Pixel calibration method and system of high-precision liquid crystal display screen

Abstract

The invention relates to the field of display screen pixel calibration, in particular to a pixel calibration method and system of a high-precision liquid crystal display screen. The method comprises the following steps: acquiring a real-time playing animation image of a high-precision liquid crystal display screen based on an internal sensor of the display screen; carrying out image scene distortion feature mining on the real-time playing animation image so as to obtain a comprehensive scene distortion feature; performing frame-by-frame pixel optical tracking on the real-time playing animation image, and performing frame-by-frame rolling offset vector calculation so as to construct a pixel point time sequence offset trajectory diagram; carrying out regional dynamic optical flow estimation on the pixel point time sequence offset trajectory diagram, carrying out dynamic trajectory distortion modeling based on comprehensive scene distortion characteristics, and constructing a pixel point offset trajectory distortion model; and carrying out dynamic color change perception on the pixel point offset track distortion model so as to generate a self-adaptive pixel point track offset compensation value. According to the invention, efficient and accurate pixel calibration is realized, and the dynamic visual effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of display screen pixel calibration, and in particular to a high-precision liquid crystal display screen pixel calibration method and system. Background Art

[0002] With the rapid development and widespread application of liquid crystal display technology, liquid crystal displays (LCDs) play an increasingly important role in various electronic devices, and are widely needed in many fields, from mobile phones, televisions, computers to car dashboards. The display effect of LCD screens directly affects the user experience, so its display quality has become one of the important criteria for measuring the performance of LCD screens. In the manufacturing and use of LCD screens, accurate calibration of pixels is crucial to ensure the display effect. Especially in high-precision LCD screens, the uniformity of pixels, color accuracy, and brightness stability are more stringent.

[0003] With the continuous advancement of liquid crystal display technology, high-resolution and ultra-high-definition display effects have become the trend of market development. However, in the production process of high-resolution liquid crystal displays, due to the influence of factors such as manufacturing process, equipment errors, and environmental changes, the deviation between pixels has gradually become the main problem affecting the display effect. These deviations manifest themselves in problems such as color distortion, uneven brightness, and reduced resolution, especially in high-precision displays. Therefore, how to accurately and effectively calibrate these deviations has become the key to improving the display effect of liquid crystal displays.

[0004] At present, the pixel calibration methods of LCD screens mostly rely on manual detection and traditional physical measurement tools. Although these methods can effectively identify and correct display deviations in some cases, the process is cumbersome and inefficient, and it is difficult to meet the needs of modern high-precision LCD display technology. In addition, traditional methods often have certain errors, especially in high-resolution and ultra-high-definition display screens, where subtle pixel deviations are often difficult to completely eliminate by traditional methods. In order to solve this problem, there is an urgent need for an intelligent, efficient and accurate LCD screen pixel calibration method to ensure that pixel calibration can be performed automatically and with high precision during the production process, and that the display effect can be monitored in real time and adjusted in time, so as to ensure that the display quality of the final product meets the standard requirements. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a high-precision pixel calibration method and system for a liquid crystal display screen to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a high-precision pixel calibration method for a liquid crystal display screen, comprising the following steps: Step S1: obtaining a high-precision liquid crystal display screen real-time animation image based on the internal sensor of the display screen; performing image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; Step S2: performing optical tracking of the real-time animation image frame by frame, and performing frame by frame rolling offset vector calculation, so as to construct a pixel point time sequence offset trajectory diagram; Step S3: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map, and performing dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to construct a pixel point offset trajectory distortion model; Step S4: performing dynamic color change perception on the pixel point offset trajectory distortion model, and then performing adaptive pixel point trajectory offset compensation calculation, thereby generating an adaptive pixel point trajectory offset compensation value; Step S5: performing pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; Step S6: Based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, the real-time animation image is subjected to instant pixel deviation calibration, and dynamic visual perception optimization is performed to obtain a dynamic visual optimized display image.

[0007] The present invention mines distortion features of high-precision real-time animation images acquired by sensors inside the display screen, and can fully understand the scene distortion of the image. This process helps to accurately analyze the color cast, blur or distortion problems that may occur when the liquid crystal display screen plays animations. The obtained comprehensive scene distortion features can provide important reference data for subsequent optimization and compensation, thereby accurately identifying the local or overall distortion that may occur in the image during playback. Frame-by-frame pixel optical tracking can track the pixel position of each frame of the image and accurately obtain the trajectory offset information of each pixel, which makes it possible to grasp the changes of each pixel in real time when the animation image is played quickly, and then accurately model its trajectory. Frame-by-frame rolling offset vector calculation can effectively reflect the motion trajectory of each pixel in the image, and provide necessary time series data for subsequent dynamic optical flow estimation and distortion modeling. Regional dynamic optical flow estimation can efficiently estimate the motion flow field of pixels in local areas and capture the motion changes in animation images. Combined with the comprehensive scene distortion features, it is helpful to identify abnormal trajectories caused by distortion during motion. Dynamic trajectory distortion modeling can accurately model these abnormal trajectories, which is a good basis for subsequent distortion compensation. The system provides a theoretical basis and mathematical model for compensation to ensure that the trajectory and motion deformation of each pixel are properly corrected. Dynamic color change perception helps the system understand the color changes on the timeline, so that the color changes can be taken into account when performing trajectory offset compensation to avoid color distortion or color difference problems. Adaptive pixel trajectory offset compensation can intelligently adjust the compensation value for different pixel motion trajectories to ensure that the motion and color changes of each pixel are effectively corrected. By analyzing the color drift evolution of multiple pixel trajectory sub-models, the color accuracy can be further optimized and the color offset can be reduced. Especially in fast-moving scenes, the effect of color fine-tuning is particularly significant. The color fine-tuning parameters of each pixel can accurately correct the color distortion caused by display limitations or motion to ensure color consistency and accuracy. Instant pixel deviation calibration can correct the pixel deviation in the displayed image in real time during playback, thereby avoiding visual distortion caused by time delay or device limitations. Dynamic visual perception optimization ensures that the image seen by the audience always has the best visual effect during the entire display process, eliminating color distortion, offset or other visual interference factors during dynamic playback.

[0008] Preferably, step S1 comprises the following steps: Step S11: obtaining a high-precision liquid crystal display screen to play an animated image in real time based on the internal sensor of the display screen; Step S12: performing time-sequential frame decomposition processing on the real-time animation image, thereby extracting multiple frame images; Step S13: Calculating the refresh frequency of the display screen for multiple frames of images, thereby obtaining a real-time frame rate of the images; Step S14: analyzing the brightness parameter changes of multiple frames of images to obtain image brightness change characteristics; Step S15: mining the image scene distortion features based on the image brightness change features and the image real-time frame rate, thereby obtaining the comprehensive scene distortion features.

[0009] The present invention can obtain the animated image being played on the liquid crystal display screen in real time with high precision through the internal sensor of the display screen. This process can ensure that the collected image data is not interfered by the outside world and reflects the actual image performance of the display screen. The real-time image provides the original image data for subsequent analysis, which can accurately reflect the dynamic changes and display problems of the animated image on the screen. The time-series frame decomposition processing can decompose the animated image into multiple frames of static images in chronological order, so as to capture the image details at each moment, which is the prerequisite for studying the changes of dynamic images and analyzing distortion. By extracting each frame of the image, the detailed performance of the image at different time points can be seen more clearly, especially in scenes with rapid dynamic changes. Frame-by-frame analysis can help identify possible distortion or deviation. By calculating the refresh frequency of multiple frames of images, the real-time frame rate and response performance of the display screen can be understood, which is crucial for evaluating the smoothness and stability of the display screen. If the real-time frame rate does not match the expected value, it may lead to Calculating and adjusting the frame rate can help optimize display performance. By analyzing the brightness changes of multiple frames, the brightness curve of the image can be accurately captured, and the pattern of brightness fluctuations in the image and its impact on the visual effect can be identified. Sharp changes in brightness in animated images may cause the picture to appear unnatural or distorted. By analyzing the brightness change characteristics, it is possible to identify which areas or time periods have brightness distortion problems, and provide targeted data support for subsequent calibration. Combining the brightness change characteristics with the real-time frame rate for analysis can more comprehensively identify the scene distortion characteristics in the image. This not only takes into account the single brightness problem, but also combines the dynamic characteristics of the animated image and the refresh frequency and other factors to form a multi-dimensional distortion feature model. Through the mining of comprehensive features, it is possible to more accurately identify the various types of distortion that may exist in the animated image (such as brightness distortion, color distortion, motion blur, etc.), thereby providing a reliable basis for subsequent pixel-level optimization and compensation.

[0010] Preferably, step S15 specifically comprises the following steps: Perform spatial optical distortion detection based on image brightness change characteristics and image real-time frame rate, thereby generating spatial optical distortion characteristics; Perform color shift statistics according to the spatial optical distortion characteristics to obtain the color shift characteristics of the dynamic image; Performing transient change visual recognition on multiple frames of images to generate transient change visual features of the images; Perform local dynamic image fuzzy analysis based on the transient visual features of the image and extract local dynamic fuzzy features; Based on the local dynamic blur feature, local blur distortion is located for multiple frames of images to generate a local dynamic blur distortion area; Quantifying the distortion distribution of the local dynamic blur distortion area, thereby generating dynamic blur distortion distribution data; The image scene distortion features are mined based on the dynamic image color shift characteristics and dynamic blur distortion distribution data to obtain the comprehensive scene distortion features.

