Backlight source dynamic refreshing method and system based on content change frequency

Abstract

The invention provides a backlight dynamic refreshing method and system based on content change frequency, and is applied to the field of image data processing. According to the method, the inter-frame variable quantity of the current image content of the display screen is pre-collected, whether the image refreshing speed is matched with the backlight response or not is judged, the pixel change rate in unit time is calculated based on the frame difference method, the content change frequency index is reasonably divided, the backlight PWM parameters are dynamically adjusted in combination with user setting, and the display quality is improved. The response coordination of the display screen under the dynamic image is effectively improved, when the frame rate fluctuation is recognized, the gray scale compensation frame is introduced and the brightness change slope is limited, and the problems of smear, splash screen and the like caused by asynchronous refreshing can be relieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image data processing, in particular to a backlight dynamic refreshing method and system based on content change frequency. BACKGROUND

[0002] The backlight driving of many current liquid crystal display screens is constant frequency or simple brightness response adjustment mode, which fails to combine the frequency characteristics of display content change for dynamic refreshing, especially in high frame rate display, high dynamic video, game or interactive interface, the screen content frequently changes, and if the backlight refreshing frequency is not adjusted synchronously, it may cause image trailing, flickering, power waste and other problems. SUMMARY

[0003] The present application aims to solve the problem of failing to follow the content refreshing frequency adjustment under dynamic image, which is prone to "trailing" or "flickering", and provides a backlight dynamic refreshing method and system based on content change frequency.

[0004] The present application adopts the following technical means to solve the technical problems: The present application provides a backlight dynamic refreshing method based on content change frequency, comprising: counting the total number of pixels Based on the inter-frame change amount of the current display content pre-acquired by the display screen, the image content refreshing speed of the current display content is obtained; It is judged whether the image content refreshing speed matches the preset backlight refreshing response speed; If not, the image difference data of the current display content and the previous frame display content is identified, the pixel change rate in unit time is calculated by applying the preset frame difference method, and the content change frequency index of the display screen is dynamically adjusted according to the pixel change rate, wherein the content change frequency index specifically includes low dynamic, medium dynamic and high frequency dynamic; It is judged whether the content change frequency index meets the display setting of the display screen by the user; If not, the PWM frequency and duty cycle of the display screen backlight are adjusted according to the mapping strategy preset by the user for the display screen, the frame rate fluctuation of the display screen is detected, the preset gray scale intermediate compensation frame is added when the display screen backlight changes, the upper limit of the brightness change slope of the display screen is dynamically limited, and the brightness change gradient of the display screen is constructed, wherein the mapping strategy specifically includes brightness strategy, image quality strategy and perception priority strategy.

[0005] Further, before the step of identifying the image difference data of the current display content and the previous frame display content, the present application further comprises: extract image data from the data buffer, and identify an image frame time interval of the display screen according to the image data, wherein the image data specifically includes color values, gray scale brightness values and transparency of pixel points; determine whether the image frame time interval is synchronized with a screen refresh cycle; if yes, analyze display content of the image data, acquire display parameters of the display screen according to the display content, and construct a frame rate change trend of the image data through the display parameters, wherein the display content specifically includes static images, video streams, fast scrolling texts and game pictures, and the display parameters specifically include frame rates, motion vectors and change area distributions.

[0006] Further, the step of dynamically adjusting the content change frequency index of the display screen according to the pixel change rate further includes: extract a change pixel quantity proportion of the grid area based on the grid area of the display screen; determine whether the change pixel quantity proportion matches the pixel change rate; if yes, construct a corresponding local pixel change weight according to the change pixel quantity proportion, identify content complexity of the grid area, and dynamically adjust a grid scale of the grid area, wherein the content complexity specifically includes dynamic content and static content, and the grid scale specifically includes coarse grid and fine grid.

[0007] Further, the step of dynamically limiting the upper limit of the brightness change slope of the display screen and constructing the brightness change gradient of the display screen further includes: extract brightness distribution statistical values of the continuous image frames based on the continuous image frames of the display screen, and generate a brightness jump point within a preset time window, wherein the brightness distribution statistical values specifically include average brightness, maximum brightness and minimum brightness; determine whether the brightness jump point exceeds a preset mutation threshold; if yes, identify a brightness mutation area corresponding to the continuous image frames according to the brightness jump point, apply a preset exponential decay filter to perform slow transition of brightness on the brightness mutation area, detect adjacent areas of the brightness mutation area in real time, and dynamically limit a brightness change amplitude between the adjacent areas.

