An image processing method suitable for variable-size TFT display screen

Abstract

The present invention relates to an image processing method suitable for a variable-size TFT display screen, comprising the following steps: obtaining the display area size and range of the TFT display screen in real time, obtaining the position coordinate range of the edge area of ​​the photo-alignment film; using an image processing method to detect the pixel brightness within the corresponding position coordinate range of the edge area of ​​the photo-alignment film; calculating and obtaining compensation curve correction parameters in combination with factors affecting the brightness distribution of the edge area; using the compensated brightness distribution curve to correct the pixel brightness of the edge area of ​​the photo-alignment film; after completing the pixel brightness compensation of the edge area, obtaining the change value of the display contrast before and after compensation; by analyzing the relationship between the change value of the display contrast and the pixel brightness compensation intensity, adjusting the compensation curve parameters of the edge area in real time. The present invention can effectively improve the brightness uniformity of the edge area of ​​the TFT display screen, improve the overall display quality, and can adapt to the needs of display screens of different sizes, and has strong practicality and adaptability.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an image processing method suitable for a variable-size TFT display screen. Background Art

[0002] When the display area of ​​a variable-size TFT display changes dynamically, the orientation stability of the photo-alignment film in the edge area faces severe challenges. Due to the change in the display area size, the distribution of environmental stress in the edge area changes, resulting in a weakening of the orientation force on the photo-alignment film in this area. The weakening of the orientation force causes the arrangement of liquid crystal molecules to become disordered, thus affecting the uniform brightness of edge pixels. In order to compensate for the change in the brightness of edge pixels, the pixel brightness compensation curve needs to be corrected. However, the correction of the compensation curve is not easy. First, it is necessary to accurately characterize the quantitative relationship between the weakening of environmental stress and the brightness of edge pixels, which requires consideration of factors such as the material properties of the photo-alignment film, the properties of liquid crystal, and the pixel electrode structure. Secondly, the correction amplitude of the compensation curve needs to be dynamically adjusted according to the change in the display size, which places high demands on the real-time and robustness of the compensation algorithm. In addition to affecting the uniform brightness of edge pixels, the correction of the compensation curve will also affect the display contrast. The change in the compensation curve will change the voltage-brightness response characteristics of the pixel, thereby causing a change in the display contrast. Especially in the low grayscale area, a slight change in the compensation curve may lead to a significant decrease in contrast.

[0003] Therefore, when correcting the compensation curve, it is necessary to consider both the uniformity of the edge pixel brightness and the overall display contrast, which further increases the difficulty of designing the compensation algorithm. In short, the problem of orientation stability of the photo-alignment film in the edge area caused by the dynamic change of the display area in the variable-size TFT display, as well as the resulting pixel brightness compensation and display contrast problems, is a complex technical problem. Summary of the invention

[0004] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide an image processing method suitable for a variable-size TFT display screen. The image processing method suitable for a variable-size TFT display screen can effectively improve the brightness uniformity of the edge area of ​​the TFT display screen, improve the overall display quality, and can meet the needs of display screens of different sizes, and has strong practicality and adaptability.

[0005] The image processing method applicable to a variable-size TFT display screen of the present invention comprises the following steps:

[0006] S1, obtaining the display area size and range of the TFT display screen in real time, and obtaining the position coordinate range of the edge area of ​​the photo-alignment film in combination with a preset fixed percentage of the edge area in the display area;

[0007] S2, using an image processing method to detect the pixel brightness within the coordinate range of the position corresponding to the edge area of ​​the photo-alignment film, and obtain the pixel brightness distribution characteristics of the edge area of ​​the photo-alignment film;

[0008] S3, calculating compensation curve correction parameters according to the pixel brightness distribution characteristics and factors affecting the brightness distribution of the edge area;

[0009] S4, using the compensated brightness distribution curve to correct the pixel brightness in the edge area of ​​the photo-alignment film;

[0010] S5, after completing the pixel brightness compensation in the edge area of ​​the photo-alignment film, evaluating the overall contrast of the display image to obtain a change value of the display contrast before and after the compensation;

[0011] S6. By analyzing the relationship between the change value of the display contrast and the pixel brightness compensation intensity, and according to the dynamic change of the display size, the compensation curve parameters of the edge area are adjusted in real time.

[0012] Preferably, the step S1 specifically includes:

[0013] Obtain the display area brightness matrix according to the output signal of the display controller interface, obtain the brightness value through the sampling points set horizontally and vertically in the display area, and use the least squares method to fit to obtain the display area boundary coordinate point set;

[0014] Performing pixel mapping transformation on the display area boundary coordinate point set, performing product operation on the preset edge area ratio value and the display area boundary coordinate point set, and obtaining the position coordinate range of the edge area of ​​the photo-alignment film;

[0015] The optical sensor arranged in the edge area is used to collect the polarized light intensity of the photo-alignment film, and the photo-alignment angle value is obtained by fitting the light intensity curve;

[0016] According to the light alignment angle value and a preset difference threshold, a proportional integral controller is used to calculate the position compensation amount, and compensation is performed by drivers arranged in the horizontal and vertical directions until the light alignment angle value difference is less than the difference threshold.

[0017] Preferably, the step S2 specifically includes:

[0018] Acquire a first image of the edge area of ​​the photo-alignment film from an area array camera, and obtain a filtered second image by processing the first image with a Gaussian filter;

[0019] Calculating a brightness threshold for the second image using the maximum inter-class variance method, dividing the brightness value into a plurality of quantization intervals using the brightness threshold, and recording a brightness gradient value of the quantization interval for which a standard deviation ratio of adjacent intervals exceeds a preset threshold;

[0020] According to the brightness gradient value, a linear fitting calculation is used to obtain a change in the alignment force of the photo-alignment film within the interval, and the change exceeding a change threshold is marked as an alignment force weakening region;

[0021] A compensation voltage value is calculated for the alignment force weakened area through a proportional integral controller, and the compensation voltage value is determined according to the difference between the brightness value of the alignment force weakened area and the reference brightness value. The brightness value is continuously collected until the brightness standard deviation in the area is less than a preset standard deviation threshold.

[0022] Preferably, the step S2 further includes:

[0023] Dividing the edge area of ​​the photo-alignment film into a number of small areas, respectively calculating the average brightness values ​​and standard deviations of the pixels in the small areas, and comparing the average brightness values ​​and standard deviations of the small areas;

[0024] The small areas whose differences exceed the preset difference threshold are judged as having uneven brightness distribution. The unevenness degrees of different small areas are obtained through the spatial differences of brightness distribution in each of the small areas. The proportion of pixel points that need to be compensated and corrected and the brightness adjustment intensity are determined according to the unevenness degrees.

[0025] Preferably, the step S3 specifically includes:

[0026] Factors that affect the brightness distribution in the edge area include the photo-alignment film material properties, liquid crystal properties, and pixel electrode structure;

[0027] Collecting refractive index values, rotation angle values, and electrode width values ​​of detection points arranged in an array in an edge region of the photo-alignment film to obtain a characteristic parameter matrix;

[0028] Comparing the refractive index value, the rotation angle value and the electrode width value with preset thresholds respectively to obtain a comparison result;

[0029] The detection points are grouped to obtain a compensation area, and a corresponding relationship between a refractive index value, a rotation angle value and a pixel voltage value in the compensation area is established by multivariate linear regression to obtain a regression coefficient matrix;

[0030] The driving voltage value of the compensation area is calculated according to the regression coefficient matrix, and the driving voltage value is adjusted according to the standard deviation of the brightness distribution curve by using a proportional-integral controller until the standard deviation is less than a standard deviation threshold.

