction and processing on the reconstructed image.

[0251] In another example, as shown in Figures 17-19, Figures 17-19 respectively show the data diagrams of three target image data before and after preprocessing. As shown in Figure 17, when the color channels of each pixel in the target image data are recombined based on the arrangement parameters, at least two reconstructed images can be formed.

[0252] In one example of this embodiment, taking Figure 3 as an example, data line S1 connects adjacent blue and green sub-pixel columns, and data line S2 connects a column of red sub-pixels; and each red sub-pixel column in the first display screen is connected to an independent data line. Thus, as shown in Figure 17, the grayscale data of the blue and green sub-pixel columns can be arranged into the same color component image to obtain the reconstructed image CM1, and the grayscale data of the red sub-pixel columns can be combined into another color component image to obtain the reconstructed image CM2; in this way, three color component images are reconstructed to obtain two reconstructed images.

[0253] The two reconstructed images contain color channels for different colors.

[0254] In another example of this embodiment, taking Figure 4 as an example, data line S1 connects adjacent blue and green sub-pixel columns, data line S2 connects adjacent red and blue sub-pixel columns, and data line S3 connects adjacent green and red sub-pixel columns. Thus, in the first display screen, multiple pixel groups can be obtained by dividing the image into groups of two sub-pixel columns. As shown in Figure 18, the grayscale data of adjacent blue and green sub-pixel columns are arranged in the same color component image to obtain reconstructed image CM1; the grayscale data of adjacent red and blue sub-pixel columns are combined in another color component image to obtain reconstructed image CM2; and the grayscale data of adjacent green and red sub-pixel columns are combined in yet another color component image to obtain reconstructed image CM3. In this way, three color component images are reconstructed to obtain three reconstructed images. In each of these three reconstructed images, there are grayscale data of some color channels from the three color channels, and the color channels corresponding to different reconstructed images are not completely identical.

[0255] Thus, the three reconstructed images can include a first reconstructed image, a second reconstructed image, and a third reconstructed image; wherein, the first reconstructed image corresponds to the blue and red channels, the second reconstructed image corresponds to the green and red channels, and the third reconstructed image corresponds to the blue and green channels.

[0256] In an exemplary embodiment, for a first display screen with a single grid structure, as shown in FIG6, since the mutually influencing sub-pixels are located in the same column and in two adjacent rows, the reassembly can also be performed in the manner described in FIG16-FIG18.

[0257] It should be noted that during the training phase of the target model, the first and second image data can also be recombined. The recombination process can be referred to Figures 16-18.

[0258] It should be noted that the preprocessing method of the first image data can be consistent with the preprocessing method of the target image data. For example, if the first image data is preprocessed into the reconstructed image shown in Figure 18, then the target image data is also preprocessed into the reconstructed image shown in Figure 18.

[0259] In particular, when the target image data is processed into three reconstructed images, compared with forming two reconstructed images and one reconstructed image as shown in Figure 16, the amount of data in each reconstructed image can be reduced. This reduces the computational cost of the target model, improves computational efficiency, and thus improves compensation efficiency.

[0260] In an exemplary embodiment, the target model can compensate for the original grayscale of a portion of the color channels in each of the reconstructed images, and can restore the compensated reconstructed images according to the color channels to restore the arrangement architecture of the original target image data. The restored color component map includes the compensated grayscale of each pixel in the same color channel, thus, three color component maps can be restored.

[0261] The reduction process can be the reverse of the preprocessing process.

[0262] After obtaining the three restored color component images, the compensated target image data can be obtained, and the target image data can be displayed based on the restored color component images.

[0263] This reduction process can be performed by the target model.

[0264] In another exemplary embodiment, after compensating the original grayscale of some color channels in each of the reconstructed images, the target model can directly output the compensated reconstructed image. Then, the processor of the first display screen or other electronic device can be used to restore the compensated reconstructed image to obtain a restored color component map; wherein the restored color component map includes the compensated grayscale of each pixel in the same color channel. This exemplary embodiment reduces the computational requirements on the target model and simplifies its workload.