[0011] The present invention can detect the spatial optical distortion problem caused by the physical characteristics of the display (such as response speed, optical contrast, brightness uniformity, etc.) by combining the brightness change characteristics and the real-time frame rate of the image. This distortion is usually manifested as brightness fluctuations or uneven light distribution during dynamic changes. The generated spatial optical distortion characteristics provide targeted numerical support for subsequent image optimization, helping the system to identify and compensate for the distortion caused by optical defects in the display screen. By analyzing the impact of the spatial optical distortion characteristics on color, a color shift feature can be generated to identify the color deviation caused by optical distortion or display characteristics (such as color gamut mismatch, brightness overexposure, etc.) in the image. This process can generate accurate The color shift data can provide a specific reference for subsequent color calibration to ensure that the color shift is minimized during the dynamic image playback process. By identifying the transient changes in the image, the transition effect between different frames of the dynamic image can be accurately captured, especially in scenes with fast motion or fast changes. The transient change visual features can reveal the abnormalities or distortions that may occur in the image during playback. This process can help the system identify subtle changes in dynamic images, especially when the image changes greatly in an instant, and can be identified and optimized in time. By analyzing transient changes, it can identify possible local dynamic blur in the image, especially image blur caused by fast motion or untimely screen refresh. Phenomenon. In dynamic images, blur is not evenly distributed, but usually appears in specific areas. This step can accurately analyze and extract local blur features to ensure that subsequent corrections can be carried out in a targeted manner. By analyzing local dynamic blur features, the specific blurred area can be accurately located in the dynamic image. This step provides a specific area reference for subsequent local optimization, which helps to perform more accurate pixel-level calibration. By identifying and locating the blurred area, pixel-level optimization can be performed in a targeted manner to avoid unnecessary global processing of the entire image, thereby improving calibration efficiency and accuracy. By quantitatively analyzing the local dynamic blur distortion area, blur distortion can be converted into specific This helps to more accurately understand and describe the severity and distribution of blur, and thus provide more efficient guidance for subsequent compensation. By quantifying the distortion data, the corresponding compensation resources can be allocated according to the degree of blur during the calibration process to ensure the best calibration effect. The color shift characteristics of the image and the dynamic blur distortion distribution data are combined and analyzed to obtain the comprehensive distortion characteristics of the image, covering all distortion problems related to dynamic playback in the image (such as color deviation, blur, uneven brightness, etc.). By extracting the comprehensive scene distortion characteristics, the global characteristics of the dynamic image can be optimized, which not only solves local problems, but also ensures the coordination of the global display effect and optimizes the overall visual performance.

[0012] Preferably, step S2 specifically comprises the following steps: Step S21: performing time-series frame fitting on multiple image frames to generate a time-series frame image sequence; Step S22: performing optical tracking of the time-series frame image sequence frame by frame, and extracting the pixel points in each frame image; Step S23: performing pixel offset identification of adjacent frames for pixels in each frame of the image, thereby obtaining inter-frame pixel offset values; Step S24: performing frame-by-frame rolling offset vector calculation based on the inter-frame pixel offset value to generate a time-series frame pixel offset vector; Step S25: Perform temporal visualization of the offset trajectory of the time frame pixel offset vector, thereby constructing a pixel point temporal offset trajectory diagram. Through the time-series frame fitting, the system of the present invention can effectively organize multiple frames of images in time sequence, ensuring that the time relationship between each frame of the image is accurately captured. This is an important basis for dynamic image analysis. By generating a sequence of time-series frame images, it can be ensured that subsequent pixel tracking, offset identification and other processing can be performed based on the real image time sequence, thereby ensuring the accuracy and consistency of the analysis results. Through frame-by-frame optical tracking, all pixel point information in each frame of the image can be accurately extracted. In this way, it can be ensured that the changes in each pixel are fully captured, which helps to discover subtle changes and distortions in the image. The pixel point extraction of each frame of the image is the basis for subsequent offset identification, vector calculation and other steps. Only by accurately extracting the pixel points can the accuracy and effectiveness of subsequent processing be ensured. By comparing the pixel points of adjacent frames, the dynamic changes of the image on the time axis can be identified, thereby extracting the offset of each pixel point. The inter-frame offset value reflects the change of the image content in the time dimension. This step helps to reveal the motion trajectory of each pixel point of the image during playback, can identify the offset of the image at different time points, and then discover the problems of motion distortion, smearing and so on of the image. Through frame-by-frame rolling offset vector calculation, the pixel point offset value of each frame can be converted into vector form. This provides very important motion information for subsequent dynamic correction and optimization of images. After converting the offset value into a vector, the motion trajectory of each frame can be more clearly understood. Especially in dynamic images, the change of the offset vector can reveal the motion mode, speed and other information of the image. By visualizing the offset vector in time series, a dynamic trajectory diagram can be generated to help analyze the motion trajectory of each pixel in the image. This not only provides an intuitive representation of the dynamic changes of the image, but also reveals the motion mode, deviation, etc. of the image during playback. Through the time series offset trajectory diagram, motion distortion, frame skipping or other possible distortion problems in the image can be clearly seen. For image optimizers, this is a very important reference tool that can help them correct the distorted area in a targeted manner.

[0013] Preferably, step S3 specifically comprises the following steps: Step S31: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map to generate regional distribution features of the pixel point offset trajectory; Step S32: calculating the screen switching trajectory offset amplitude based on the pixel point offset trajectory area distribution characteristics of the multiple frame images, thereby generating the pixel point offset trajectory amplitude; Step S33: performing trajectory deviation distribution trend evolution based on the pixel point deviation trajectory amplitude to generate pixel point trajectory deviation trend data; Step S34: Dynamic trajectory distortion modeling is performed on the pixel point trajectory deviation trend data based on the comprehensive scene distortion characteristics to construct a pixel point deviation trajectory distortion model.

[0014] The present invention uses dynamic optical flow estimation to extract dynamic change characteristics of different areas from the pixel point timing offset trajectory diagram, and identify which areas have experienced a large degree of pixel offset during image playback. This can help accurately locate the affected areas in the dynamic image and optimize targeted compensation. Dynamic optical flow estimation can identify the motion patterns of various areas in the image by tracking the motion trajectory of pixel points, revealing problems such as motion blur and ghosting in the image, thereby providing an important basis for subsequent distortion compensation and optimization. By analyzing the image screen switching, the offset amplitude of the pixel points during the screen switching can be calculated. This process can help identify the offset caused by frame skipping, ghosting and other phenomena during image transition or refresh, and provide specific quantitative data for subsequent optimization. Converting the offset amplitude into specific numerical data can provide an accurate reference for subsequent offset compensation, ensuring that the correction process can be targeted at offsets of different amplitudes. By taking appropriate corrective measures and analyzing the evolution trend of the offset amplitude, the offset trend of the pixel points in the dynamic image can be depicted, which helps to reveal the offset phenomenon that gradually intensifies or weakens over time in the image and identify potential dynamic distortion problems. The trajectory offset trend data can provide key information for subsequent dynamic distortion modeling, ensuring that the distortion model can accurately reflect the dynamic changes of the dynamic image, and then make effective compensation. By combining the comprehensive scene distortion characteristics of the image and the evolution trend of the offset trajectory, a more sophisticated dynamic trajectory distortion model can be constructed. The model can comprehensively reflect the various distortion problems (such as offset, blur, color distortion, etc.) that may occur when the display screen plays dynamic images, and provide specific and targeted correction strategies for subsequent image optimization. Through distortion modeling, accurate modeling and compensation of image dynamic distortion can be achieved, avoiding image quality degradation due to excessive or insufficient compensation, and ensuring the best optimization of display effects.

[0015] Preferably, the specific steps of step S4 are: Step S41: performing regional pixel density calculation on multiple frames of images, thereby obtaining pixel density in different regions; Step S42: Dynamically perceive color changes of multiple frames of images, thereby generating picture change color change features; Step S43: performing region segmentation on the pixel point offset trajectory distortion model to generate a plurality of pixel point offset trajectory sub-models; Step S44: performing adaptive pixel point trajectory offset compensation calculation on multiple pixel point offset trajectory sub-models based on the pixel point density of different regions and the color change characteristics of the picture transformation, so as to generate an adaptive pixel point trajectory offset compensation value.

[0016] The present invention can clearly understand the pixel distribution of images in different areas and identify which areas have higher or lower pixel density through regional pixel density calculation, which is helpful for subsequent targeted processing of different areas, especially for areas with uneven pixel density, which can be specially optimized. Understanding the pixel density of each area in the image is helpful for optimizing the fineness of image calibration, avoiding over-compensation of high-density areas or neglect of low-density areas, thereby achieving a more balanced correction effect. Through dynamic color change perception, the color change characteristics in dynamic images can be monitored in real time, and color shift, color aberration and other problems in the image can be identified, which helps to find color distortion in dynamic display. This step can identify color changes caused by frame-to-frame changes, scene transitions, etc. during the playback of dynamic images, which helps to timely adjust the color settings of the display screen to avoid the visual effect of inconsistent colors. By performing regional segmentation on the pixel offset trajectory distortion model, the distortion of the entire image can be decomposed into Sub-models of different regions are formed to facilitate more accurate regional-level optimization. The offset trajectory of each region can be analyzed and corrected independently to avoid the impact of global correction. Multiple sub-models after regional segmentation help to achieve customized regional-level optimization, especially in complex dynamic images. The motion patterns and offset characteristics of different regions may vary greatly. This step can effectively improve the accuracy and detail of image calibration. By combining the pixel density and color change characteristics to compensate for the offset trajectory sub-model of each region, adaptive pixel trajectory offset compensation can be achieved. This adaptive method can ensure dynamic optimization in different regions and different scenes without causing quality degradation in other areas of the picture. By dynamically compensating according to regional density and color changes, distortion in the image can be effectively reduced, especially in scenes such as fast motion and screen switching, avoiding problems such as excessive color offset and motion blur, thereby improving the stability and smoothness of the display effect.

[0017] Preferably, the specific steps of step S5 are: Step S51: performing color quantization after pixel-by-pixel displacement on multiple pixel offset trajectory sub-models, thereby generating pixel-by-pixel color change values; Step S52: performing initial pixel color analysis on the pixels in each frame of the image to obtain initial pixel color features; Step S53: performing pixel color drift evolution on the color change values ​​of each pixel according to the initial pixel color feature to generate a color drift feature for each pixel; Step S54: fine-tuning the color of each pixel point on a pixel-by-pixel basis according to the adaptive pixel point trajectory offset compensation value, thereby generating a color fine-tuning parameter for each pixel.