[0031] Preferably, the step S4 specifically includes:

[0032] Acquire a target brightness value, generate a compensation voltage increment matrix according to a difference between the target brightness value and an actual brightness value, and set a driving voltage in an edge area driving circuit by using the compensation voltage increment matrix;

[0033] Acquiring brightness data collected by an optical detector array from an edge region of the photo-alignment film and generating a brightness data matrix;

[0034] The standard deviation of the brightness data matrix is ​​calculated. If the calculated standard deviation is greater than a preset standard deviation threshold, a voltage increment is added based on the compensation voltage increment matrix until the standard deviation is less than the preset standard deviation threshold or a maximum number of iterations is reached.

[0035] Preferably, the step S4 further includes:

[0036] According to the relative position of the pixel point in the edge area of ​​the photo-alignment film, the horizontal and vertical coordinate values ​​of the corresponding compensation curve are substituted into the compensation curve to obtain the target pixel brightness value after the pixel point is adjusted;

[0037] Compare the difference between the current pixel brightness value and the target pixel brightness value, and adjust the RGB parameters of the pixel point in a gradual manner until the target value is reached;

[0038] Brightness compensation is performed on all selected pixel points in the edge region of the photo-alignment film, and the display is refreshed after a certain number of pixel points are adjusted.

[0039] Preferably, the step S5 specifically includes:

[0040] Collecting a display image through an optical detector, dividing the display image into grid blocks and counting the brightness distribution of pixels in the blocks to obtain a contrast matrix of the edge area of ​​the photo-alignment film;

[0041] Calculating the contrast difference before and after compensation in a block according to the contrast matrix, normalizing the contrast difference, and obtaining a contrast change map of the edge area of ​​the photo-alignment film by linear interpolation;

[0042] A linear regression mapping relationship is established according to the contrast change graph, and contrast change data before and after compensation are input to obtain a contrast score of the overall display area;

[0043] A scoring data buffer area is set in the display controller, and the contrast score is recorded in the scoring data buffer area. If the contrast score is lower than a preset contrast score threshold, the compensation process is triggered to restart.

[0044] Preferably, the step S6 specifically includes:

[0045] Acquire a real-time image of the edge area of ​​the photo-alignment film by an optical imager, and generate a compensation point matrix of the edge area according to the real-time image;

[0046] Extracting contrast values ​​between adjacent points from the compensation point matrix, and using multiple linear regression to obtain the corresponding relationship between edge compensation parameters and contrast changes;

[0047] A proportional-integral controller is set according to the edge compensation parameter, and the proportional-integral controller outputs a compensation voltage value to the edge area;

[0048] If the display size change exceeds a preset display size change threshold, a horizontal component and a vertical component correction matrix are constructed according to the edge compensation parameter, and the compensation parameter value is updated by linear interpolation.

[0049] The image processing method applicable to a variable-size TFT display screen described in the present invention has the advantages that:

[0050] The image processing method for a variable-size TFT display screen of the present invention can automatically adapt to display screens of different sizes by acquiring the display area size and range of the TFT display screen in real time, without the need for separate settings or adjustments for each size, thereby improving the flexibility and versatility of processing; by detecting the pixel brightness distribution characteristics of the edge area of ​​the light alignment film, and calculating the compensation curve correction parameters accordingly, the brightness of the edge area can be accurately corrected, thereby improving the overall uniformity and consistency of the display image and enhancing the visual experience; by correcting the pixel brightness of the edge area of ​​the light alignment film, the display quality degradation caused by edge effects or uneven brightness can be reduced, thereby improving the overall display quality; after completing the pixel brightness compensation of the edge area, the overall contrast of the display image can also be evaluated, and the compensation curve parameters can be adjusted in real time according to the dynamic changes in the size of the display screen. This dynamic adjustment capability ensures that the TFT display screen can achieve the best display uniformity and contrast performance at different sizes. The image processing method for a variable-size TFT display screen can effectively improve the brightness uniformity of the edge area of ​​the TFT display screen, improve the overall display quality, and can adapt to the needs of display screens of different sizes, and has strong practicality and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention is a flowchart of an image processing method suitable for a variable-size TFT display screen. DETAILED DESCRIPTION

[0052] like Figure 1 As shown, the image processing method applicable to a variable-size TFT display screen of the present invention comprises the following steps:

[0053] S1. Real-time acquisition of the display area size and range of the TFT display screen, combined with a preset fixed percentage of the edge area in the display area, to obtain a position coordinate range of the edge area of ​​the photo-alignment film; the position coordinate range of the edge area of ​​the photo-alignment film is obtained by combining a preset fixed percentage of the edge area in the display area to calculate a pixel coordinate range of the edge area;

[0054] S2, using an image processing method to detect the pixel brightness within the coordinate range of the position corresponding to the edge area of ​​the photo-alignment film, and obtain the pixel brightness distribution characteristics of the edge area of ​​the photo-alignment film;

[0055] S3, calculating compensation curve correction parameters according to pixel brightness distribution characteristics and factors affecting edge area brightness distribution;

[0056] S4, using the compensated brightness distribution curve to correct the pixel brightness in the edge area of ​​the photo-alignment film;

[0057] S5, after completing the pixel brightness compensation in the edge area of ​​the photo-alignment film, evaluating the overall contrast of the display image to obtain a change value of the display contrast before and after the compensation;

[0058] S6. By analyzing the relationship between the change in display contrast and the pixel brightness compensation intensity, and adjusting the compensation curve parameters of the edge area in real time according to the dynamic change of the display size; the relationship between the change in display contrast and the pixel brightness compensation intensity reveals the influence mechanism of the pixel brightness compensation curve correction on the display contrast; real-time adjustment of the compensation curve parameters of the edge area can ensure that the TFT display screen can achieve the best display uniformity and contrast performance at different sizes.

[0059] Furthermore, in this embodiment, step S1 specifically includes:

[0060] Obtain the display area brightness matrix according to the output signal of the display controller interface, obtain the brightness value through the sampling points set horizontally and vertically in the display area, and use the least squares method to fit to obtain the display area boundary coordinate point set;

[0061] Performing pixel mapping transformation on the display area boundary coordinate point set, performing product operation on the preset edge area ratio value and the display area boundary coordinate point set, and obtaining the position coordinate range of the edge area of ​​the photo-alignment film;

[0062] The optical sensor arranged in the edge area is used to collect the polarized light intensity of the photo-alignment film, and the photo-alignment angle value is obtained by fitting the light intensity curve;

[0063] According to the threshold of the difference between the light orientation angle value and the preset value, a proportional integral controller is used to calculate the position compensation amount, and the compensation is performed by the drivers arranged in the horizontal and vertical directions until the difference between the light orientation angle value is less than the threshold of the difference;

[0064] Specifically, the display area brightness matrix is ​​read from the display controller interface, 16 sampling points are evenly distributed in the horizontal and vertical directions of the display area, a photoelectric detector is used to obtain the brightness value of each sampling point, and a display area boundary coordinate point set is obtained by least squares fitting;

[0065] According to the preset edge area ratio value parameter, pixel mapping transformation is performed on the display area boundary coordinate point set to obtain the edge area coordinate range;

[0066] If the display area is a rectangle, the horizontal and vertical coordinate values ​​of the edge area are obtained by multiplying the boundary point set by the preset proportion value;

[0067] Five optical sensors are arranged in the edge area to collect the angle data of the photo-alignment film. The optical sensors are perpendicular to the boundary of the display area, collect the polarized light intensity of the photo-alignment film, and calculate the photo-alignment angle value by fitting the light intensity curve.