[0265] In one exemplary embodiment, the target image data after target model compensation can continue to undergo pixel-based compensation, such as compensation based on the characteristics of the pixel driving circuit. Specifically, this compensation can be based on a LUT (Look-Up Table), where the LUT stores the compensation value for each grayscale level of each sub-pixel, allowing for lookup-based compensation. The target image data after these two compensations can then be driven for display.

[0266] Based on the same inventive concept, this disclosure also provides a screen compensation device, as shown in FIG20, which is a schematic diagram of the structure of the screen compensation device. As shown in FIG20, the screen compensation device includes:

[0267] The caching module is used to acquire target image data corresponding to the first display screen. The target image data includes the original grayscale of each pixel in different color channels. The different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen.

[0268] The compensation module is used to compensate the original grayscale of each pixel in the target image data in some or all color channels using a target model; wherein the target model is obtained by deep learning using multiple datasets as training samples, the datasets include first image data and second image data, and the second image data is obtained by compensating the original grayscale of each pixel in the first image data in some or all color channels.

[0269] The compensated color channels in the pixels are determined based on the arrangement parameters of the first display screen, which include the arrangement of sub-pixels of different colors and the driving structure of multiple sub-pixels.

[0270] In an exemplary embodiment, the compensation module is configured with a target model. Specifically, the compensation module is used to compensate the original grayscale corresponding to the target sub-pixel in the target image data using the target model. The target model is obtained by deep learning using multiple datasets as training samples. The datasets include first image data and second image data. The second image data is obtained by compensating the original grayscale of each pixel in the first image data in some or all color channels. The compensated color channels are obtained through the arrangement parameters.

[0271] In one exemplary embodiment, the first display screen includes a data cable, and the compensation module includes:

[0272] The first determining unit is configured to determine, based on the arrangement parameters, a first sub-pixel and a second sub-pixel that are connected to the same data line and are adjacent to each other from a plurality of sub-pixels;

[0273] The second determining unit is used to obtain the difference between the original gray levels corresponding to the first sub-pixel and the second sub-pixel in the target image data;

[0274] The third determining unit is used to determine the target sub-pixel from the first sub-pixel and the second sub-pixel based on the difference and the driving order of the first sub-pixel and the second sub-pixel.

[0275] In an exemplary embodiment, the third determining unit is specifically used for:

[0276] If the difference is greater than a preset difference, and the original grayscale of the sub-pixel with the later driving order is higher than that of the first sub-pixel and the second sub-pixel, then the sub-pixel with the later driving order is determined as the target sub-pixel.

[0277] If the difference is less than the preset difference, the sub-pixel that is driven later in the driving order is determined as the target sub-pixel;

[0278] In one exemplary embodiment, the data line connects multiple adjacent columns of sub-pixels; wherein the first sub-pixel and the second sub-pixel are located in the same row.

[0279] In an exemplary embodiment, the compensation module is specifically used to obtain a target model corresponding to the arrangement parameters based on the arrangement parameters, wherein different arrangement parameters correspond to different target models; and to compensate the original grayscale corresponding to the target sub-pixel in the target image data using the target model corresponding to the arrangement parameters.

[0280] In an exemplary embodiment, the first display screen includes data lines extending along a column direction, and one data line connects a first pixel group and a second pixel group that are adjacent in position and different in color. The device may further include a preprocessing module for preprocessing the target image data; wherein, the preprocessing module is specifically used for:

[0281] Based on the arrangement parameters, the color channels of each pixel in the target image data are recombined to obtain at least one reconstructed image. The reconstructed image includes the original grayscale corresponding to at least two color channels, and the two adjacent color channels correspond to the first pixel group and the second pixel group, respectively.

[0282] The compensation module is specifically used to compensate for the original grayscale of a portion of the color channels in each of the reconstructed images using the target model; wherein the color channels being compensated are the target sub-pixels;

[0283] The first image data is the image data after preprocessing.

[0284] In one exemplary embodiment, the preprocessing module includes:

[0285] The acquisition unit is used to acquire multiple color component maps based on the original gray levels of each pixel in different color channels in the target image data, and each color component map includes the original gray levels of each pixel in the same color channel.