[0018] The present invention analyzes the position of each pixel in the offset trajectory submodel and quantifies the color change of each pixel, so as to accurately capture the color change of each pixel in the image caused by motion or other factors. This method can help to accurately identify the color error caused by position offset in dynamic scenes. By quantifying each pixel, it is possible to avoid uniformizing the entire image during the compensation process, thereby reducing unnecessary color distortion caused by global compensation. By analyzing the color of the initial pixels of each frame of the image, the color features of each pixel can be extracted, including parameters such as brightness, saturation, and hue. This provides a reference value for subsequent color adjustment and drift compensation, ensuring the accuracy and rationality of compensation. By obtaining the initial color features, the offset of the pixel color can be accurately compared and identified in the subsequent steps, helping to optimize the compensation strategy and reduce errors. By quantifying each pixel based on the initial color features, the color features of each pixel can be extracted. The analysis of color change values ​​can track the color drift process of each pixel in real time, which makes it possible to accurately predict and correct the color change trend of the entire image, thereby effectively avoiding distortion caused by color shift. By generating color drift features, the color drift pattern in dynamic images can be identified, so that targeted color compensation can be performed on different pixels to ensure that during dynamic playback of the image, the color will not drift excessively due to screen refresh or image movement. By fine-tuning the color drift of each pixel in combination with the adaptive pixel trajectory offset compensation value, it can be ensured that each pixel is personalized according to its trajectory offset and color change. This can significantly improve the accuracy of the image and avoid color imbalance caused by global compensation. The fine-tuning parameters of each pixel can be customized according to its specific trajectory and color drift, which can make color restoration more accurate, especially in dynamic scenes, the color restoration effect is significantly improved.

[0019] Preferably, the specific steps of step S6 are: Step S61: performing instant pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel to construct a pixel deviation calibrated display image; Step S62: performing regional display content analysis on the real-time playing animation image to generate display content data of different regions in the image; Step S63: performing human visual perception simulation on the display content data of different regions in the image to obtain human visual perception features; Step S64: dynamically optimizing the pixel deviation calibrated display image according to the visual perception characteristics of the human body to obtain a dynamically optimized display image.

[0020] The present invention can perform instant deviation calibration for each pixel by combining the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel. This step ensures that during the dynamic image playback process, distortion phenomena such as color drift and position offset can be effectively corrected, enhancing the realism and details of the displayed image. Pixel deviation calibration can significantly reduce image artifacts, smears and color distortion caused by display refresh or image motion, and improve image stability and clarity. By analyzing the regional display content of the dynamic image, the content features of different regions in the image can be identified. For example, the central area of ​​the image may contain more moving elements, while the edge area may be relatively static. This step helps to provide targeted information for subsequent processing steps. By generating display content data for different regions, different optimization strategies can be customized for each region to avoid global processing leading to loss of details or improper processing of certain regions. By simulating human visual perception, different regions can be optimized for more natural and comfortable vision according to the characteristics of the human eye. The human eye's perception of different image features (such as brightness, color, contrast, etc.) is not uniform. Therefore, optimization based on human visual perception can make the image display more in line with the actual experience of the viewer and avoid visual fatigue or discomfort. For example, the human visual system has different sensitivities to brightness changes, color contrast, etc. Through human perception simulation, we can avoid areas that are too bright or too dark, optimize the brightness and contrast of the image, and make the image softer and less dazzling when viewed. Through dynamic optimization based on human visual perception characteristics, we can make fine adjustments to elements such as motion and transition in dynamic images to improve the visual coherence and comfort of the image. The optimization effect is not limited to the clarity of static images, but can also improve the visual performance in dynamic situations such as fast movement and scene switching. Through human perception optimization, we can avoid over-enhancing certain elements of the image (such as too bright background, over-sharpened edges, etc.), thereby reducing visual discomfort and fatigue when viewing and improving user experience.

[0021] In this specification, a high-precision liquid crystal display pixel calibration system is provided, which is used to perform the high-precision liquid crystal display pixel calibration method as described above, including: An image distortion module is used to obtain a high-precision LCD screen real-time animation image based on an internal sensor of the screen; perform image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; An offset track module is used to perform frame-by-frame pixel optical tracking on the real-time animation image and perform frame-by-frame rolling offset vector calculation, thereby constructing a pixel point time-series offset track diagram; The trajectory model module is used to estimate the regional dynamic optical flow of the pixel point temporal offset trajectory map, and to perform dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to build a pixel point offset trajectory distortion model; The trajectory offset compensation module is used to perceive the dynamic color change of the pixel point offset trajectory distortion model, and then perform adaptive pixel point trajectory offset compensation calculation to generate an adaptive pixel point trajectory offset compensation value; A color fine-tuning module is used to perform pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; The real-time calibration module is used to perform real-time pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, and perform dynamic visual perception optimization to obtain a dynamic visual optimized display image.

[0022] The present invention can analyze the distortion phenomenon (such as color distortion, geometric distortion, etc.) in the image through the image distortion module, and mine the comprehensive scene distortion characteristics to provide basic data for subsequent image correction. Through in-depth analysis of the scene distortion characteristics, the module can accurately identify the distortion type existing in the image, and provide a directional optimization solution for subsequent offset trajectory, color fine-tuning and other processing, reducing unnecessary global adjustments. By tracking each pixel point in the image frame by frame, the offset of each pixel in the time dimension can be identified. The time-series offset trajectory diagram generated by the trajectory module provides detailed information on the movement of the pixel points, which helps the subsequent trajectory modeling and compensation. The offset trajectory module can clearly display the changes of each pixel in the dynamic image, which is convenient for the subsequent dynamic optical flow estimation and distortion modeling, thereby improving the accuracy of image correction. By estimating the optical flow (i.e., the motion pattern of the pixel) in the image area, the dynamic characteristics of the image area are identified, and the movement and changes of each area in the image can be accurately analyzed and simulated. Based on the distortion characteristics of the image, the trajectory model module can construct a dynamic trajectory distortion model to provide for subsequent offset compensation. Precise mathematical basis ensures accuracy in the distortion repair process. By sensing color changes and dynamic features in dynamic images, the trajectory offset compensation module can dynamically adjust colors according to the motion trajectory of each pixel to ensure a smooth transition of image color changes. Through the adaptive compensation method, the module can generate appropriate trajectory offset compensation values ​​according to the actual conditions of different areas and different pixels to avoid unnatural color changes caused by global correction. By performing independent color fine-tuning on each pixel, the color drift phenomenon in dynamic images can be effectively corrected to ensure that the color of each pixel matches its actual position. The module can identify and repair color deviations caused by offset trajectories, avoid color inconsistencies or drift in the image, and improve the overall display quality. By instantly calibrating pixel deviations in the image and optimizing it in combination with human visual perception characteristics, it ensures that each picture can maintain high-quality visual effects during playback. By sensing the sensitivity of human vision, the brightness, contrast and color of the image are optimized to make the image have the best effect under different viewing conditions and enhance the viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic flow chart of the steps of a pixel calibration method for a high-precision liquid crystal display screen according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0024] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0025] The present application example provides a high-precision LCD pixel calibration method and system. The execution subject of the high-precision LCD pixel calibration method and system includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio image management system, an information management system, and a cloud data management system.

[0026] See also Figures 1 to 4 The present invention provides a pixel calibration method for a high-precision liquid crystal display screen, and the pixel calibration method for a high-precision liquid crystal display screen comprises the following steps: Step S1: obtaining a high-precision liquid crystal display screen real-time animation image based on the internal sensor of the display screen; performing image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; Step S2: performing optical tracking of the real-time animation image frame by frame, and performing frame by frame rolling offset vector calculation, so as to construct a pixel point time sequence offset trajectory diagram; Step S3: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map, and performing dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to construct a pixel point offset trajectory distortion model; Step S4: performing dynamic color change perception on the pixel point offset trajectory distortion model, and then performing adaptive pixel point trajectory offset compensation calculation, thereby generating an adaptive pixel point trajectory offset compensation value; Step S5: performing pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; Step S6: Based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, the real-time animation image is subjected to instant pixel deviation calibration, and dynamic visual perception optimization is performed to obtain a dynamic visual optimized display image.

[0027] The present invention mines distortion features of high-precision real-time animation images acquired by sensors inside the display screen, and can fully understand the scene distortion of the image. This process helps to accurately analyze the color cast, blur or distortion problems that may occur when the liquid crystal display screen plays animations. The obtained comprehensive scene distortion features can provide important reference data for subsequent optimization and compensation, thereby accurately identifying the local or overall distortion that may occur in the image during playback. Frame-by-frame pixel optical tracking can track the pixel position of each frame of the image and accurately obtain the trajectory offset information of each pixel, which makes it possible to grasp the changes of each pixel in real time when the animation image is played quickly, and then accurately model its trajectory. Frame-by-frame rolling offset vector calculation can effectively reflect the motion trajectory of each pixel in the image, and provide necessary time series data for subsequent dynamic optical flow estimation and distortion modeling. Regional dynamic optical flow estimation can efficiently estimate the motion flow field of pixels in local areas and capture the motion changes in animation images. Combined with the comprehensive scene distortion features, it is helpful to identify abnormal trajectories caused by distortion during motion. Dynamic trajectory distortion modeling can accurately model these abnormal trajectories, which is a good basis for subsequent distortion compensation. The system provides a theoretical basis and mathematical model for compensation to ensure that the trajectory and motion deformation of each pixel are properly corrected. Dynamic color change perception helps the system understand the color changes on the timeline, so that the color changes can be taken into account when performing trajectory offset compensation to avoid color distortion or color difference problems. Adaptive pixel trajectory offset compensation can intelligently adjust the compensation value for different pixel motion trajectories to ensure that the motion and color changes of each pixel are effectively corrected. By analyzing the color drift evolution of multiple pixel trajectory sub-models, the color accuracy can be further optimized and the color offset can be reduced. Especially in fast-moving scenes, the effect of color fine-tuning is particularly significant. The color fine-tuning parameters of each pixel can accurately correct the color distortion caused by display limitations or motion to ensure color consistency and accuracy. Instant pixel deviation calibration can correct the pixel deviation in the displayed image in real time during playback, thereby avoiding visual distortion caused by time delay or device limitations. Dynamic visual perception optimization ensures that the image seen by the audience always has the best visual effect during the entire display process, eliminating color distortion, offset or other visual interference factors during dynamic playback.