[0068] According to the collected photo-alignment angle value, a proportional-integral controller is set to perform position compensation calculation, and the controller outputs the edge area coordinate offset. Four drivers are set in the horizontal and vertical directions of the photo-alignment film position to perform compensation until the difference in the photo-alignment angle is less than a preset threshold;

[0069] Here is an example:

[0070] The display controller obtains the brightness matrix data through the preset display driver interface. The 16 sampling points set horizontally and vertically in the display area form a 16x16 sampling matrix. The brightness value of each sampling point ranges from 0 to 255. The pixel coordinate value of the sampling point is provided by the display controller. The brightness value collected by the photodetector needs to be calibrated with the brightness matrix data of the display controller. The calibration coefficient is set to 1.2 to obtain the actual brightness value and store it in the brightness data matrix.

[0071] The preset ratio value of the display area boundary is usually set between 0.15 and 0.2. Taking 0.18 as an example, when the display area has a resolution of 1920x1080, the pixel width of the edge area is 346 pixels and the height is 194 pixels. The coordinate range of the edge area is obtained by multiplying the boundary point set by the preset ratio value. The coordinate point set of the edge area forms a closed rectangular area.

[0072] When collecting the polarized light intensity of the photo-alignment film through the optical sensor, the optical sensor is kept 90 degrees perpendicular to the boundary of the display area, the sampling frequency of the optical sensor is 100 Hz, the polarized light intensity value ranges from 0 to 100, the photo-alignment angle value is obtained by fitting the light intensity curve, and the angle value ranges from 0 to 90 degrees. When the sensor collects data, a sampling delay of 10 milliseconds is set;

[0073] The proportional coefficient of the proportional-integral controller is set to 0.8, and the integral time constant is set to 0.5 seconds. The controller calculates the position compensation amount according to the expected value and the actual value of the light alignment angle. The maximum value of the compensation amount does not exceed 50 microns. The response time of the driver is less than 5 milliseconds. The position compensation is performed by 4 drivers in the horizontal and vertical directions respectively. The movement accuracy of the driver is 1 micron. The preset threshold of the light alignment angle is plus or minus 2 degrees.

[0074] When compensating the position of the photo-alignment film, the driver is driven by a stepper motor with a stepping accuracy of 0.1 micron per step. The travel range of the driver is plus or minus 100 microns. The response delay of the driver is less than 1 millisecond. The position compensation adopts a closed-loop control method. The feedback signal comes from the angle data collected by the optical sensor. The control cycle is 10 milliseconds. The photo-alignment angle value is monitored in real time during the compensation process.

[0075] The least squares method is used to fit the coordinate points of the display area boundary. The fitting curve uses a quadratic polynomial. The fitting accuracy is set to 0.1 pixels. The fitting result obtains a set of closed boundary curve equations. The boundary curve is multiplied by the preset ratio value to obtain the coordinate range of the edge area. The edge area forms a closed area, and the set of pixel points in the area constitutes the working area of ​​the photo-alignment film.

[0076] Furthermore, in this embodiment, step S2 specifically includes:

[0077] Collecting a first image of the edge area of ​​the photo-alignment film from an area array camera, and processing the first image with a Gaussian filter to obtain a filtered second image;

[0078] The brightness threshold of the second image is calculated by the maximum inter-class variance method, and the brightness value is divided into multiple quantization intervals by the brightness threshold. For the quantization interval whose standard deviation ratio of adjacent intervals exceeds the preset threshold, the brightness gradient value of the quantization interval is recorded;

[0079] According to the brightness gradient value, a linear fitting calculation is used to obtain a change in the alignment force of the photo-alignment film within the interval, and a change exceeding a change threshold is marked as an alignment force weakening region;

[0080] The compensation voltage value is calculated by the proportional integral controller for the area where the orientation force is weakened. The compensation voltage value is determined according to the difference between the brightness value of the area where the orientation force is weakened and the reference brightness value. The brightness value is continuously collected until the brightness standard deviation in the area is less than the preset standard deviation threshold value.

[0081] Here is an example:

[0082] An area array camera with a resolution of 4096x3072 is used to collect images of the edge area of ​​the photo-alignment film. The collection area is set within a 1000x1000 pixel range outside the edge of the display screen. A Gaussian filter with a kernel size of 5x5 pixels and a standard deviation of 1.5 is used to process the brightness value matrix in the collected image to obtain a brightness distribution map of the edge area of ​​the photo-alignment film after filtering.

[0083] For the filtered brightness distribution map, the maximum inter-class variance method is used to calculate the brightness threshold. The brightness value range is divided into 8 quantization intervals between 0 and 255. The standard deviation of the brightness difference is calculated for the pixels in each interval. The uniformity of the brightness distribution is judged by the standard deviation value. If the standard deviation ratio of adjacent intervals is greater than 2.0, the brightness gradient value of the interval is recorded.

[0084] Based on the recorded brightness gradient values, linear fitting is used to calculate the change trend of the alignment force of the photo-alignment film in each interval. The area where the alignment force change in the interval exceeds the preset threshold of 0.5 is marked as the alignment force weakened area, and the pixel coordinates of the marked area are located;

[0085] According to the difference between the brightness value of the area where the orientation force is weakened and the preset reference brightness value, a proportional integral controller with a proportional coefficient of 0.8 is set to calculate the compensation voltage value. The output voltage range of the driver is between 0 and 5 volts, and the voltage adjustment step value is 0.1 volt. During the compensation process, the brightness value is continuously collected until the brightness standard deviation in the area is less than 5;

[0086] In the image acquisition of the edge area of ​​the photo-alignment film, the area array camera's 4096x3072 resolution provides fine pixel information. The camera is set to sample within the 1000x1000 pixel range at the edge of the display screen. The sampling area covers the key orientation area of ​​the photo-alignment film. The Gaussian filter uses a 5x5 pixel kernel to eliminate image noise. The standard deviation setting of 1.5 can not only retain edge details, but also effectively suppress random noise.

[0087] The maximum inter-class variance method is used to calculate the brightness threshold. The brightness values ​​from 0 to 255 are divided into 8 intervals, and the width of each interval is 32. In actual applications, interval 1 corresponds to the dark area of ​​0 to 31, interval 4 corresponds to the mid-tone area of ​​96 to 127, and interval 8 corresponds to the highlight area of ​​224 to 255. The uniformity is determined by calculating the standard deviation of the brightness value of the pixel points in each interval. If the standard deviation ratio exceeds 2.0, it indicates that there is an obvious brightness jump at that location.

[0088] The variation trend of the orientation force of the photo-alignment film is reflected by linear fitting. Taking a local block of 100x100 pixels as a unit, the slope of the brightness gradient value in the block with the position is calculated. When the slope exceeds 0.5, it indicates that the orientation force is significantly weakened. In a typical 1000x1000 pixel sampling area, there may be 5 to 10 blocks with weakened orientation force, and these blocks often show cluster distribution characteristics.