[0286] The recombination unit is used to transfer the second pixel group from the color component map where the second pixel group is located to the color component map where the first pixel group is located, based on the arrangement parameters, so that the original gray levels corresponding to the first sub-pixel and the second sub-pixel are arranged in the same color component map to obtain the recombined image.

[0287] In one exemplary embodiment, the system includes multiple reconstructed images, each of which corresponds to two different color channels, and the color channels corresponding to different reconstructed images are not exactly the same.

[0288] In one exemplary embodiment, the image includes a first reconstructed image, a second reconstructed image, and a third reconstructed image;

[0289] The first reconstructed image corresponds to the blue and red channels, the second reconstructed image corresponds to the green and red channels, and the third reconstructed image corresponds to the blue and green channels.

[0290] In one exemplary embodiment, the apparatus further includes:

[0291] The restoration module is used to restore the compensated reconstructed image based on the arrangement parameters to obtain a restored color component image; wherein the restored color component image includes the compensated grayscale of each pixel in the same color channel.

[0292] In one exemplary embodiment, the process of obtaining the target model includes:

[0293] The first image data is input into a preset network, which is used to compensate the original grayscale of the pixels in the first image data in some or all color channels to output predicted image data.

[0294] The predicted image data output by the preset network is obtained, and the parameters of the preset network are updated based on the difference between the predicted image data and the second image data.

[0295] In one exemplary embodiment, the dataset includes a first dataset and a second dataset, the first dataset corresponds to the first display screen, the second dataset corresponds to a plurality of second display screens, the first display screen is one of the second display screens, or the first display screen does not belong to the second display screens;

[0296] The process of obtaining the target model includes:

[0297] Based on the second dataset, deep learning is performed on the preset network to obtain an initial model;

[0298] Based on the first dataset, deep learning is performed on the initial model to obtain the target model adapted to the first display screen.

[0299] In one exemplary embodiment, the apparatus further includes a debugging module, the debugging module being configured to acquire second image data, including:

[0300] The first processing unit is configured to determine the third sub-pixel to be compensated based on the first image data and the arrangement parameters of the second display screen; wherein the second display screen is different from the first display screen, or the first display screen and the second display screen are the same display screen;

[0301] The second processing unit is used to perform at least one iterative compensation on the third sub-pixel until the termination condition is met, and to use the compensated image obtained when the termination condition is met as the second image data.

[0302] In each iteration of the compensation process, the grayscale of the third sub-pixel is increased or decreased according to a preset compensation value, and the first display brightness difference between the actual display brightness of the compensated third sub-pixel and the preset display brightness is determined. The preset display brightness is the ideal display brightness corresponding to the original grayscale of the third sub-pixel.

[0303] The termination condition includes: the first display brightness difference is less than the first preset brightness difference.

[0304] In one exemplary embodiment, the apparatus further includes:

[0305] The brightness difference determination module is used to determine the second display brightness difference between the actual display brightness of sub-pixels of the same color and gray level in the compensated first image data;

[0306] The termination condition also includes: the second display brightness difference is less than the second preset brightness difference;

[0307] The sub-pixels having the same color and grayscale include: the third sub-pixel and the uncompensated sub-pixels.

[0308] In one exemplary embodiment, the debugging module further includes:

[0309] An image segmentation unit is used to divide the first image data into multiple image regions, each of which corresponds to a sub-region in the display area of ​​the second display screen;

[0310] The second processing unit is specifically used to perform at least one iterative compensation on the third sub-pixel in each of the sub-regions until the termination condition is met;

[0311] The compensated image data of each of the sub-regions is stitched together when the termination condition is met to obtain the second image data.

[0312] In an exemplary embodiment, the second processing unit is specifically configured to acquire resistivity change curves of signal lines on the second display screen at different regions of the display area; and based on the resistivity change curves, divide the first image data into a plurality of image regions; wherein the signal lines include at least data lines.