[0028] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a pixel calibration method for a high-precision liquid crystal display screen of the present invention. In this example, the steps of the pixel calibration method for a high-precision liquid crystal display screen include: Step S1: obtaining a high-precision liquid crystal display screen real-time animation image based on the internal sensor of the display screen; performing image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; In this embodiment, the high-precision sensors (such as light sensors, temperature sensors, motion sensors, etc.) built into the liquid crystal display are used to monitor the animated images being played in real time. These sensors can capture information such as screen brightness, color temperature, and color distribution. The sensors of the display screen are connected to a computer or a data processing unit to ensure that the sensor data can be read in real time. This can be achieved through serial communication, a USB interface, or other data transmission protocols. While the sensor is acquiring data, an image acquisition card or related software (such as OpenCV) is used to capture the animated images played on the display screen in real time to ensure that the captured images are of high quality and without delay to ensure the accuracy of subsequent analysis. The captured images are format converted and preprocessed to ensure that they are suitable for subsequent distortion feature analysis, which may include image resizing, denoising, and color correction. The scene distortion features of the acquired real-time animated images are mined. This step can be achieved by the following method: Apply image processing techniques (such as SSIM, PSNR, etc.) to evaluate the degree of image distortion. These algorithms can compare the similarity between the original image and the captured image and identify the distortion area. domain, use computer vision methods (such as edge detection, texture analysis, etc.) to extract important features in the image, including color distribution, brightness changes, contrast, etc. These features can help analyze the overall quality of the image, perform statistical analysis on the extracted features, and calculate the comprehensive scene distortion features, including the average value and standard deviation of the distortion degree, which will help quantify the distortion of the image, integrate the mined distortion features, and generate a comprehensive scene distortion feature data structure. The data structure should contain the values ​​of each distortion feature and related image information (such as timestamp, frame number, etc.). For each frame of the image, record its comprehensive scene distortion features for subsequent dynamic analysis and optimization. You can use a database or file system to store this data to ensure the persistence and queryability of the data, output the analysis results of the comprehensive scene distortion features, and generate a visualization report. The visualization can display the distribution of distortion features in the form of charts, heat maps, etc., to help identify image quality problems, and associate the result data with the real-time animated image to ensure that the data can be quickly accessed and analyzed in subsequent steps.

[0029] Step S2: performing optical tracking of the real-time animation image frame by frame, and performing frame by frame rolling offset vector calculation, so as to construct a pixel point time sequence offset trajectory diagram; In this embodiment, an image acquisition card or related software (such as OpenCV) is used to capture each frame of the animated image in real time, ensuring that the captured frame rate is high enough to avoid motion blur or image delay. The common frame rate is 30fps or 60fps. Each frame of the image is preprocessed, including denoising, normalization, and resizing, so that the image is suitable for subsequent optical tracking analysis. This can be achieved through filtering techniques (such as Gaussian filtering) to reduce noise interference in the image. For each frame of the image, an optical flow algorithm (such as the Lucas-Kanade method or the Horn-Schunck method) is used to perform frame-by-frame pixel tracking. These algorithms can estimate the pixel motion between consecutive frames and generate a motion vector for each pixel. When implementing optical tracking, the gradient information of the image is first calculated to identify the motion area. The speed and direction of each pixel are calculated by comparing the pixel value changes of adjacent frames. For each pixel, The rolling offset vector is calculated between adjacent frames, that is, the displacement of the pixel in the time series. The offset vector = the current frame pixel position − the previous frame pixel position. The offset vector of each pixel is recorded in an array. The offset vector not only contains the size of the displacement, but also includes the direction of movement information. The rolling offset vectors of all pixels are integrated to generate a time series offset trajectory diagram, which can show the motion trajectory of each pixel in the time dimension and help visualize the motion pattern. The offset trajectory diagram can be visualized using image processing technology, such as using color coding to represent different motion speeds and directions to generate a dynamic trajectory diagram. The time series offset trajectory data of each pixel is saved to a database or file system for subsequent analysis to ensure that the data structure is clear and easy to query and process. The constructed offset trajectory diagram is statistically analyzed to identify the main motion patterns and features in the image, providing a basis for subsequent image optimization.

[0030] Step S3: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map, and performing dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to construct a pixel point offset trajectory distortion model; In this embodiment, the temporal offset trajectory map is divided into regions, which can adopt a content-based segmentation method, such as k-means clustering or an image segmentation algorithm (such as SLIC superpixel segmentation), to ensure that the pixels in each region have similar dynamic characteristics, and adopt a dynamic optical flow estimation algorithm (such as Horn-Schunck or Lucas-Kanade method) to estimate the optical flow in each region. This method can generate an optical flow field for each region by calculating the pixel motion between adjacent frames, record the speed and direction of the optical flow, and extract key features from the estimated optical flow field, including the amplitude and direction of the optical flow. The average speed, maximum speed, and speed distribution of the optical flow in each region can be calculated to identify the main motion direction and dynamic characteristics in the region, and these features are combined with the temporal offset trajectory to form a data structure containing motion information and optical flow features for subsequent analysis. Combined with the comprehensive scene distortion features obtained in the previous step, the optical flow field is divided. Analysis, comprehensive scene distortion features can provide the degree and type of image distortion in different areas (such as blur, color distortion, etc.). According to the distortion features, the weight of optical flow estimation is adjusted. For example, for areas with a high degree of distortion, the uncertainty of optical flow estimation can be increased to reflect the actual motion state. Based on regional dynamic optical flow estimation and comprehensive scene distortion features, a dynamic trajectory distortion model is constructed. The model should consider the relationship between the change of optical flow and the distortion features. Machine learning methods (such as regression analysis or neural networks) can be used to establish the model. The input is optical flow features and distortion features, and the output is the distorted trajectory. This model can simulate the distortion of pixel offset in actual scenes. To verify the constructed trajectory distortion model, the actual captured image data can be used for testing, and the difference between the trajectory output by the model and the actual observed trajectory can be compared. The model is adjusted and optimized according to the verification results to ensure that the model can accurately reflect the dynamic features and distortion effects in the image.

[0031] Step S4: performing dynamic color change perception on the pixel point offset trajectory distortion model, and then performing adaptive pixel point trajectory offset compensation calculation, thereby generating an adaptive pixel point trajectory offset compensation value; In this embodiment, the pixel offset trajectory distortion model constructed previously is used to analyze the color change of each pixel in the dynamic scene. This can be achieved by comparing the change of color value frame by frame. Specifically, for each pixel, its color change between adjacent frames is calculated. This can be achieved by calculating the color difference (such as the difference in RGB or HSV color space) color change = | current frame color - previous frame color |. By analyzing the amplitude and direction of the color change, the color change area with a greater impact on visual perception in the dynamic scene is identified, and these change characteristics are recorded for subsequent compensation calculations. Before performing the adaptive compensation calculation, the results of dynamic color change perception are integrated with the previous distortion model data to ensure that the color change information of each pixel corresponds to its offset trajectory distortion characteristics. At this time, focus on those areas with significant color changes. These areas are usually more susceptible to motion blur and distortion. According to the characteristics of dynamic color changes and the distortion model, an adaptive offset compensation value is calculated. The compensation value = ∑(color change×distortion feature) / regional weight. In this formula, the weights of color change and distortion feature can be dynamically adjusted according to their impact on the human eye. For areas with larger color changes, higher weights are given to ensure that their importance is reflected in the compensation calculation. After the calculation is completed, the adaptive pixel trajectory offset compensation value is recorded in a data structure. The compensation value of each pixel should include its displacement information and its color change characteristics in the dynamic scene, forming a mapping table of comprehensive compensation values ​​for rapid search and application in subsequent steps. The generated adaptive pixel trajectory offset compensation value can be verified by comparing it with the dynamic changes in the real scene to evaluate the accuracy and effectiveness of the compensation value. According to the verification results, the parameters and weights of the compensation calculation are adjusted to improve the accuracy and reliability of the model.

[0032] Step S5: performing pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; In this embodiment, first, multiple pixel offset trajectory sub-models are extracted from the previous steps. These sub-models record the displacement and corresponding color information of each pixel in different time frames, ensuring that the data structure of these models is clear and easy to access. In this process, the historical color value and displacement information of each pixel are integrated into an array or dictionary for subsequent processing. For each pixel, its color change in the time series is analyzed, and its color drift is calculated. Color drift is a color change caused by factors such as motion and lighting changes. By comparing the color values ​​of adjacent frames, the color drift is calculated. Color drift = current frame color − previous frame color. The color drift value of each pixel is recorded to ensure that the color change trend of each pixel in the dynamic scene can be reflected. Based on the calculation result of the color drift, the color of each pixel is fine-tuned. The purpose of the fine-tuning is to adjust the color of the pixel. The color value is used to enhance the overall visual effect. A weighted fine-tuning algorithm is used to make adjustments based on the drift amplitude of each pixel. The color value after fine-tuning = original color value + k × color drift. After fine-tuning, the color fine-tuning parameters of each pixel are recorded. These parameters should include the original color value, the color value after drift, and the color value after fine-tuning. These fine-tuning parameters are integrated into a data structure for quick access in subsequent steps of image processing. Dictionaries or arrays can be used to organize these parameters to ensure the readability and ease of use of the structure. The generated color fine-tuning parameters are verified. By comparing the images before and after fine-tuning, the effectiveness of fine-tuning and the improvement of visual effects are evaluated. Based on the verification results, the parameters of the fine-tuning algorithm are adjusted. If necessary, more complex algorithms (such as regression analysis based on machine learning) can be introduced to further optimize the fine-tuning process to ensure that the color performance of the final image is optimal.

[0033] Step S6: Based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, the real-time animation image is subjected to instant pixel deviation calibration, and dynamic visual perception optimization is performed to obtain a dynamic visual optimized display image.

[0034] In this embodiment, the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel are extracted from the previous steps. These data should include the displacement of each pixel and the adjusted color value to ensure a clear data structure and facilitate subsequent processing and application. Arrays or dictionaries can be used to organize these data so that the compensation value and the fine-tuning parameter of each pixel can be easily corresponded one by one. For each frame of the real-time playback animation image, the adaptive pixel point trajectory offset compensation value is applied pixel by pixel. Specifically, for each pixel, the new pixel position is calculated: the new position = the original position + the compensation value, and the color value is updated: the new color value = the original color value + the color fine-tuning parameter. After completing the pixel deviation calibration, dynamic visual perception optimization is performed. This step aims to further improve the image effect by analyzing the perception characteristics of the human visual system to the image. The dynamic visual The visual perception model evaluates the visual performance of the calibrated image in dynamic scenes. The HVS (human visual system) model can be used to analyze the impact of factors such as image brightness, contrast, and color saturation on visual perception. According to the evaluation results, the brightness, contrast, and color distribution of the image are adjusted. For example, for areas with poor visual perception in dynamic scenes, the contrast and saturation are appropriately enhanced to improve their attractiveness. The calibrated and optimized image frames are integrated into a dynamic stream for subsequent display and processing. This can be achieved by compressing the image frames and saving them as video files through video encoding technology to ensure image quality and smooth playback. During the output process, ensure that the image format is compatible with the target playback device to ensure the best display effect. You can choose to use standard video formats such as H.264 or HEVC to balance quality and file size.