[0008] Further, the step of determining whether the image content refresh speed matches the preset backlight refresh response speed further includes: extract pixel differences between the continuous image frames based on the continuous image frames of the display screen, and construct a dynamic hot area map of the display screen according to the pixel differences; determining whether the dynamic heat map can be mapped to physical display screen coordinates of the display screen; If yes, identifying high-frequency irregular jitter of the dynamic heat map, marking the high-frequency irregular jitter as an abnormal frame of the display screen, demarcating a corresponding to-be-reconstructed region according to the abnormal frame, and constructing a candidate frame block of the to-be-reconstructed region by using a preset pixel texture similarity.

[0009] Further, in the step of determining whether the content change frequency indicator conforms to a display setting of the display screen by a user, the step further includes: identifying a scene type of the display screen based on a preset display mode of the display screen by the user, wherein the display mode specifically includes a power saving mode, a standard mode and a high performance mode, and the scene type specifically includes a reading scene, a video scene and a game scene; determining whether the display mode matches a refresh requirement of the scene type; If no, constructing a corresponding pixel motion region according to pixel differences between continuous image frames, generating a displacement path of each pixel point by using the pixel motion region, collecting a motion non-continuous region of the pixel motion region according to the displacement path, detecting a corresponding missing pixel from the motion non-continuous region, performing vector offset distance scaling on the displacement path in proportion to obtain a target position of each pixel required for frame insertion, and performing pixel combination on the missing pixel by taking the target position of each pixel as frame insertion content between original frames.

[0010] Further, in the step of obtaining an image content refresh speed of the current display content based on an interframe change amount of the current display content pre-acquired by the display screen, the step further includes: intercepting a frame data structure of the current display content based on a preset frequency of the display screen, wherein the frame data structure specifically includes a pixel matrix, brightness information and a color channel; determining whether the frame data structure matches a preset image heat area of the display screen; If yes, obtaining a pixel motion speed of the current display content, and dynamically adjusting a refresh period of the image heat area according to the pixel motion speed.

[0011] The application further provides a backlight dynamic refresh system based on a content change frequency, including: an obtaining module configured to obtain an image content refresh speed of current display content based on an interframe change amount of the current display content pre-acquired by a display screen; a determining module configured to determine whether the image content refresh speed matches a preset backlight refresh response speed; The execution module is configured to, if no, identify image difference data of the current display content and previous frame display content, apply a preset frame difference method to calculate a pixel change rate within a unit time, and dynamically adjust a content change frequency index of the display screen according to the pixel change rate, wherein the content change frequency index specifically includes low dynamic, medium dynamic and high frequency dynamic; The second judgment module is configured to judge whether the content change frequency index meets a display setting of the display screen by a user. The second execution module is configured to, if not, adjust a PWM frequency and a duty cycle of a backlight source of the display screen according to a preset mapping strategy of the display screen by the user, detect a frame rate fluctuation of the display screen, add a preset intermediate compensation frame of a gray scale when the backlight source of the display screen changes, dynamically limit an upper limit of a brightness change slope of the display screen, and construct a brightness change gradient of the display screen, wherein the mapping strategy specifically includes a brightness strategy, a picture quality strategy and a perception priority strategy.

[0012] Further, the method further comprises: The identification module is configured to extract image data from a data buffer area preset by the display screen, and identify an image frame time interval preset by the display screen according to the image data, wherein the image data specifically includes color values, gray scale brightness values and transparency of pixel points. The third judgment module is configured to judge whether the image frame time interval is synchronized with a screen refresh cycle. The third execution module is configured to, if yes, analyze display content of the image data, obtain display parameters of the display screen according to the display content, and construct a frame rate change trend of the image data through the display parameters, wherein the display content specifically includes a static image, a video stream, a fast scrolling text and a game picture, and the display parameters specifically include a frame rate, a motion vector and a change area distribution.