[0089] In the compensation process of the proportional-integral controller, the proportional coefficient of 0.8 provides a relatively smooth response characteristic. The output voltage range of the driver from 0 to 5 volts covers the adjustment range required for orientation force compensation. The adjustment step value of 0.1 volt ensures the fineness of compensation. When the standard deviation of regional brightness drops below 5, it indicates that the brightness distribution of the area has reached a uniform level. In the actual compensation process, it usually takes 50 to 100 iterative adjustments from detection to completion of compensation.

[0090] In the monitoring of the orientation force stability in the edge area of ​​the photo-alignment film, the area array camera collects 10 frames of images per second, and obtains the brightness distribution changes through real-time processing. When the orientation force weakening phenomenon is detected in 30 consecutive frames of images, compensation is started. The compensation voltage is adjusted in a ramp change manner, and the voltage change rate is limited to 0.5 volts per second. This can avoid overshoot during the orientation force compensation process and ensure smooth orientation adjustment of the photo-alignment film.

[0091] Furthermore, in this embodiment, step S2 also includes:

[0092] Divide the edge area of ​​the photo-alignment film into several small areas, count the average brightness and standard deviation of the pixels in the several small areas, and compare the average brightness and standard deviation of the small areas;

[0093] The small areas whose differences exceed the preset difference threshold are judged as uneven brightness distribution. The unevenness of different small areas is obtained through the spatial differences of brightness distribution in each small area. The proportion of pixels that need compensation correction and the brightness adjustment intensity are determined according to the unevenness.

[0094] Specifically, the total number of pixels in the edge area of ​​the photo-alignment film is obtained, and multiple small areas are obtained according to the fixed-size grid division, and the brightness values ​​of the pixels in the small areas are accumulated to obtain the average brightness of the small areas;

[0095] For small areas, by calculating the standard deviation of the pixel brightness value and the small area brightness mean, if the brightness mean difference between adjacent small areas exceeds the brightness mean difference threshold, or the standard deviation ratio exceeds the set value, the brightness mutation point is obtained;

[0096] According to the brightness mutation point, linear regression is used to calculate the slope of the brightness distribution curve of the small area and the surrounding area, and the small area whose absolute value of the slope exceeds the preset absolute value of the slope is determined as the area to be compensated;

[0097] For the area to be compensated, the proportional-integral controller is set to output a compensation voltage according to the difference between the brightness mean and the reference brightness value, and the area to be compensated is compensated by the compensation voltage until the brightness standard deviation of the area to be compensated is less than a preset brightness threshold;

[0098] Here is an example:

[0099] According to the total number of pixels in the edge area of ​​the photo-alignment film, the area is divided into small grid areas of fixed size 100x100 pixels. The brightness values ​​of the pixels in each small area are accumulated and summed, and the number of pixels is recorded. The average brightness of the small area is obtained by dividing the accumulated sum by the number of pixels and recorded in the brightness average matrix;

[0100] For each pixel in a small area, the sum of the squares of the difference between the brightness value and the mean brightness of the small area is calculated, and then divided by the number of pixels to obtain the standard deviation of the brightness of the small area. If the difference in the mean brightness between adjacent small areas exceeds the set threshold of 20, or the standard deviation ratio exceeds 2.0, then the location is recorded as a brightness mutation point;

[0101] Starting from the brightness mutation point, linear regression is used to calculate the brightness distribution trend of the current small area and the surrounding 8 small areas, and the slope of the brightness distribution curve is calculated. The brightness unevenness of the small area is judged by the slope value. The area with an absolute slope value greater than 0.5 is marked as the area to be compensated;

[0102] According to the difference between the average brightness of the area to be compensated and the reference brightness value, a proportional integral controller with a proportional coefficient of 0.8 is set to output a compensation voltage, with a voltage range of 0 to 5 volts and an adjustment step of 0.1 volt. During the compensation process, the brightness standard deviation of the area to be compensated is continuously collected until it is less than a preset threshold of 5;

[0103] In the implementation of fixed-size grid division, it is reasonable to choose 100x100 pixels as the basic unit. In a 1920x1080 resolution display, the edge area is usually 200 pixels wide, so a 2x10 grid array can be formed at the edge. Each grid contains enough pixels for statistical calculations, while not ignoring local brightness change characteristics due to too large an area.

[0104] The calculation of the mean and standard deviation of the brightness of a small area is done by pixel-by-pixel accumulation. In a 100x100 pixel grid, there are 10,000 pixels to be processed. The brightness value of each pixel ranges from 0 to 255. The brightness values ​​of these pixels are accumulated and divided by 10,000 to get the mean. In practical applications, the mean is usually between 120 and 180, indicating that the edge area is at a medium brightness level.

[0105] The brightness mutation point is determined by comparing adjacent small areas. The central small area is used as the benchmark and compared with the four adjacent areas above, below, left and right. When the mean difference of the adjacent areas exceeds 20 or the standard deviation ratio exceeds 2.0, it indicates that there is a significant brightness jump. In practical applications, a typical edge area may have 3 to 5 brightness mutation points, which often show continuous distribution characteristics.

[0106] When linear regression is used to calculate the brightness distribution trend, a 9-point data set consisting of the central area and the surrounding 8 areas is selected. The brightness change curve is obtained by least squares fitting. The slope of the curve reflects the severity of the brightness change. In practical applications, the absolute value of the slope usually does not exceed 1.0. When the slope exceeds 0.5, it indicates that there is obvious brightness unevenness in the area.

[0107] During the compensation process, the input signal of the proportional-integral controller is the difference between the average brightness of the area to be compensated and the reference value. The proportional coefficient of 0.8 provides a relatively gentle response characteristic. The controller outputs a voltage signal of 0 to 5 volts, which is adjusted step by step with a step size of 0.1 volt. The complete compensation process usually requires 20 to 30 iterations. The compensation is completed when the brightness standard deviation in the area drops below 5.

[0108] The evaluation of brightness uniformity adopts multi-level indicators. The standard deviation at the pixel level reflects the most microscopic uniformity, the mean difference between small areas reflects the uniformity at the mesoscale, and the overall brightness distribution trend reflects the uniformity at the macroscale. This multi-scale evaluation method can comprehensively reflect the display quality. In actual display devices, microscopic uniformity usually requires the standard deviation to be less than 5, mesoscopic uniformity requires the mean difference between adjacent areas to be less than 20, and macroscopic uniformity requires the slope of the brightness distribution curve to be less than 0.5.