[0035] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: obtaining a high-precision liquid crystal display screen to play an animated image in real time based on the internal sensor of the display screen; Step S12: performing time-sequential frame decomposition processing on the real-time animation image, thereby extracting multiple frame images; Step S13: Calculating the refresh frequency of the display screen for multiple frames of images, thereby obtaining a real-time frame rate of the images; Step S14: analyzing the brightness parameter changes of multiple frames of images to obtain image brightness change characteristics; Step S15: mining the image scene distortion features based on the image brightness change features and the image real-time frame rate, thereby obtaining the comprehensive scene distortion features.

[0036] In this embodiment, in order to achieve high-precision real-time animation playback, it is necessary to install high-precision sensors inside the LCD screen. These sensors may include optical sensors and motion sensors to capture detailed information of the animated images played on the display screen. Select a suitable high-precision sensor, such as a high-resolution camera or a photoelectric sensor, that can capture dynamic images on the LCD screen in real time. Ensure that the sensor has sufficient sampling rate and resolution to provide clear image data. When installing the sensor, ensure that its position can fully cover the display area to avoid the influence of shadows and reflections on data capture. Set up a data acquisition system through programming (such as using Python or C++) to record the animated images captured by the sensor in real time. The data should be stored in an appropriate format (such as JPEG or PNG) for subsequent processing. Configure real-time data transmission to ensure that the image data can be transmitted to the processing unit without delay. Run the data acquisition system to record the animated images played on the LCD screen in real time. Ensure that the system can continue to work under different playback conditions (such as brightness and contrast changes). Regularly check the working status of the sensor to ensure the accuracy and stability of data acquisition. Perform time-series frame decomposition on the real-time animated images to extract individual frames. Image, which will provide basic data for subsequent frame rate calculation and brightness change analysis. Select appropriate image processing algorithms for time-series frame decomposition. Common methods include background subtraction and optical flow. These methods can effectively extract single-frame images from video data. Ensure that the selected algorithm can handle high-frequency dynamic changes and adapt to the refresh characteristics of LCD screens. Use image processing libraries (such as OpenCV) to read real-time animated image data, convert the acquired dynamic image data into a processable format, record the timestamp of each frame, so as to analyze the timing changes later, and apply the selected frame decomposition algorithm to extract animation frame by frame. The extracted images are stored locally or in a database to ensure data integrity and the efficiency of the frame extraction process for real-time analysis and subsequent processing. The timestamp information of each frame is read from the extracted multi-frame images, which will be used to calculate the time interval between each frame to ensure the accuracy of the timestamp for subsequent calculations. The time interval between each pair of adjacent frames can be calculated using a simple time difference calculation formula: frame interval = current frame timestamp − previous frame timestamp. All frame intervals are recorded in a list or array for subsequent calculation of the frame rate. The real-time frame rate is calculated using the frame interval data. The frame rate = 1 / average frame interval. The calculated frame rate is compared with the standard frame rate of the display to evaluate the display effect. The brightness parameter changes of the extracted multi-frame images are analyzed to identify the image brightness characteristics, and then the stability of the display effect is evaluated. The appropriate brightness calculation method is selected, such as converting RGB values ​​to grayscale values, or directly calculating the average brightness of each frame. The formula is used to convert RGB to brightness: brightness = 0.299×R+0.587×G+0.114×B, calculate the brightness of each frame of the extracted image, record the brightness value of each frame, which can be achieved by traversing each pixel and applying the above formula, store the brightness value of each frame in an array or list for subsequent analysis, analyze the brightness value change trend with the frame, calculate the average value, standard deviation and maximum and minimum values ​​of the brightness to evaluate the brightness stability of the image, visualize the brightness change data, generate a brightness change curve chart to help identify potential problems, integrate the image brightness change characteristics and real-time frame rate, and mine the image scene distortion characteristics to evaluate the display quality of the LCD screen, integrate the brightness change characteristics and real-time frame rate data into a comprehensive feature set, and the characteristics of each frame can include the brightness mean, standard deviation and frame rate, etc., to ensure that the format of feature data is unified for subsequent analysis, and to select appropriate distortion feature analysis methods, such as structural similarity index (SSIM) or peak signal-to-noise ratio (PSNR), to evaluate the degree of image distortion. These methods can quantify changes in image quality and compare them with standard images. Distortion features are calculated for each frame of the image, and the similarity between the current frame and the previous frame is compared. Each frame is analyzed using the SSIM or PSNR formula, and the calculation results are recorded in an array. The change trend of the distortion features is analyzed, and the comprehensive scene distortion features, including the distortion index and change trend of each frame, are output, and a visual report is generated, which will provide an important basis for the performance evaluation and quality control of LCD screens. .

[0037] In this embodiment, the specific steps of step S15 are: Perform spatial optical distortion detection based on image brightness change characteristics and image real-time frame rate, thereby generating spatial optical distortion characteristics; Perform color shift statistics according to the spatial optical distortion characteristics to obtain the color shift characteristics of the dynamic image; Performing transient change visual recognition on multiple frames of images to generate transient change visual features of the images; Perform local dynamic image fuzzy analysis based on the transient visual features of the image and extract local dynamic fuzzy features; Based on the local dynamic blur feature, local blur distortion is located for multiple frames of images to generate a local dynamic blur distortion area; Quantifying the distortion distribution of the local dynamic blur distortion area, thereby generating dynamic blur distortion distribution data; The image scene distortion features are mined based on the dynamic image color shift characteristics and dynamic blur distortion distribution data to obtain the comprehensive scene distortion features.

[0038] In this embodiment, the previously extracted image brightness change features (such as mean, standard deviation) and real-time frame rate data are integrated into a data set. Ensure that each frame of the image contains the corresponding brightness and frame rate information for comprehensive analysis. Use a data frame (such as pandas DataFrame) to organize the integrated data for subsequent processing. Select a suitable spatial optical distortion detection algorithm, such as a gradient-based edge detection algorithm or wavelet transform. Edge detection can help identify distorted areas in an image, while wavelet transform can effectively capture changes in different frequencies. Ensure that the selected algorithm can handle dynamic image features and can adapt to different display conditions. Implement the selected detection algorithm and analyze the integrated data. Identify potential optical distortion by calculating the relationship between the brightness change and the frame rate of each frame of the image. Record the specific location and degree of distortion and generate optical distortion feature data. These features will be used for subsequent color shift statistics. Extract RGB color values ​​from each frame of the image and calculate the color features of each pixel. This can be achieved by traversing each pixel of the image and recording its RGB value. Integrate the extracted color data into a list or array for subsequent analysis. Calculate the average color value of each frame and compare it with the previous frame to calculate the color shift. Record the color shift value of each frame and associate it with the spatial optical distortion characteristics. Perform statistical analysis on the color shift values ​​to calculate the average color shift, maximum, minimum, and standard deviation. These statistics can be used to evaluate the stability and consistency of color. Generate a visualization chart to show the trend of color shift over time to help identify potential problem areas. Select an appropriate transient change recognition algorithm, such as optical flow or motion estimation. These methods can effectively identify rapidly changing parts of the image. Make sure that the selected algorithm can adapt to different lighting conditions and motion patterns. Prepare the extracted multi-frame image data to ensure that the data format is uniform and can be directly used for algorithm processing. Load the image data into the corresponding processing framework (such as OpenCV) to run the transient change recognition algorithm. Apply the selected recognition algorithm to analyze the image frame by frame and extract transient change features. These features can include motion speed, change direction, and change amplitude. Record the identified transient change features in a data frame for subsequent analysis. Select an appropriate blur detection algorithm, such as Laplace operator or Fourier transform. These algorithms can effectively identify blur features in images. Make sure the selected algorithm can handle dynamic images and adapt to rapidly changing scenes. Combine the extracted transient visual features with the blur detection algorithm, analyze the image frame by frame, and identify the blurry area. Calculate the blur level for each local area and record the blur feature values, including blur level and blur direction. Record the location and characteristics of the blurry area in each frame of the image, and integrate the data in a table for subsequent analysis. Generate a visual chart to show the blurry area and its changing trend in each frame of the image. Select a suitable positioning algorithm, such as a region growing-based algorithm or cluster analysis.These algorithms can effectively identify blurry and distorted areas and accurately locate them. Ensure that the selected algorithm can handle different degrees of blur and distortion and adapt to dynamically changing scenes.

[0039] Apply the selected positioning algorithm to analyze the local dynamic blur characteristics frame by frame and identify the blurred and distorted areas. Record the location, size and blur degree of each blurred and distorted area to provide data support for subsequent analysis. Integrate the data of the blurred and distorted areas into a table to facilitate subsequent statistical analysis. Generate a visual chart to show the distribution and dynamic changes of the blurred and distorted areas. Select a suitable quantification method, such as histogram analysis or distribution function. These methods can effectively quantify the distribution of blurred and distorted areas. Ensure that the selected method can adapt to situations with different blur levels and provide comprehensive analysis results. Input the identified blurred and distorted area data into the selected quantification method to calculate the distortion distribution characteristics. Record the characteristic values ​​of each blurred area, including blur degree, distribution frequency, etc. Integrate the quantified distortion distribution data into a table to facilitate subsequent analysis and comparison. Generate a visual chart to show the distribution and statistical results of the blurred and distorted areas. Output dynamic blur distortion distribution data, including the distribution characteristics of each blurred area, and generate relevant visual reports to help evaluate the quality of the display effect. Combine the dynamic image color shift characteristics and dynamic blur distortion distribution data to mine image scene distortion characteristics and obtain comprehensive scene distortion characteristics. Combine color shift features and motion blur distortion distribution data into a comprehensive feature set for further analysis. Ensure that the format of feature data is unified to facilitate subsequent processing and analysis. Select appropriate mining methods, such as principal component analysis (PCA) or cluster analysis. These methods can effectively discover potential patterns and relationships in the data. Ensure that the selected method can handle different types of feature data and provide clear analysis results. Apply the selected mining method to analyze the integrated feature data and identify potential distortion features. Record the contribution and impact of each feature for subsequent evaluation and optimization of display effects. Output comprehensive scene distortion feature data, including the main distortion features and their changing trends, and generate relevant visualization reports to show the dynamic changes of distortion features.