[0013] Further, the execution module further comprises: The extraction unit is configured to extract a change pixel quantity proportion of the grid area based on the grid area preset by the display screen. The judgment unit is configured to judge whether the change pixel quantity proportion matches the pixel change rate. The execution unit is configured to, if yes, construct a corresponding local pixel change weight according to the change pixel quantity proportion, identify a content complexity of the grid area, and dynamically adjust a grid scale of the grid area, wherein the content complexity specifically includes dynamic content and static content, and the grid scale specifically includes a coarse grid and a fine grid.

[0014] The application provides a backlight dynamic refreshing method and system based on content change frequency, which has the following beneficial effects: The application pre-collects the interframe change amount of the current image content of the display screen, judges whether the image refreshing speed matches the backlight response, and then calculates the pixel change rate in a unit time based on the frame difference method, reasonably divides the content change frequency index, and dynamically adjusts the backlight PWM parameter in combination with the user setting, effectively improves the response coordination of the display screen under dynamic images, introduces the gray scale compensation frame and limits the brightness change slope when the frame rate fluctuation is identified, and can relieve the ghosting, screen flashing and other problems caused by different refreshing speeds, and the overall mechanism can realize dynamic adjustment of the brightness change gradient, thereby improving the visual stability and viewing comfort under dynamic images. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 It is a flowchart of an embodiment of the backlight dynamic refreshing method based on content change frequency of the application. Figure 2 It is a structural block diagram of an embodiment of the backlight dynamic refreshing system based on content change frequency of the application. DETAILED DESCRIPTION

[0016] It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application, and the purpose, function characteristics and advantages of the application will be further described with reference to the embodiments and the accompanying drawings.

[0017] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings of the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0018] Reference is made to the accompanying drawings Figure 1 The backlight dynamic refreshing method based on content change frequency in an embodiment of the application comprises: S1: based on the interframe change amount of the current display content pre-collected by the display screen, acquiring the image content refreshing speed of the current display content; S2: judging whether the image content refreshing speed matches the preset backlight refreshing response speed; S3: if not, identifying the image difference data of the current display content and the previous frame display content, applying the preset frame difference method to calculate the pixel change rate in a unit time, and dynamically adjusting the content change frequency index of the display screen according to the pixel change rate, wherein the content change frequency index specifically comprises low dynamic, medium dynamic and high frequency dynamic; S4: judging whether the content change frequency index meets the display setting of the display screen by the user; S5: if not, adjusting the PWM frequency and duty cycle of the display screen backlight source according to the mapping strategy preset by the user for the display screen, detecting the frame rate fluctuation of the display screen, adding the preset intermediate compensation frame in the gray scale when the display screen backlight source changes, dynamically limiting the upper limit of the brightness change slope of the display screen, and constructing the brightness change gradient of the display screen through the frame rate fluctuation, wherein the mapping strategy specifically includes a brightness strategy, a picture quality strategy and a perception priority strategy.