[0109] Furthermore, in this embodiment, step S3 specifically includes:

[0110] Factors that affect the brightness distribution in the edge area include the photo-alignment film material properties, liquid crystal properties, and pixel electrode structure;

[0111] Collecting refractive index values, rotation angle values, and electrode width values ​​of detection points arranged in an array in an edge region of the photo-alignment film to obtain a characteristic parameter matrix;

[0112] The refractive index value, the rotation angle value and the electrode width value are respectively compared with the preset threshold value to obtain a comparison result;

[0113] The detection points are grouped to obtain the compensation area, and the corresponding relationship between the refractive index value, the rotation angle value and the pixel voltage value in the compensation area is established through multivariate linear regression to obtain the regression coefficient matrix;

[0114] The driving voltage value of the compensation area is calculated according to the regression coefficient matrix, and the driving voltage value is adjusted according to the standard deviation of the brightness distribution curve by using a proportional integral controller until the standard deviation is less than the standard deviation threshold;

[0115] Here is an example:

[0116] By using an optical detector, 32 detection points are arranged in a 4x8 array in the edge area, with a spacing of 50 microns between each detection point. The refractive index value of the photo-alignment film at each point is collected, ranging from 1.4 to 1.6, the rotation angle value of the liquid crystal molecules, ranging from 0 to 90 degrees, and the pixel electrode width value, ranging from 2 to 5 microns, to form a characteristic parameter matrix;

[0117] For the characteristic parameter matrix, the refractive index threshold is set to 1.5, the rotation angle threshold is set to 45 degrees, and the electrode width threshold is set to 3.5 microns. The detection points are grouped, and the edge area is divided into 4 compensation areas with similar physical characteristics. The pixel brightness value is collected in each compensation area to obtain the brightness distribution curve after compensation.

[0118] The corresponding relationship between the refractive index, rotation angle and pixel voltage in the compensation area is established through multivariate linear regression. The regression coefficient matrix records the weights of various characteristic parameters. The driving voltage value of each compensation area is calculated according to the weights, and the voltage value range is between 0 and 5 volts.

[0119]

[0120] in, represents the regression coefficient matrix, Indicates The compensation area is The weight coefficient of the feature parameter, represents the number of compensation areas, Indicates the number of feature parameters;

[0121]

[0122] in, Indicates the output drive voltage value, Indicates The weight coefficient of the feature parameter, Indicates The input value of the characteristic parameter, represents the bias term, Indicates the total number of feature parameters;

[0123] For each compensation area, the upper limit of the compensation voltage is set to 5V, the lower limit is 0.5V, the step value is 0.1V, and the proportional integral controller with a proportional coefficient of 0.8 adjusts the driving voltage according to the standard deviation of the brightness distribution curve until the standard deviation is less than the preset threshold value of 5;

[0124] The optical detector forms a regular sampling grid with 4x8 array detection points arranged in the edge area. The 50 micron spacing between each point ensures that the sampling density matches the LCD pixel size. In practical applications, the edge area width of the LCD display is between 200 and 400 microns, and the 4x8 detection array can fully cover this area.

[0125] The refractive index value of the photo-alignment film reflects the optical properties of the material. The refractive index range of 1.4 to 1.6 corresponds to the common polyimide alignment film material. When the refractive index is close to 1.5, the alignment film exhibits the best optical anisotropy. Under this condition, the liquid crystal molecules can obtain a stable pre-tilt angle. In actual measurements, the refractive index value is usually concentrated between 1.48 and 1.52.

[0126] The rotation angle of the liquid crystal molecules characterizes the orientation state. 0 degrees corresponds to vertical orientation, 90 degrees corresponds to parallel orientation, and the 45-degree threshold division in the edge area reflects the critical state of orientation abnormality. In actual work, the rotation angle of the normal area is usually maintained between 85 and 90 degrees, while the abnormal area may drop to 30 to 40 degrees.

[0127] The width of the pixel electrode structure determines the electric field distribution. The range of 2 to 5 microns covers the design parameters of mainstream LCDs. The threshold of 3.5 microns corresponds to the most common electrode structure. In actual manufacturing, the electrode width is usually controlled between 3.2 and 3.8 microns. This range can balance the aperture ratio and driving capability.

[0128] In the compensation model established by multivariate linear regression, every 0.01 change in refractive index corresponds to a 0.2 volt voltage adjustment, and every 1 degree change in rotation angle corresponds to a 0.05 volt voltage adjustment. This linear correspondence shows a good correction effect within a small range of deviations. In the actual compensation process, the voltage adjustment amount rarely exceeds 2 volts.

[0129] The compensation voltage is controlled with a fine step of 0.1 volts, and the proportional coefficient of 0.8 provides a smooth adjustment process to avoid compensation overshoot. In practical applications, it usually takes 15 to 20 iterations from detection to completion of compensation, and the whole process takes no more than 1 second. The brightness standard deviation after compensation can usually be reduced to between 3 and 4, showing good uniformity.

[0130] The compensation area of ​​the photo-alignment film is divided into four blocks with similar physical properties. Each block contains eight detection points. This division method balances the control accuracy and complexity. In practical applications, the compensation voltage difference between adjacent blocks usually does not exceed 0.5 volts, ensuring a smooth transition of the display brightness in the edge area.

[0131] Furthermore, in this embodiment, step S4 specifically includes:

[0132] Obtaining a target brightness value, generating a compensation voltage increment matrix according to a difference between the target brightness value and an actual brightness value, and setting a driving voltage in an edge region driving circuit through the compensation voltage increment matrix;

[0133] Acquire brightness data collected by the optical detector array from the edge area of ​​the photo-alignment film and generate a brightness data matrix;

[0134] Calculate the standard deviation of the brightness data matrix. If the calculated standard deviation is greater than a preset standard deviation threshold, increase the voltage increment based on the compensation voltage increment matrix until the standard deviation is less than the preset standard deviation threshold or the maximum number of iterations is reached, so that the brightness area of ​​the edge area of ​​the photo-alignment film is uniform.

[0135] Here is an example:

[0136] Read the compensation parameter matrix of 32x32 pixels in the edge area of ​​the photo-alignment film from the correction parameter database, obtain the target brightness value 160 of each pixel by table lookup, and generate the compensation voltage increment matrix according to the difference between the actual brightness value currently collected and the target brightness value at a step value of 0.1 volt;

[0137] According to the compensation voltage increment matrix, the voltage output value is set in the 32x32 array of the edge area driving circuit, the upper limit of the driving voltage is 5 volts, and the lower limit is 0.5 volts. By looking up the preset compensation curve data table, the driving voltage of each pixel is adjusted step by step by 0.1 volt;

[0138] The brightness data is collected from an 8x8 optical detector array evenly arranged in the edge area. Each detector covers a 4x4 pixel area. The sampling period is set to 10 milliseconds. 100 data points are collected continuously and recorded in a 64x100 brightness data matrix.

[0139] Calculate the standard deviation of the brightness data matrix within a 100 millisecond window. If the standard deviation is greater than a preset threshold of 5, add a 0.1 volt voltage increment based on the current compensation parameters and repeat the compensation process until the standard deviation is less than the preset threshold or the maximum number of iterations is 50.

[0140] The compensation parameter matrix of 32x32 pixels in the edge area forms a fine control grid. Each pixel has independent compensation parameters. This dense parameter distribution can cope with local brightness unevenness. In actual display, when the typical width of the edge area is 1 mm, each pixel occupies about 31 microns, forming a sufficiently fine control granularity.

[0141] The target brightness value of 160 is located in the middle of the grayscale range of 0 to 255. This choice balances the display dynamic range. At this brightness level, the human eye is most sensitive to brightness changes. Therefore, it is particularly suitable as a reference point for uniformity evaluation. In actual measurements, the brightness value of the edge area usually fluctuates between 140 and 180.

[0142] The 0.1 volt step value of the driving voltage provides fine control capability. The upper limit of 5 volts and the lower limit of 0.5 volts cover the typical operating range of the LCD. In the actual compensation process, the driving voltage of most pixels is between 2 and 4 volts. The response of the liquid crystal molecules is most linear within this range.