[0040] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing time-series frame fitting on multiple image frames to generate a time-series frame image sequence; Step S22: performing optical tracking of the time-series frame image sequence frame by frame, and extracting the pixel points in each frame image; Step S23: performing pixel offset identification of adjacent frames for pixels in each frame of the image, thereby obtaining inter-frame pixel offset values; Step S24: performing frame-by-frame rolling offset vector calculation based on the inter-frame pixel offset value to generate a time-series frame pixel offset vector; Step S25: Perform temporal visualization of the offset trajectory of the time frame pixel offset vector, thereby constructing a pixel point temporal offset trajectory diagram. In this embodiment, it is ensured that a series of continuous time-series frame images have been obtained. These images can be extracted from a video stream or a static image sequence. The images should have the same resolution and size for subsequent processing. An image processing library (such as OpenCV) is used to load the image sequence into memory for frame-by-frame analysis. For each frame of the image, its pixel data is traversed, and the rows and columns of each pixel are traversed through nested loops to extract the RGB value or grayscale value of each pixel. The extracted pixel data is stored in a suitable data structure (such as a list or array) for further analysis of each frame of the image. The pixel point data of each frame of the image is organized into a multidimensional array, and each element represents the color information and position of a pixel. This structure is convenient for subsequent offset identification and analysis, and the position information of each pixel (such as x, y coordinates) in order to track the movement of pixels between adjacent frames. After completing the frame-by-frame pixel extraction, the next step is to identify the offset of adjacent frames for the pixels in each frame. The goal of this process is to calculate the movement of each pixel between consecutive frames and identify the offset value of the pixel. For each pair of adjacent frames, the pixel of the current frame needs to be matched with the pixel of the previous frame. The optical flow method (such as the Lucas-Kanade optical flow algorithm) can be used. The algorithm estimates the motion vector by using the pixel intensity change in the local neighborhood. The offset value of each pixel is obtained by calculating the position change of each pixel in the adjacent frames. In the implementation process, the adjacent frames are first preprocessed, including denoising and grayscale, to improve the accuracy of matching. When using the optical flow method, a suitable window size needs to be selected so that feature matching can be performed in a sufficient area. For each matched pixel, its offset in the x and y directions is calculated and these offset values ​​are recorded. For each pixel, the offset vector is defined as the displacement from the previous frame to the current frame. The offset vector can be represented by a two-tuple: offset vector = (dx, dy);Among them, dx and dy represent the offset in the x and y directions respectively. Traverse all pixel points and calculate the offset vector of each pixel based on the offset value identified in the previous step. Store the offset vectors of all pixels in a three-dimensional array. The first dimension represents the frame number, and the second and third dimensions represent the pixel position. If the offset value of a pixel is less than the set threshold, it can be regarded as stationary and its offset vector is set to zero. After calculating the offset vector of each frame, the overall movement trend can be further analyzed, such as calculating the average offset vector of all pixels, identifying the direction and amplitude of the overall movement, and outputting the offset vector data of each frame into a structured data format for subsequent processing and visualization. Select a suitable visualization tool or library, such as Matplotlib, Seab orn et al., implemented the visualization of the offset trajectory in the Python environment, determined the basic settings of the chart, including the coordinate axis, title, and legend, etc., in order to clearly display the data, traversed the pixel offset vector of each frame, calculated the movement trajectory of each pixel point on the two-dimensional plane, and drew the movement trajectory of each pixel by connecting the initial position and the offset position of the pixel. Different colors or marks were used to represent different pixels to highlight the change of the offset. In order to better show the change of the trajectory, a dynamically updated visualization effect can be achieved. For example, the offset of each frame of the image is displayed through animation to help users intuitively understand the dynamic change process of the image. Interactive functions are added to the visualization interface to allow users to select specific frames for detailed observation. Finally, for each pair of adjacent frames, a list of offset values ​​is generated to record the movement of all pixels. These offset values ​​will become the basis for the subsequent frame-by-frame scrolling offset vector calculation. ;

[0041] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map to generate regional distribution features of the pixel point offset trajectory; Step S32: calculating the screen switching trajectory offset amplitude based on the pixel point offset trajectory area distribution characteristics of the multiple frame images, thereby generating the pixel point offset trajectory amplitude; Step S33: performing trajectory deviation distribution trend evolution based on the pixel point deviation trajectory amplitude to generate pixel point trajectory deviation trend data; Step S34: Dynamic trajectory distortion modeling is performed on the pixel point trajectory deviation trend data based on the comprehensive scene distortion characteristics to construct a pixel point deviation trajectory distortion model.

[0042] In this embodiment, a suitable dynamic optical flow algorithm is selected, such as the Lucas-Kanade method or the Horn-Schunck method. These algorithms can effectively estimate the motion vector of each pixel in the image sequence and calculate the optical flow field. Ensure that the selected algorithm is suitable for processing dynamic image data and can work accurately under different lighting and motion conditions. For the extracted pixel point time series offset trajectory diagram, the selected optical flow algorithm is used to calculate the motion vector of each pixel. This involves comparing the pixels between adjacent frames and identifying their displacement. Generate an optical flow field diagram to represent the movement of each pixel in the time series. Each vector in the optical flow field represents the direction and amplitude of the movement of the pixel. According to the calculated optical flow field, the motion characteristics of the pixels in the area are further analyzed. The image can be divided into multiple regions (such as by cluster analysis) to calculate the optical flow characteristics of each region. Record the motion characteristics of each region, including the mean, standard deviation and direction of the offset, and generate the regional dynamic optical flow estimation result. The calculation method for determining the trajectory offset amplitude can usually be achieved by calculating the displacement amplitude of all pixels in the region before and after switching. The trajectory offset amplitude can be expressed by the formula: offset amplitude = , where Δx and Δy represent displacements on the x-axis and y-axis, respectively. Traverse the pixel points in each region, calculate the displacement of each pixel between frames, and calculate the offset amplitude of each pixel according to the above formula. Summarize the offset amplitude of each region, calculate the average offset amplitude and total offset amplitude of the region, and record these data for subsequent analysis. Integrate the calculated offset amplitude data into a table to facilitate statistical analysis and subsequent processing. Generate statistical charts to show the trend of offset amplitude changes when different screens are switched. Determine the distribution features to be analyzed, such as the range of variation, distribution frequency, and average offset of the offset amplitude. These features can help understand the evolution of the trajectory. Use statistical methods (such as histograms and quantile analysis) to perform distribution analysis on the offset amplitude and identify the trend of offset amplitude changes. Arrange the offset amplitude data in chronological order and analyze the changes in offset amplitude in different time periods. Time series analysis methods (such as autocorrelation analysis) can be used to identify periodic changes and trends. Record important change points and analyze their possible causes (such as screen switching, object movement, etc.). Generate visual charts (such as line charts or scatter plots) to show the trend of trajectory offset amplitude changes over time. This will help identify potential anomalies and patterns and provide intuitive analysis results. Determine the type of dynamic trajectory distortion model. You can choose a statistical model (such as a linear regression model) or a machine learning model (such as a support vector machine, a neural network). Define the input features of the model, including pixel offset amplitude, time series data, and comprehensive scene distortion features. Train the selected model using the previously extracted pixel offset trajectory trend data and scene distortion feature data. Depending on the model type, select an appropriate loss function and optimization algorithm. Evaluate model performance through methods such as cross-validation and adjust model parameters to improve prediction accuracy. Apply the trained model to new image sequence data to predict the trajectory offset and distortion features of the pixel points. Record the output of the model and compare it with the actual situation to evaluate the effectiveness of the model. Generate relevant visualization charts to show the difference between the trajectory predicted by the model and the actual trajectory to help understand the distortion phenomenon.

[0043] In this embodiment, step S4 includes the following steps: Step S41: performing regional pixel density calculation on multiple frames of images, thereby obtaining pixel density in different regions; Step S42: Dynamically perceive color changes of multiple frames of images, thereby generating picture change color change features; Step S43: performing region segmentation on the pixel point offset trajectory distortion model to generate a plurality of pixel point offset trajectory sub-models; Step S44: performing adaptive pixel point trajectory offset compensation calculation on multiple pixel point offset trajectory sub-models based on the pixel point density of different regions and the color change characteristics of the picture transformation, so as to generate an adaptive pixel point trajectory offset compensation value.