[0313] Based on the same inventive concept, this disclosure also provides a display device. Referring to FIG21, a structural schematic diagram of the display device is shown. As shown in FIG21, the display device can be an LCD screen, including a display panel, a data processing module, and a display driving module. The data processing module is connected to the display driving module, and the display driving module is connected to the display panel.

[0314] The data processing module is used to execute the image quality compensation method for the display screen;

[0315] The display driving module is used to drive the display panel to display images based on the compensated target image data output by the data processing module.

[0316] In this embodiment, the display panel may include a color filter substrate, an array substrate, and a liquid crystal layer located between the color filter substrate and the array substrate. In one example, it may also include a backlight panel located on the side of the array substrate away from the color filter substrate.

[0317] In one example of this embodiment, as shown in FIG21, the display panel includes multiple gate lines and multiple data lines, and the sub-pixel is located in the area defined by the intersection of the gate lines and the data lines;

[0318] The display driving module includes a gate driving unit connected to the gate line and a data driving unit connected to the data line.

[0319] The data processing module includes a processing unit and a timing controller connected to the processing unit. The timing controller is connected to the gate driving unit and the data driving unit, respectively.

[0320] The processing unit is configured with a target model, which is used to execute the image quality compensation method for the display screen.

[0321] As shown in Figure 21, the data processing module may include a SOC (System on Chip) in the display device. For example, the processing unit may be an SOC, and the target model may be located within the SOC.

[0322] In this way, the target image data, after compensation by the target model, can continue to enter the timing controller. The timing controller then compensates for the grayscale of each sub-pixel in the compensated target image data according to the LUT table. This compensation can be based on pixel characteristics, such as compensation based on the characteristics of the pixel driving circuit. After two compensations, the target image data can be driven for display. The analog electrical signal output by the source driver is based on the grayscale of each sub-pixel in the target image data after the two compensations.

[0323] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0324] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0325] The above provides a detailed description of a display screen image quality compensation method, image quality compensation device, and display device provided by this disclosure. Specific examples have been used to illustrate the principles and implementation methods of this disclosure. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this disclosure. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this disclosure. Therefore, the content of this specification should not be construed as a limitation of this disclosure.

[0326] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0327] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0328] The terms "an embodiment," "embodiment," or "one or more embodiments" as used herein mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of this disclosure. Furthermore, please note that the examples of the phrase "in one embodiment" do not necessarily all refer to the same embodiment.

[0329] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this disclosure may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0330] In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This disclosure can be implemented by means of hardware comprising a plurality of different elements and by means of a suitably programmed computer. In a unit claim enumerating a plurality of means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0331] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit them. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A method for image quality compensation of a display screen, wherein, The method includes: Acquire target image data to be displayed on the first display screen. The target image data includes the original grayscale of each pixel in different color channels. The different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen. Based on the arrangement parameters corresponding to the first display screen and the target image data, the target sub-pixel to be compensated is obtained from the multiple sub-pixels, and the original grayscale corresponding to the target sub-pixel in the target image data is compensated. The arrangement parameters include the driving structure of the sub-pixels, which characterizes the connection architecture between the data lines and each sub-pixel in the first display screen.

2. The image quality compensation method for a display screen according to claim 1, wherein, The step of obtaining the target sub-pixel to be compensated from a plurality of sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data, and compensating the original grayscale corresponding to the target sub-pixel in the target image data, includes: The original grayscale corresponding to the target sub-pixel in the target image data is compensated using the target model; The target model is obtained by deep learning using multiple datasets as training samples. The datasets include first image data and second image data. The second image data is obtained by compensating the original grayscale of each pixel in the first image data in some or all color channels. The compensated color channels are obtained through the arrangement parameters.

3. The image quality compensation method for a display screen according to claim 1 or 2, wherein, The step of obtaining the target sub-pixel to be compensated from a plurality of sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data includes: Based on the arrangement parameters, a first sub-pixel and a second sub-pixel that are connected to the same data line and are adjacent to each other are determined from a plurality of sub-pixels; the first sub-pixel and the second sub-pixel correspond to different colors; Obtain the difference between the original gray levels corresponding to the first sub-pixel and the second sub-pixel in the target image data; The target sub-pixel is determined from the first sub-pixel and the second sub-pixel based on the difference and the driving order of the first sub-pixel and the second sub-pixel.