[0143] The arrangement of the 8x8 optical detector array optimizes spatial sampling. Each detector is responsible for regional monitoring of 4x4 pixels, which not only ensures the spatial resolution of the measurement but also avoids excessive data volume. In actual operation, the 64 detection points can accurately reflect the spatial variation characteristics of the brightness distribution.

[0144] The 10-millisecond sampling period matches the response time of the liquid crystal. The time window formed by continuously collecting 100 data points reaches 1 second. This time span is sufficient to observe the dynamic process of brightness changes. At the same time, the 64x100 data matrix size also balances storage efficiency and data integrity.

[0145] The standard deviation threshold of 5 represents the limit of brightness uniformity that can be perceived by the human eye. When the standard deviation exceeds this threshold, the compensation loop increases the voltage increment of 0.1 volt for correction. This progressive compensation strategy avoids overshoot. The maximum number of iterations of 50 ensures that the compensation process is completed within 5 seconds.

[0146] During the compensation process, the driving voltage difference between adjacent pixels usually does not exceed 0.3 volts. This smooth voltage distribution prevents jagged brightness jumps at the display edge. In actual display, the compensated edge area of ​​the photo-alignment film can achieve a standard deviation of less than 3, showing excellent uniformity.

[0147] Furthermore, in this embodiment, step S4 also includes:

[0148] According to the relative position of the pixel point in the edge area of ​​the photo-alignment film, the horizontal and vertical coordinate values ​​of the corresponding compensation curve are substituted into the compensation curve to obtain the target pixel brightness value after the pixel point is adjusted;

[0149] Compare the difference between the current pixel brightness value and the target pixel brightness value, and adjust the RGB parameters of the pixel point in a gradual manner until the target value is reached;

[0150] Performing brightness compensation on all selected pixels in the edge area of ​​the photo-alignment film and refreshing the display after adjusting a certain number of pixels;

[0151] Specifically, an image of the edge region of the photo-alignment film collected by an optical scanner is obtained, the first pixel point at one end of the image of the edge region of the photo-alignment film is selected as a coordinate origin, and a position mapping table is calculated according to the selected coordinate origin;

[0152] Read the target brightness value from the compensation curve database according to the position mapping table, and calculate the linear mapping relationship between the red, green and blue component values ​​of the current pixel and the target brightness value according to the target brightness value;

[0153] If there is a difference between the target component value and the current component value, the step-by-step adjustment value is calculated according to the ratio of the red, green and blue components, and the step-by-step adjustment value is used to generate a compensation parameter table;

[0154] The compensation parameter table writes compensation data to the display controller, and the display controller refreshes the display cache according to the compensation interval until the red, green and blue component values ​​reach the target values;

[0155] Here is an example:

[0156] An optical scanner with a pixel resolution of 4096x3072 is used to collect an image of the edge area of ​​the photo-alignment film, and the first pixel point in the upper left corner is selected as the coordinate origin. The horizontal and vertical coordinate distances of each pixel point relative to the origin are calculated, and a position mapping table of the edge area of ​​the photo-alignment film is generated at an interval of recording a position value every 10 microns;

[0157] Read the target brightness value corresponding to the position mapping table from the compensation curve database, calculate the linear mapping relationship between the red, green and blue component values ​​of the current pixel and the target brightness value, the value range of the three components of red, green and blue is between 0 and 255, and generate the target component value through linear correspondence;

[0158] According to the difference between the target component value and the current component value, the increase and decrease ratio of the three components of red, green and blue is set to 3:6:1, and the step-by-step adjustment value is calculated for each pixel point. The adjustment range each time does not exceed 8 grayscale values, and a compensation parameter table of the red, green and blue components is generated;

[0159] Write the compensation parameter table to the display controller, set the compensation interval to 5 milliseconds, generate a display cache refresh signal every time 128 pixels are adjusted, and the refresh cycle is 16.7 milliseconds. The compensation parameter update process continues until the red, green and blue component values ​​of all pixels reach the target value;

[0160] The optical scanner's 4096x3072 resolution provides fine pixel acquisition capabilities. Within the 1 mm width of the edge area of ​​the display screen, a position value is recorded every 10 microns, forming 100 sampling points. This dense sampling ensures the accuracy of position mapping. In actual applications, the scanner usually updates data at a frequency of 200 Hz to ensure real-time performance.

[0161] The position mapping of the edge area of ​​the photo-alignment film adopts a relative coordinate system, with the upper left corner pixel point as the origin to establish a rectangular coordinate system, and the horizontal and vertical relative positions are expressed in microns. This representation method is convenient for finding and mapping the compensation curve. In the actual edge area, the horizontal coordinate range is usually 0 to 1000 microns, and the vertical range is 0 to 200 microns;

[0162] The compensation curve database stores the target brightness values ​​corresponding to different positions, usually expressed in 255 grayscale. In the edge area, the target brightness value is gradually distributed from the center to the edge. The center area is maintained at 160 to 180 grayscale, and the edge area is reduced to 140 to 160 grayscale. This gradual distribution avoids sudden changes in brightness.

[0163] The 3:6:1 ratio setting of the red, green and blue components is based on the fact that the human eye is most sensitive to green. In actual compensation, if the target brightness is 10 gray levels higher than the current value, the red component increases by 3 gray levels, the green component increases by 6 gray levels, and the blue component increases by 1 gray level. This ratio ensures color balance.

[0164] Each adjustment during the compensation process is limited to 8 grayscales. This limit value matches the response time of the liquid crystal. Within the compensation interval of 5 milliseconds, the liquid crystal molecules can stably reach the target state. In actual compensation, it usually takes 10 to 15 iterations from recognition to completion of compensation.

[0165] The 16.7 millisecond refresh cycle of the display controller corresponds to a 60 Hz refresh rate. The setting of refreshing every 128 pixels balances visual smoothness and processing load. In actual display, this batch processing mechanism avoids the jump feeling of the picture and also reduces the bandwidth requirement of the data bus.

[0166] The compensation effect of the edge area of ​​the photo-alignment film can be evaluated by the standard deviation. Before compensation, the standard deviation of the brightness of the pixels in the area is usually between 10 and 15. After compensation, it can be reduced to between 3 and 5. The standard deviations of the red, green and blue components are controlled between 2, 4 and 1 respectively, showing good uniformity.

[0167] Furthermore, in this embodiment, step S5 specifically includes:

[0168] The display image is collected by an optical detector, the display image is divided into grid blocks, and the brightness distribution of pixels in the blocks is counted to obtain a contrast matrix of the edge area of ​​the photo-alignment film;

[0169] The contrast difference before and after compensation in the block is calculated according to the contrast matrix, the contrast difference is processed by normalization, and a contrast change map of the edge area of ​​the photo-alignment film is obtained by linear interpolation;

[0170] A linear regression mapping relationship is established according to the contrast change graph, and the contrast change data before and after compensation are input to obtain the contrast score of the overall display area;

[0171] A scoring data buffer is set in the display controller, and the contrast score is recorded in the scoring data buffer. If the contrast score is lower than a preset contrast score threshold, the compensation process is triggered to restart;

[0172] Here is an example:

[0173] The display images before and after edge area compensation are collected by an optical detector with a sampling frequency of 100 Hz, and the collected images are divided into 32x32 pixel grid blocks. The pixel brightness distribution is statistically analyzed using a histogram in each block, and the ratio of the maximum brightness to the minimum brightness is recorded as the block contrast value to generate a contrast matrix of the edge area of ​​the photo-alignment film;

[0174] According to the contrast matrix data, the contrast difference before and after compensation in each block is calculated, the difference data is normalized according to the window size of 8x8 blocks, and the contrast change trend between blocks is calculated by linear interpolation to obtain the contrast change map of the edge area of ​​the photo-alignment film;

[0175] Linear regression is used to establish the mapping relationship between the contrast change of the block and the overall display quality. The contrast change data before and after compensation is input to calculate the contrast score of the overall display area. The contrast score range is set between 0 and 100.