[0044] In this embodiment, multiple frames of images are read and preprocessed, including denoising and normalization. Commonly used image processing libraries such as OpenCV or PIL can efficiently load and process images. Denoising can use median filtering or Gaussian filtering to reduce noise interference in the image and ensure the accuracy of subsequent analysis. Next, the image is divided into multiple regions, which can be achieved through a fixed grid or an adaptive segmentation method. The fixed grid method divides the image into several small blocks evenly, while the adaptive segmentation can be dynamically adjusted based on the image content (such as edges or textures) to better capture regional features. For each region, the number of pixels therein, that is, the number of non-background pixels in the region, can be calculated by traversing the pixel matrix in the region and counting the number of non-zero pixels. Finally, the number of pixels in each area is calculated by the ratio of the area to the area to obtain the pixel density. After completing this calculation, the generated pixel density data will provide a basis for the subsequent dynamic color change perception and trajectory compensation calculation, helping to identify active areas and potential change areas. After obtaining the regional pixel density, the next step is to perceive the dynamic color changes of multiple frames of images. This process aims to capture and analyze the color change characteristics in the image to understand the dynamics of scene changes. It is necessary to perform color space conversion on multiple frames of images, usually from RGB color space to HSV or Lab color space. This conversion can better separate color information and make the perception of color changes more sensitive. Using the color space in OpenCV The color conversion function can easily achieve this operation. By calculating the color difference between each frame of the image and the previous frame, the dynamically changing areas can be identified. The pixel-level difference analysis method can be used, such as calculating the color difference of each pixel (such as CIEDE2000) to evaluate the degree of color change. By setting a threshold, the areas with significant changes can be marked to form a mask of the changed area. According to the color characteristics of the changed area, the statistical data of the color change can be extracted, such as the change amplitude, change rate and total area of ​​the changed area. These characteristics will help to describe the dynamic changes of the scene and provide the necessary information for the subsequent pixel trajectory offset compensation. After completing the perception of dynamic color changes, the next step is to compensate for the pixel offset trajectory distortion. The model is segmented into regions. The purpose of this step is to divide the overall distortion model into multiple sub-models for more detailed analysis and processing. It is necessary to analyze the previously constructed pixel trajectory distortion model to determine its characteristics and distribution. The distortion model can be displayed through visualization tools to better understand the structure and motion characteristics of the model. On this basis, a suitable segmentation method is selected, such as a clustering algorithm (such as K-means clustering or spectral clustering), to segment the distortion model into multiple regions. The clustering algorithm can cluster similar regions together based on the characteristics of the pixel points (such as position, speed, degree of distortion, etc.). Reasonable clustering parameters can be set to ensure that each sub-model contains pixels with similar characteristics, thereby improving the accuracy of segmentation. After completing the regional segmentation,Each sub-model will represent an independent motion trajectory area, which is convenient for subsequent analysis and compensation calculation. These sub-models will provide basic data for adaptive pixel trajectory offset compensation calculation, helping to better understand the distortion of different regions. After obtaining the sub-model of regional segmentation, the last step is to perform adaptive pixel trajectory offset compensation calculation based on the pixel density and color change characteristics of different regions. This process aims to accurately compensate for the motion of each region to reduce the impact of distortion. Combining the regional pixel density and dynamic color change characteristics, the compensation parameters of each region are determined. An adaptive adjustment algorithm can be set to dynamically adjust the compensation value according to the activity of the region (assessed by pixel density) and the amplitude of color change. The compensation value of the active area can be relatively large, while the static or less changing area uses a smaller compensation value. For each pixel offset trajectory sub-model, the calculated compensation value is applied to compensate for the trajectory offset. This process can be achieved by simple vector addition, that is, the offset of each pixel is added to the corresponding compensation value to obtain the corrected trajectory. ,

[0045] In this embodiment, the specific steps of step S5 are: Step S51: performing color quantization after pixel-by-pixel displacement on multiple pixel offset trajectory sub-models, thereby generating pixel-by-pixel color change values; Step S52: performing initial pixel color analysis on the pixels in each frame of the image to obtain initial pixel color features; Step S53: performing pixel color drift evolution on the color change values ​​of each pixel according to the initial pixel color feature to generate a color drift feature for each pixel; Step S54: fine-tuning the color of each pixel point on a pixel-by-pixel basis according to the adaptive pixel point trajectory offset compensation value, thereby generating a color fine-tuning parameter for each pixel.

[0046] In this embodiment, color quantization is performed on multiple pixel offset trajectory sub-models after pixel displacement one by one to generate a color change value for each pixel. Color quantization is to simplify the color space in the image into a finite number of colors for ease of analysis and processing. Based on the previously generated pixel offset trajectory sub-model, displacement calculation is performed pixel by pixel. This process involves adjusting the original position of each pixel according to its offset to obtain the pixel value of the new position. An interpolation method (such as bilinear interpolation or cubic interpolation) can be used to estimate the color value at the new pixel position to ensure that information loss is minimized during the displacement process. For color quantization, a common color quantization algorithm, such as K-means clustering or Octree quantization, can be used to calculate the value of each pixel. The color value is mapped to a limited color set. This process will create a color mapping table to help convert the color value of each pixel into a quantized value. The generated pixel-by-pixel color change values ​​will be recorded in an array, reflecting the difference between each pixel and the original color after displacement. These change values ​​will provide a data basis for subsequent color analysis and drift evolution, so that we can further understand the dynamic changes of color in the image. After completing the color quantization of each pixel, the next step is to perform initial color analysis on the pixels in each frame of the image to obtain the color features of the initial pixels. This analysis will provide a reference for the subsequent color drift evolution. It is necessary to extract the color features of each frame of the image pixel by pixel. The color histogram method can be used. The color value of each pixel is counted to generate the frequency distribution of each color. This method can help identify the main color tone and color distribution characteristics in the image and classify the color of each pixel. The K-means clustering algorithm can be used to divide the color space into several cluster centers to extract representative color features. These cluster centers will help understand the color characteristics of different regions and provide a basis for subsequent color drift evolution. During the analysis process, different color spaces (such as RGB, HSV or Lab) can be considered for comparison to select the space that best reflects the color characteristics of the image. Finally, the generated initial pixel color features will include the color value of each pixel, the frequency distribution and its representation in different color spaces. To lay a solid foundation for the subsequent color change analysis, after obtaining the initial pixel color features, the next step is to perform pixel color drift evolution on the color change values ​​of each pixel based on these features, quantify and describe the trend of color change over time, and use the initial pixel color features to compare the color change value of each pixel with its initial color. The drift feature can be obtained by calculating the color drift of each pixel, that is, the difference between the current color value and the initial color value. The specific operation can be performed using the following formula: Color drift = current color − initial color. In this process, pay attention to the choice of color space. Usually, calculations are performed in RGB space, but you can consider mapping the results to other more suitable color spaces (such as HSV) for analysis.In order to better capture the changes in brightness and saturation, the color drift characteristics of each pixel are generated by statistically analyzing the drift of all pixels. These characteristics can include the mean, standard deviation and change rate of the drift, etc., to help identify the trend and characteristics of color changes. The color drift characteristics finally generated will provide basic data for the subsequent calculation of color fine-tuning parameters, so that the color of each pixel can be adjusted more accurately, thereby achieving higher quality image processing. After completing the generation of color drift characteristics, the last step is to fine-tune the color of each pixel according to the adaptive pixel trajectory offset compensation value. The goal of this process is to ensure that the image maintains the naturalness and consistency of color during dynamic changes. The adaptive pixel trajectory offset compensation value calculated previously is combined with the color drift characteristics of each pixel. The degree of effect of the compensation value in color fine-tuning can be determined by setting a weight coefficient. Specifically, the weight can be dynamically adjusted according to the activity of the area and the amplitude of color change to ensure that more compensation is given to areas with large dynamic changes. The color of each pixel is fine-tuned. The color value of each pixel is adjusted according to the calculated drift characteristics and compensation values. The specific operation can be achieved through simple vector addition, that is, the drift characteristics of each pixel are added to the corresponding compensation value to obtain the corrected color value. Note that during the adjustment process, it is necessary to ensure that the corrected color value is still within the valid color range to avoid unnatural colors. ,

[0047] In this embodiment, the specific steps of step S6 are: Step S61: performing instant pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel to construct a pixel deviation calibrated display image; Step S62: performing regional display content analysis on the real-time playing animation image to generate display content data of different regions in the image; Step S63: performing human visual perception simulation on the display content data of different regions in the image to obtain human visual perception features; Step S64: dynamically optimizing the pixel deviation calibrated display image according to the visual perception characteristics of the human body to obtain a dynamically optimized display image.

[0048] In this embodiment, the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel are extracted from the previous steps. These data should include the displacement of each pixel and the corresponding color adjustment value, ensuring that the data format is consistent and can correspond one-to-one with the animated image played in real time. For each frame of the animated image, the compensation value and the color fine-tuning parameter are applied pixel by pixel. Specifically, for each pixel, its position and color value are updated: the new pixel position is calculated: the new position = the original position + the compensation value, the color value is updated: the new color value = the original color value + the color fine-tuning parameter, and a new image frame is generated according to the adjusted pixel value, ensuring that the change of each pixel is immediate to avoid visual delay, and the calibrated image frame is stored in the memory and is ready for real-time playback. As shown in the figure, you can use image processing libraries (such as OpenCV) to render and display images in real time, divide the real-time animated image into several regions, which can be achieved through fixed grid division or content-based segmentation methods (such as image segmentation algorithms), ensure that the size and position of each region can represent the content characteristics of the overall image, and perform content analysis on each region, including calculating color histograms, texture features, shape features, etc. This can use computer vision technology (such as LBP, Gabor filters, etc.) to extract regional features, record the display content data of each region, including color distribution, texture complexity and shape information, integrate the display content data of different regions, and generate a comprehensive data structure containing the content of each region. Feature information, which can perform statistical analysis on the content features of the region to identify the main content and structural features in the image, output the display content data of each region, including color, texture and shape features, generate a visualization report, and display the content analysis results of different regions. Select a suitable visual perception model, such as a physiological vision model (such as the HVS model) or a psychological vision model that can simulate how humans perceive color, brightness and contrast, ensure that the selected model can process the regional display content data and capture the key features of human visual perception. For the display content data of each region, input it into the selected visual perception model and calculate the corresponding visual perception features, which may include the evaluation of indicators such as contrast, brightness, and color saturation. , record the visual perception characteristics of each area, analyze its impact on human visual perception, integrate the visual perception characteristics of different areas into a data structure to facilitate subsequent analysis and optimization, statistically analyze the perceptual characteristics of each area, identify areas that are more attractive to the audience to help optimize the image display, output the visual perception feature data of different areas in the image, including perceptual intensity, contrast, and brightness, etc., generate a visualization report to display the simulated visual perception results, and design a dynamic optimization model based on the human visual perception characteristics obtained in the previous steps. The model can dynamically adjust the display parameters of the image according to the visual perception intensity and other characteristics of the area. You can consider using a weighted algorithm to adjust the color, brightness, and contrast according to the weight of the perceptual characteristics.For the display image of each area, dynamically calculate the optimization parameters according to its visual perception characteristics, which may include adjusting brightness, enhancing contrast or adjusting color saturation, apply the optimization parameters to update the image, ensure that the visual effect of each area can be improved, evaluate the image effect after optimization, compare the visual perception characteristics before and after optimization, ensure that the optimization process improves the audience's visual experience, and if necessary, adjust the parameters of the optimization model to further improve the image quality, output the display image after dynamic visual perception optimization, ensure that the visual effect of the image is improved, and generate a visualization report to show the visual comparison effect before and after optimization.