4. [Amended according to Rule 26, 11.04.2025] The image quality compensation method for the display screen according to claim 3, wherein, The step of determining the target sub-pixel among the first and second sub-pixels based on the difference and the driving order of the first and second sub-pixels includes: If the difference is greater than a preset difference, and the original grayscale of the sub-pixel with the later driving order is higher than that of the first sub-pixel and the second sub-pixel, then the sub-pixel with the later driving order is determined as the target sub-pixel. If the difference is less than the preset difference, the sub-pixel that is driven later in the driving order is determined as the target sub-pixel.

5. The image quality compensation method for a display screen according to claim 3, wherein, The data line connects multiple adjacent columns of sub-pixels, and the connected columns of sub-pixels correspond to different colors; wherein the first sub-pixel and the second sub-pixel are located in the same row.

6. The image quality compensation method for a display screen according to claim 2, wherein, The step of compensating for the original grayscale corresponding to the target sub-pixel in the target image data using the target model includes: Based on the arrangement parameters, a target model corresponding to the arrangement parameters is obtained, and different arrangement parameters correspond to different target models; The original grayscale corresponding to the target sub-pixel in the target image data is compensated using the target model corresponding to the arrangement parameters.

7. The image quality compensation method for a display screen according to claim 2, wherein, A data line connects adjacent and different color first and second pixel groups. Before compensating the original grayscale of the target sub-pixels in the target image data using the target model, the method further includes: preprocessing the target image data; wherein the preprocessing includes: Based on the arrangement parameters, the grayscale data of each pixel in the target image data in multiple color channels are recombined to obtain at least one reconstructed image. The reconstructed image includes the original grayscale corresponding to at least two color channels, and the two adjacent color channels correspond to the first pixel group and the second pixel group, respectively. The step of compensating for the original grayscale of the target sub-pixels in the target image data using a target model includes: The original grayscale of a portion of the color channels in each of the reconstructed images is compensated using the target model, and the compensated color channels are the target sub-pixels. The first image data is the image data after preprocessing.

8. The image quality compensation method for a display screen according to claim 7, wherein, Based on the arrangement parameters, the grayscale data of each pixel in the target image data across multiple color channels are recombined to obtain at least one reconstructed image, including: Based on the original gray levels of each pixel in the target image data in different color channels, multiple color component maps are obtained, and each color component map includes the original gray levels of each pixel in the same color channel. Based on the arrangement parameters, the second pixel group is transferred from the color component map where the second pixel group is located to the color component map where the first pixel group is located, so that the grayscale data corresponding to the first pixel group and the second pixel group are located in the same color component map, thereby obtaining the reconstructed image.

9. The image quality compensation method for a display screen according to claim 7, wherein, The image includes multiple reconstructed images, each of which corresponds to two different color channels. The color channels corresponding to different reconstructed images are not exactly the same.

10. The image quality compensation method for a display screen according to claim 7, wherein, Including the first reconstructed image, the second reconstructed image, and the third reconstructed image; The first reconstructed image corresponds to the blue and red channels, the second reconstructed image corresponds to the green and red channels, and the third reconstructed image corresponds to the blue and green channels.

11. The image quality compensation method for a display screen according to claim 7, wherein, After compensating the original grayscale of some color channels in each reconstructed image using the target model, the method further includes: Based on the arrangement parameters, the compensated reconstructed image is restored to obtain the restored color component map; The target image data is displayed based on the restored color component map. The restored color component map includes the compensated grayscale of each pixel in the same color channel.

12. The image quality compensation method for a display screen according to claim 2, wherein, The process of obtaining the target model includes: The first image data is input into a preset network, which is used to compensate the original grayscale of the pixels in the first image data in some or all color channels to output predicted image data. The predicted image data output by the preset network is obtained, and the parameters of the preset network are updated based on the difference between the predicted image data and the second image data.