[0176] According to the contrast score, a score data buffer is set in the display controller to record the score value every 10 milliseconds, and 64 data points are recorded continuously. When the score value is lower than the preset threshold of 80, the compensation process is triggered to restart;

[0177] The 100 Hz sampling frequency of the optical detector ensures timely capture of display changes. In practical applications, the response time of LCD displays is usually between 5 and 10 milliseconds. 100 Hz sampling can fully record the brightness change process. In a typical compensation cycle, 10 to 20 frames of image data can be collected.

[0178] The 32x32 pixel grid division forms fine evaluation units in the edge area. Each grid contains 1024 pixels. This scale not only ensures the reliability of statistics, but also avoids excessive calculation. In actual display, the physical size of a grid is about 1 square millimeter, which matches the resolution characteristics of the human eye.

[0179] The block contrast value is calculated by the ratio of the maximum and minimum brightness. This method is particularly sensitive to the extreme values ​​of brightness distribution. In actual measurements, the contrast value of the normal display area is usually between 1.2 and 1.5, while the edge area may drop to between 0.8 and 1.0 when not compensated, and can be restored to the level of 1.3 to 1.4 after compensation.

[0180] The normalized window of 8x8 blocks realizes the feature extraction of mesoscopic scale. The window contains 64 basic grids, covering a display area of ​​about 8 square millimeters. This scale can reflect local changes and smooth random fluctuations. In actual processing, the overlap rate of adjacent windows is set to 50% to ensure the continuity of features.

[0181] The mapping relationship established by linear regression converts local features into an overall score. In the score range of 0 to 100, scores above 90 indicate excellent display effects, scores between 80 and 90 indicate acceptable ranges, and the setting of triggering compensation restart when scores below 80 provides a quality assurance mechanism. In actual operation, the display after compensation can usually be maintained between 85 and 95 points.

[0182] The score buffer of the display controller records data at intervals of 10 milliseconds. The 64 data points constitute an observation window of 640 milliseconds. This time span is sufficient to reflect the stability of display quality. In actual monitoring, the score fluctuation of stable display usually does not exceed 5 points. If there is a fluctuation of more than 10 points, it often indicates a deterioration trend of display quality.

[0183] The real-time nature of the contrast assessment method is reflected in the entire processing chain. The total delay from image acquisition to score output is controlled within 50 milliseconds. This fast response ensures timely detection and correction of display anomalies. During continuous operation, the processing load of the assessment method occupies no more than 5% of the processor time.

[0184] Furthermore, in this embodiment, step S6 specifically includes:

[0185] Acquire a real-time image of the edge area of ​​the photo-alignment film by an optical imager, and generate a compensation point matrix of the edge area according to the real-time image;

[0186] The contrast values ​​between adjacent points are extracted from the compensation point matrix, and the corresponding relationship between edge compensation parameters and contrast changes is obtained by multivariate linear regression.

[0187] A proportional-integral controller is set according to the edge compensation parameter, and the proportional-integral controller outputs a compensation voltage value to the edge area;

[0188] If the display size change exceeds a preset display size change threshold, a horizontal component and a vertical component correction matrix are constructed according to the edge compensation parameter, and the compensation parameter value is updated by linear interpolation;

[0189] Here is an example:

[0190] The real-time images of the edge area of ​​the photo-alignment film are collected by an optical imager, the scanning frequency is set to 200 Hz, the actual horizontal and vertical dimensions of the display area are measured, the proportional relationship between the dimension value and the standard dimension is recorded, and a 32x32 edge area compensation point matrix is ​​generated;

[0191] Extract the contrast values ​​between adjacent points from the compensation point matrix, use multivariate linear regression to calculate the corresponding relationship between edge compensation parameters and contrast changes, the parameter range is between 0.5 and 2.0, the slope of the compensation curve is limited to between 0.2 and 1.0, and establish a point compensation parameter data table;

[0192] For the values ​​in the compensation parameter data table, a proportional-integral controller with a proportional coefficient of 0.8 is set, the output compensation voltage value range is between 0 and 5 volts, the step value is 0.1 volt, the compensation period is set to 10 milliseconds, and the compensation parameter is adjusted when the contrast change value deviates from the preset interval;

[0193] The correction matrix is ​​constructed from the compensation parameter data table. The horizontal component and the vertical component are recorded in two 32x32 data tables respectively. When the display size changes by more than 1%, the parameter update is triggered. The new compensation parameter value is calculated by linear interpolation. The update interval is set to 100 milliseconds.

[0194] The 200 Hz sampling frequency of the optical imager provides real-time monitoring of the size changes of the display area. In actual operation, when the display is bent or vibrated, the size change rate of the edge area can reach 0.5% to 2%. The 200 Hz sampling can capture such tiny changes, and the 32x32 compensation point matrix provides sufficiently dense compensation control points.

[0195] In the compensation model established by multivariate linear regression, the parameter range of 0.5 to 2.0 covers the compensation range of normal operation of the display. In practical applications, the compensation parameters of the edge area are usually concentrated between 0.8 and 1.2. The slope of the compensation curve is limited to between 0.2 and 1.0 to ensure the smoothness of the compensation and avoid sudden changes in the compensation process.

[0196] The proportional-integral controller is designed with a proportional coefficient of 0.8, which shows good stability during the compensation process. The output range of 0 to 5 volts covers the operating voltage range of the LCD, and the step value of 0.1 volt ensures the fineness of the compensation. The compensation period of 10 milliseconds matches the response characteristics of the LCD.

[0197] The correction matrix records the horizontal and vertical compensation parameters respectively, forming two 32x32 data tables. In actual display, the horizontal compensation parameters are often greater than the vertical parameters, which is related to the structural characteristics of the display. The 1% size change trigger threshold balances the response sensitivity and stability, and the 100 millisecond update interval avoids frequent adjustments.

[0198] The relationship between contrast change and size change is particularly evident in the edge area. When the display size changes by 1%, the contrast in the edge area usually decreases by 5% to 10%. Through real-time adjustment of compensation parameters, the contrast drop can be controlled within 2%, and the display quality is maintained.

[0199] The update of compensation parameters adopts linear interpolation calculation to form a smooth transition between adjacent compensation points. In the actual compensation process, the calculation delay of parameter update is controlled within 1 millisecond. The total delay of the entire compensation link from size detection to parameter update does not exceed 15 milliseconds, meeting the real-time requirements.

[0200] The display uniformity of the edge area of ​​the photo-alignment film is closely related to the contrast performance. The dynamic adjustment of the compensation parameters maintains the display uniformity while maintaining a high contrast level. The measured data shows that the contrast of the compensated edge area can reach more than 95% of the central area, and the uniformity deviation is controlled within 3%.