[0049] In this embodiment, a high-precision liquid crystal display pixel calibration system is provided, which is used to execute the high-precision liquid crystal display pixel calibration method as described above, including: An image distortion module is used to obtain a high-precision LCD screen real-time animation image based on an internal sensor of the screen; perform image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; An offset track module is used to perform frame-by-frame pixel optical tracking on the real-time animation image and perform frame-by-frame rolling offset vector calculation, thereby constructing a pixel point time-series offset track diagram; The trajectory model module is used to estimate the regional dynamic optical flow of the pixel point temporal offset trajectory map, and to perform dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to build a pixel point offset trajectory distortion model; The trajectory offset compensation module is used to perceive the dynamic color change of the pixel point offset trajectory distortion model, and then perform adaptive pixel point trajectory offset compensation calculation to generate an adaptive pixel point trajectory offset compensation value; A color fine-tuning module is used to perform pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; The real-time calibration module is used to perform real-time pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, and perform dynamic visual perception optimization to obtain a dynamic visual optimized display image.

[0050] The present invention can analyze the distortion phenomenon (such as color distortion, geometric distortion, etc.) in the image through the image distortion module, and mine the comprehensive scene distortion characteristics to provide basic data for subsequent image correction. Through in-depth analysis of the scene distortion characteristics, the module can accurately identify the distortion type existing in the image, and provide a directional optimization solution for subsequent offset trajectory, color fine-tuning and other processing, reducing unnecessary global adjustments. By tracking each pixel point in the image frame by frame, the offset of each pixel in the time dimension can be identified. The time-series offset trajectory diagram generated by the trajectory module provides detailed information on the movement of the pixel points, which helps the subsequent trajectory modeling and compensation. The offset trajectory module can clearly display the changes of each pixel in the dynamic image, which is convenient for the subsequent dynamic optical flow estimation and distortion modeling, thereby improving the accuracy of image correction. By estimating the optical flow (i.e., the motion pattern of the pixel) in the image area, the dynamic characteristics of the image area are identified, and the movement and changes of each area in the image can be accurately analyzed and simulated. Based on the distortion characteristics of the image, the trajectory model module can construct a dynamic trajectory distortion model to provide for subsequent offset compensation. Precise mathematical basis ensures accuracy in the distortion repair process. By sensing color changes and dynamic features in dynamic images, the trajectory offset compensation module can dynamically adjust colors according to the motion trajectory of each pixel to ensure a smooth transition of image color changes. Through the adaptive compensation method, the module can generate appropriate trajectory offset compensation values ​​according to the actual conditions of different areas and different pixels to avoid unnatural color changes caused by global correction. By performing independent color fine-tuning on each pixel, the color drift phenomenon in dynamic images can be effectively corrected to ensure that the color of each pixel matches its actual position. The module can identify and repair color deviations caused by offset trajectories, avoid color inconsistencies or drift in the image, and improve the overall display quality. By instantly calibrating pixel deviations in the image and optimizing it in combination with human visual perception characteristics, it ensures that each picture can maintain high-quality visual effects during playback. By sensing the sensitivity of human vision, the brightness, contrast and color of the image are optimized to make the image have the best effect under different viewing conditions and enhance the viewing experience.

[0051] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0052] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A high-precision pixel calibration method for a liquid crystal display, characterized in that: The following steps are involved: Step S1: obtaining a high-precision liquid crystal display screen real-time animation image based on the internal sensor of the display screen; performing image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; Step S2: performing optical tracking of the real-time animation image frame by frame, and performing frame by frame rolling offset vector calculation, so as to construct a pixel point time sequence offset trajectory diagram; Step S3: performing regional dynamic optical flow estimation on the pixel point temporal offset trajectory map, and performing dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to construct a pixel point offset trajectory distortion model; Step S4: performing dynamic color change perception on the pixel point offset trajectory distortion model, and then performing adaptive pixel point trajectory offset compensation calculation, thereby generating an adaptive pixel point trajectory offset compensation value; Step S5: performing pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; Step S6: Based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, the real-time animation image is subjected to instant pixel deviation calibration, and dynamic visual perception optimization is performed to obtain a dynamic visual optimized display image.

2. The pixel calibration method for a high-precision liquid crystal display according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: obtaining a high-precision liquid crystal display screen to play an animated image in real time based on the internal sensor of the display screen; Step S12: performing time-sequential frame decomposition processing on the real-time animation image, thereby extracting multiple frame images; Step S13: Calculating the refresh frequency of the display screen for multiple frames of images, thereby obtaining a real-time frame rate of the images; Step S14: analyzing the brightness parameter changes of multiple frames of images to obtain image brightness change characteristics; Step S15: mining the image scene distortion features based on the image brightness change features and the image real-time frame rate, thereby obtaining the comprehensive scene distortion features.

3. The pixel calibration method for a high-precision liquid crystal display according to claim 2, characterized in that: The specific steps of step S15 are: Perform spatial optical distortion detection based on image brightness change characteristics and image real-time frame rate, thereby generating spatial optical distortion characteristics; Perform color shift statistics according to the spatial optical distortion characteristics to obtain the color shift characteristics of the dynamic image; Performing transient change visual recognition on multiple frames of images to generate transient change visual features of the images; Perform local dynamic image fuzzy analysis based on the transient visual features of the image and extract local dynamic fuzzy features; Based on the local dynamic blur feature, local blur distortion is located for multiple frames of images to generate a local dynamic blur distortion area; Quantifying the distortion distribution of the local dynamic blur distortion area, thereby generating dynamic blur distortion distribution data; The image scene distortion features are mined based on the dynamic image color shift characteristics and dynamic blur distortion distribution data to obtain the comprehensive scene distortion features.

4. The pixel calibration method for a high-precision liquid crystal display according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing time-series frame fitting on multiple image frames to generate a time-series frame image sequence; Step S22: performing optical tracking of the time-series frame image sequence frame by frame, and extracting the pixel points in each frame image; Step S23: performing pixel offset identification of adjacent frames for pixels in each frame of the image, thereby obtaining inter-frame pixel offset values; Step S24: performing frame-by-frame rolling offset vector calculation based on the inter-frame pixel offset value to generate a time-series frame pixel offset vector; Step S25: Perform temporal visualization of the offset trajectory of the time frame pixel offset vector, thereby constructing a pixel point temporal offset trajectory diagram.

5. The pixel calibration method for a high-precision liquid crystal display according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing regional dynamic optical flow estimation on the pixel point time series offset trajectory map to generate regional distribution features of the pixel point offset trajectory; Step S32: calculating the screen switching trajectory offset amplitude based on the pixel point offset trajectory area distribution characteristics of the multiple frame images, thereby generating the pixel point offset trajectory amplitude; Step S33: performing trajectory deviation distribution trend evolution based on the pixel point deviation trajectory amplitude to generate pixel point trajectory deviation trend data; Step S34: Dynamic trajectory distortion modeling is performed on the pixel point trajectory deviation trend data based on the comprehensive scene distortion characteristics to construct a pixel point deviation trajectory distortion model.

6. The high-precision pixel calibration method for a liquid crystal display according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: performing regional pixel density calculation on multiple frames of images, thereby obtaining pixel density in different regions; Step S42: Dynamic color change perception is performed on multiple frames of images, thereby generating picture change color change features; Step S43: performing region segmentation on the pixel point offset trajectory distortion model to generate a plurality of pixel point offset trajectory sub-models; Step S44: performing adaptive pixel point trajectory offset compensation calculation on multiple pixel point offset trajectory sub-models based on the pixel point density of different regions and the color change characteristics of the picture transformation, so as to generate an adaptive pixel point trajectory offset compensation value.

7. The pixel calibration method for a high-precision liquid crystal display according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing color quantization after pixel-by-pixel displacement on multiple pixel offset trajectory sub-models, thereby generating pixel-by-pixel color change values; Step S52: performing initial pixel color analysis on the pixels in each frame of the image to obtain initial pixel color features; Step S53: performing pixel color drift evolution on the color change values ​​of each pixel according to the initial pixel color feature to generate a color drift feature for each pixel; Step S54: fine-tuning the color of each pixel point on a pixel-by-pixel basis according to the adaptive pixel point trajectory offset compensation value, thereby generating a color fine-tuning parameter for each pixel.

8. The pixel calibration method for a high-precision liquid crystal display according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing instant pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel to construct a pixel deviation calibrated display image; Step S62: performing regional display content analysis on the real-time playing animation image to generate display content data of different regions in the image; Step S63: performing human visual perception simulation on the display content data of different regions in the image to obtain human visual perception features; Step S64: dynamically optimizing the pixel deviation calibrated display image according to the visual perception characteristics of the human body to obtain a dynamically optimized display image.

9. A high-precision pixel calibration system for a liquid crystal display, characterized in that: The method for performing pixel calibration of a high-precision liquid crystal display screen as claimed in claim 1 comprises: An image distortion module is used to obtain a high-precision LCD screen real-time animation image based on an internal sensor of the screen; perform image scene distortion feature mining on the real-time animation image to obtain a comprehensive scene distortion feature; An offset track module is used to perform frame-by-frame pixel optical tracking on the real-time animation image and perform frame-by-frame rolling offset vector calculation, thereby constructing a pixel point time-series offset track diagram; The trajectory model module is used to estimate the regional dynamic optical flow of the pixel point temporal offset trajectory map, and to perform dynamic trajectory distortion modeling based on the comprehensive scene distortion characteristics to build a pixel point offset trajectory distortion model; The trajectory offset compensation module is used to perceive the dynamic color change of the pixel point offset trajectory distortion model, and then perform adaptive pixel point trajectory offset compensation calculation to generate an adaptive pixel point trajectory offset compensation value; A color fine-tuning module is used to perform pixel color drift evolution and pixel-by-pixel color fine-tuning on multiple pixel offset trajectory sub-models, thereby generating color fine-tuning parameters for each pixel; The real-time calibration module is used to perform real-time pixel deviation calibration on the real-time animation image based on the adaptive pixel point trajectory offset compensation value and the color fine-tuning parameter of each pixel, and perform dynamic visual perception optimization to obtain a dynamic visual optimized display image.

Images