13. The image quality compensation method for a display screen according to claim 2, wherein, The dataset includes a first dataset and a second dataset. The first dataset corresponds to the first display screen, and the second dataset corresponds to multiple second display screens. The first display screen is one of the second display screens, or the first display screen does not belong to any of the second display screens. The process of obtaining the target model includes: Based on the second dataset, deep learning is performed on the preset network to obtain an initial model; Based on the first dataset, deep learning is performed on the initial model to obtain the target model adapted to the first display screen.

14. The image quality compensation method for a display screen according to claim 2, wherein, The process of acquiring the second image data includes: Based on the first image data and the arrangement parameters of the second display screen, the third sub-pixel to be compensated is determined; wherein, the second display screen is different from the first display screen, or the first display screen and the second display screen are the same display screen; The third sub-pixel is compensated at least once until the termination condition is met, and the compensated image obtained when the termination condition is met is used as the second image data. In each iteration of the compensation process, the grayscale of the third sub-pixel is increased or decreased according to a preset compensation value, and the first display brightness difference between the actual display brightness of the compensated third sub-pixel and the preset display brightness is determined. The preset display brightness is the ideal display brightness corresponding to the original grayscale of the third sub-pixel. The termination condition includes: the first display brightness difference is less than the first preset brightness difference.

15. The image quality compensation method for a display screen according to claim 14, wherein, The method further includes: The second display brightness difference between the actual display brightness of each sub-pixel corresponding to the same color and gray level in the compensated first image data is determined. The termination condition also includes: the second display brightness difference is less than the second preset brightness difference.

16. The image quality compensation method for a display screen according to claim 15, wherein, The sub-pixels corresponding to the same color and gray level include: the third sub-pixel and the uncompensated sub-pixels.

17. The image quality compensation method for a display screen according to claim 14, wherein, The method further includes: The first image data is divided into multiple image regions, and each image region corresponds to a sub-region in the display area of ​​the second display screen; The step of performing at least one iteration of compensation on the third sub-pixel until the termination condition is met includes: Perform at least one iteration compensation on the third sub-pixel within each sub-region until the termination condition is met; The compensated image data of each of the sub-regions is stitched together when the termination condition is met to obtain the second image data.

18. The image quality compensation method for a display screen according to claim 17, wherein, The step of dividing the first image data into multiple image regions includes: Obtain the resistivity variation curves of the signal lines on the second display screen at different regions of the display area; Based on the resistivity change curve, the first image data is divided into multiple image regions; wherein, the signal lines include at least data lines.

19. A picture compensation device, wherein, The image compensation device includes: The caching module is used to acquire target image data corresponding to the first display screen. The target image data includes the original grayscale of each pixel in different color channels. The different color channels of a pixel correspond to different sub-pixels of a pixel unit in the first display screen. The compensation module is used to obtain the target sub-pixel to be compensated from multiple sub-pixels based on the arrangement parameters corresponding to the first display screen and the target image data, and to compensate the original grayscale corresponding to the target sub-pixel in the target image data. The arrangement parameters include the driving structure of the sub-pixels, which characterizes the connection architecture between the data lines and each sub-pixel in the first display screen.

20. A display device, wherein, include: The system comprises a display panel, a data processing module, and a display driver module, wherein the data processing module is connected to the display driver module, and the display driver module is connected to the display panel; wherein... The data processing module is used to execute the image quality compensation method for the display screen according to any one of claims 1-18; The display driving module is used to drive the display panel to display images based on the compensated target image data output by the data processing module.

21. The display device according to claim 20, wherein, The display panel includes multiple gate lines and multiple data lines, and the sub-pixel is located in the area defined by the intersection of the gate lines and the data lines; The display driving module includes a gate driving unit connected to the gate line and a data driving unit connected to the data line. The data processing module includes a processing unit and a timing controller connected to the processing unit. The timing controller is connected to the gate driving unit and the data driving unit, respectively. The processing unit is configured with a target model, which is used to execute the image quality compensation method for the display screen according to any one of claims 1-18.

Images