[0201] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific direction or be constructed and operated in a specific direction. Therefore, they cannot be understood as limiting the scope of protection of the present invention.

[0202] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.

Claims

1. An image processing method suitable for a variable-size TFT display screen, characterized in that: The following steps are involved: S1, obtaining the display area size and range of the TFT display screen in real time, and obtaining the position coordinate range of the edge area of ​​the photo-alignment film in combination with a preset fixed percentage of the edge area in the display area; S2, using an image processing method to detect the pixel brightness within the coordinate range of the position corresponding to the edge area of ​​the photo-alignment film, and obtain the pixel brightness distribution characteristics of the edge area of ​​the photo-alignment film; S3, calculating compensation curve correction parameters according to the pixel brightness distribution characteristics and factors affecting the brightness distribution of the edge area; S4, using the compensated brightness distribution curve to correct the pixel brightness in the edge area of ​​the photo-alignment film; S5, after completing the pixel brightness compensation in the edge area of ​​the photo-alignment film, evaluating the overall contrast of the display image to obtain a change value of the display contrast before and after the compensation; S6, by analyzing the relationship between the change value of the display contrast and the pixel brightness compensation intensity, and adjusting the compensation curve parameters of the edge area in real time according to the dynamic change of the display size; The step S1 specifically includes: Obtain the display area brightness matrix according to the output signal of the display controller interface, obtain the brightness value through the sampling points set horizontally and vertically in the display area, and use the least squares method to fit to obtain the display area boundary coordinate point set; Performing pixel mapping transformation on the display area boundary coordinate point set, performing product operation on the preset edge area ratio value and the display area boundary coordinate point set, and obtaining the position coordinate range of the edge area of ​​the photo-alignment film; The optical sensor arranged in the edge area is used to collect the polarized light intensity of the photo-alignment film, and the photo-alignment angle value is obtained by fitting the light intensity curve; According to the light alignment angle value and the preset difference threshold, a proportional integral controller is used to calculate the position compensation amount, and the compensation is performed by a driver arranged in the horizontal and vertical directions until the light alignment angle value difference is less than the difference threshold; The step S3 specifically includes: Factors that affect the brightness distribution in the edge area include the photo-alignment film material properties, liquid crystal properties, and pixel electrode structure; Collecting refractive index values, rotation angle values, and electrode width values ​​of detection points arranged in an array in an edge region of the photo-alignment film to obtain a characteristic parameter matrix; Comparing the refractive index value, the rotation angle value and the electrode width value with preset thresholds respectively to obtain a comparison result; The detection points are grouped to obtain a compensation area, and a corresponding relationship between a refractive index value, a rotation angle value and a pixel voltage value in the compensation area is established by multivariate linear regression to obtain a regression coefficient matrix; The driving voltage value of the compensation area is calculated according to the regression coefficient matrix, and the driving voltage value is adjusted according to the standard deviation of the brightness distribution curve by using a proportional-integral controller until the standard deviation is less than a standard deviation threshold.

2. The image processing method for a variable-size TFT display screen according to claim 1, characterized in that: The step S2 specifically includes: Acquire a first image of the edge area of ​​the photo-alignment film from an area array camera, and obtain a filtered second image by processing the first image with a Gaussian filter; Calculating a brightness threshold for the second image using the maximum inter-class variance method, dividing the brightness value into a plurality of quantization intervals using the brightness threshold, and recording a brightness gradient value of the quantization interval for which a standard deviation ratio of adjacent intervals exceeds a preset threshold; According to the brightness gradient value, a linear fitting calculation is used to obtain a change in the alignment force of the photo-alignment film within the interval, and the change exceeding a change threshold is marked as an alignment force weakening region; A compensation voltage value is calculated for the alignment force weakened area through a proportional integral controller, and the compensation voltage value is determined according to the difference between the brightness value of the alignment force weakened area and the reference brightness value. The brightness value is continuously collected until the brightness standard deviation in the area is less than a preset standard deviation threshold.

3. The image processing method for a variable-size TFT display screen according to claim 2, characterized in that: The step S2 further comprises: Dividing the edge area of ​​the photo-alignment film into a number of small areas, respectively calculating the average brightness values ​​and standard deviations of the pixels in the small areas, and comparing the average brightness values ​​and standard deviations of the small areas; The small areas whose differences exceed the preset difference threshold are judged as having uneven brightness distribution. The unevenness degrees of different small areas are obtained through the spatial differences of brightness distribution in each of the small areas. The proportion of pixel points that need to be compensated and corrected and the brightness adjustment intensity are determined according to the unevenness degrees.

4. The image processing method for a variable-size TFT display screen according to claim 1, characterized in that: The step S4 specifically includes: Acquire a target brightness value, generate a compensation voltage increment matrix according to a difference between the target brightness value and an actual brightness value, and set a driving voltage in an edge area driving circuit by using the compensation voltage increment matrix; Acquiring brightness data collected by an optical detector array from the edge region of the photo-alignment film and generating a brightness data matrix; The standard deviation of the brightness data matrix is ​​calculated. If the calculated standard deviation is greater than a preset standard deviation threshold, a voltage increment is added based on the compensation voltage increment matrix until the standard deviation is less than the preset standard deviation threshold or a maximum number of iterations is reached.

5. The image processing method for a variable-size TFT display screen according to claim 4, characterized in that: The step S4 further comprises: According to the relative position of the pixel point in the edge area of ​​the photo-alignment film, the horizontal and vertical coordinate values ​​of the corresponding compensation curve are substituted into the compensation curve to obtain the target pixel brightness value after the pixel point is adjusted; Compare the difference between the current pixel brightness value and the target pixel brightness value, and adjust the RGB parameters of the pixel point in a gradual manner until the target value is reached; Brightness compensation is performed on all selected pixel points in the edge region of the photo-alignment film, and the display is refreshed after a certain number of pixel points are adjusted.

6. The image processing method suitable for a variable-size TFT display screen according to claim 1, characterized in that: The step S5 specifically includes: Collecting a display image through an optical detector, dividing the display image into grid blocks and counting the brightness distribution of pixels in the blocks to obtain a contrast matrix of the edge area of ​​the photo-alignment film; Calculating the contrast difference before and after compensation in a block according to the contrast matrix, normalizing the contrast difference, and obtaining a contrast change map of the edge area of ​​the photo-alignment film by linear interpolation; A linear regression mapping relationship is established according to the contrast change graph, and contrast change data before and after compensation are input to obtain a contrast score of the overall display area; A scoring data buffer area is set in the display controller, and the contrast score is recorded in the scoring data buffer area. If the contrast score is lower than a preset contrast score threshold, the compensation process is triggered to restart.

7. The image processing method suitable for a variable-size TFT display screen according to claim 1, characterized in that: The step S6 specifically includes: Acquire a real-time image of the edge area of ​​the photo-alignment film by an optical imager, and generate a compensation point matrix of the edge area according to the real-time image; Extracting contrast values ​​between adjacent points from the compensation point matrix, and using multiple linear regression to obtain the corresponding relationship between edge compensation parameters and contrast changes; A proportional-integral controller is set according to the edge compensation parameter, and the proportional-integral controller outputs a compensation voltage value to the edge area; If the display size change exceeds a preset display size change threshold, a horizontal component and a vertical component correction matrix are constructed according to the edge compensation parameter, and the compensation parameter value is updated by linear interpolation.

Images