[0019] In the embodiment, the system acquires the image content refresh speed of the current display content based on the interframe variation quantity of the current display content pre-acquired by the display screen, and then judges whether the image content refresh speed matches the backlight refresh response speed set in advance to execute corresponding steps; for example, when the system determines that the image content refresh speed of the current display content can match the backlight refresh response speed set in advance, it is considered that the interframe image variation frequency of the current display content is within the range that the backlight source can respond in time, and there is no overshoot or delay phenomenon, so the system keeps the current backlight PWM modulation strategy and refresh parameter, does not perform additional frame insertion or gamma adjustment, and at the same time, due to good synchronism, the "energy saving compensation mode" can be started, the local backlight weakening is started for the image static area, the frame rate limiting is started for the area with low refresh intensity, the backlight power output is controlled, the energy consumption and heat are reduced, and the inertia tracking strategy is executed for the refresh direction and speed of the local hot area to avoid frequent brightness jumps caused by slight changes; for example, when the system determines that the image content refresh speed of the current display content cannot match the backlight refresh response speed set in advance, it is considered that the interframe image variation frequency of the current display content is not within the response range of the backlight source, the system identifies the image difference data of the current display content and the previous frame display content, applies the pre-set frame difference method to calculate the pixel variation rate in unit time, dynamically adjusts the content variation frequency index according to different pixel variation rates, and the content variation frequency index specifically includes low dynamic, medium dynamic and high frequency dynamic; the system can realize intelligent identification of the dynamic level of the display content by judging whether the current image content refresh speed matches the backlight response speed and adjusting the content variation frequency index (such as low dynamic, medium dynamic and high frequency dynamic) according to the interframe image difference, and this classification can effectively optimize the backlight adjustment strategy, so that the change of the display content and the response of the backlight are more coordinated, thereby improving the overall picture smoothness, clarity and transition naturalness, avoiding visual fatigue problems such as trailing and flickering caused by unmatched refresh, and at the same time, when the system identifies that the image variation is low dynamic or static, the backlight refresh frequency can be reduced to reduce unnecessary power consumption; under high frequency dynamic variation, the backlight response speed is improved to ensure that the brightness and the image content are updated synchronously to avoid ghosting or brightness lag, the scheme can flexibly adjust the backlight strategy according to the content dynamic level to balance energy saving and performance, and is particularly suitable for scenes with high energy efficiency requirements such as mobile devices and smart televisions, and the pixel variation rate in unit time is calculated by the frame difference method to establish the content variation frequency index, so that the system has stronger image perception and analysis capability, which can not only judge the variation trend at the content level, but also feed back the judgment results to the backlight source in real time to realize the content-driven display optimization mechanism, which improves the intelligence of the system and provides a structural basis for further combining AI image recognition, scene analysis and other functions.Then the system determines whether the content change frequency index meets the user's display settings for the display screen to perform corresponding steps; for example, when the system determines that the content change frequency index of the display screen meets the user's display settings for the display screen, the system considers that the current display state meets the user's expectations for display clarity, smoothness, response speed, or energy-saving mode, etc. The system detects the display mode of the display screen. If the user sets it to high refresh, dynamic clarity mode, the system maintains a higher content update rate and backlight synchronization. If the user sets it to energy-saving or eye-care mode, the system ensures a lower change frequency to avoid power consumption and visual fatigue caused by frequent changes, while keeping the current refresh and backlight parameters stable, entering a balance state of power consumption and experience, pausing unnecessary refresh adjustment calculations, reducing resource consumption, and recording the current configuration as a recommended reference parameter for similar content. For example, when the system determines that the content change frequency index of the display screen does not meet the user's display settings for the display screen, the system considers that the current display state cannot meet the user's expectations for display clarity, smoothness, response speed, or energy-saving mode, etc. The system adjusts the PWM frequency and duty cycle of the display screen backlight according to the user's pre-set mapping strategy for the display screen, which includes brightness strategy, image quality strategy, and perception priority strategy. The system detects the frame rate fluctuation of the display screen and inserts pre-set intermediate compensation frames when the display screen backlight changes to dynamically limit the upper limit of the brightness change slope and build the brightness change gradient of the display screen. By detecting the content change frequency index and combining the pre-set brightness strategy and image quality strategy, the system dynamically adjusts the PWM frequency and duty cycle of the backlight, which helps to reduce the brightness instability problem caused by frame rate fluctuation. When the system identifies frame rate fluctuation, it inserts intermediate compensation frames and limits the upper limit of the brightness change slope to effectively smooth the brightness transition process and avoid flicker or tearing caused by sudden brightness changes, thereby enhancing display stability and improving overall image quality. Through the perception priority strategy, the system can dynamically match the display effect according to the user's preference for clarity, response speed, or energy-saving in different scenarios. For example, in reading mode, it tends to reduce frame rate and brightness change rate to reduce visual fatigue, while in video playback or game mode, it increases refresh response to ensure smoothness. This strategy ensures that the display effect always matches the user's current usage intention, optimizing visual perception and operation feedback. When the system determines that the current display state does not meet the user's energy-saving requirements, it can automatically enter an energy efficiency optimization path, such as reducing backlight PWM frequency, reducing unnecessary dynamic brightness adjustment, or suppressing high-frequency frame rate fluctuations, thereby reducing power consumption. In addition, by building a brightness change gradient and limiting the upper limit of the brightness change slope, it avoids power waste caused by frequent brightness jumps, helps to balance the overall system energy consumption, and prolongs the battery life of mobile devices such as smartphones and tablets.

[0020] It should be noted that the image difference data between the current displayed content and the previous frame displayed content is recognized, and the pixel change rate per unit time is calculated by applying the preset frame difference method. According to the pixel change rate, the content change frequency index of the display screen is dynamically adjusted. The specific example is as follows: Assume that the system obtains the current frame displayed content and the content of the previous frame , and the two frames of images are compared pixel by pixel. Subsequently, the difference value is calculated using the gray difference (or RGB difference) of pixel-by-pixel difference:

[0021] By setting a change threshold T, if the pixel position D(x,y)>T, it is considered that the pixel has changed; After calculating the differences for all pixel points, count the number of changed pixels , and count the total number of pixels , then the pixel change rate (change frequency index) is defined as:

[0022] And the unit time can be set as the frame time Δt (for example, 1 / 60 seconds); Subsequently, dynamically adjust the refresh frequency, and adjust the refresh strategy according to the size of the pixel change rate R; for example, set the refresh frequency range [30Hz, 120Hz]: If R>0.8, that is, the display screen is a high-speed dynamic picture, set the refresh rate to 120Hz; If 0.3<R≤0.8, set it to 60Hz; If R≤0.3, it is a low-dynamic or static picture, and the refresh rate is reduced to 30Hz to save energy; Since the pre-set image size is 1920×1080 (Full HD), a total of about 2,073,600 pixels, the difference between the current frame and the previous frame image is as follows (using grayscale image for simplified processing). Assume that the gray difference>15 is considered "the pixel has changed", and 1,656,320 pixels are detected to meet the conditions; That is, it is necessary to calculate the pixel change rate:

[0023] Judging according to the refresh strategy: R≈0.80≈80%, which belongs to a high change rate, and the system automatically adjusts the refresh frequency to 120Hz to avoid ghosting or stuttering; In summary, the system can also perform automatic brightness adjustment based on different rates of change. If the rate of change is high, it means that it is a dynamic picture, and the system can appropriately increase the brightness and contrast to enhance the visual experience, perform local refresh optimization, extract the pixel positions that have changed (such as the rectangular hot area), and only refresh these areas to reduce the power consumption of the entire screen and maintain energy-saving control logic. If the rate of change of consecutive frames is lower than a set threshold (such as R < 0.1 for 10 consecutive frames), the system enters a "low power consumption mode" and the screen refresh rate is automatically reduced to the minimum (such as 24 Hz).

[0024] In this embodiment, the PWM frequency and duty cycle of the display screen backlight are adjusted according to the mapping strategy preset by the user, the frame rate fluctuation of the display screen is detected, a preset gray scale intermediate compensation frame is added when the display screen backlight changes, the upper limit of the brightness change slope of the display screen is dynamically limited, and the brightness change gradient of the display screen is constructed. For example, the specific implementation is as follows: Suppose a user of a mobile phone sets a "eye protection mode" brightness strategy to limit the screen brightness change amplitude during the night period (22:00-06:00); when the screen brightness decreases, it should be slowly transitioned to avoid sudden changes that stimulate the eyes; at this time, the system will use the "gray scale transition frame + limit brightness change slope" method to meet the above requirements; The mapping function between the user brightness demand (L_user) and the backlight PWM duty cycle (D) is as follows:

[0025] Wherein, Lmin=10nits, Lmax=400nits, and the PWM frequency is fixed at 20kHz, so the maximum brightness change slope limit is not more than 50nits / s; At this time, the user reduces the screen from 300 nits to 100 nits, and needs to calculate the target change slope. The target brightness change amount is:

[0026] The maximum allowed brightness change slope is:

[0027] Therefore, the brightness must be reduced within at least 4 seconds, and cannot be suddenly reduced; Then a sequence of gray scale intermediate compensation frames is constructed. In order to smoothly transition, the system will construct a set of intermediate gray scale compensation frames, with a frame rate of 60 frames per second, and the total number of frames in 4 seconds is: 60x4=240 frames. That is, the brightness change amount of each frame is:

[0028] The PWM duty cycle change per frame (according to the mapping function) is: Current brightness 300 nits→

[0029] Target brightness 100 nits→

[0030] Duty cycle change amount:

[0031] Frame rate fluctuation detection and dynamic frame insertion are then performed. If frame rate fluctuation (such as a transient decrease to 30 fps) is detected during the transition period, the system will extend the compensation time and insert more gray scale frames to maintain the unit time brightness slope unchanged. The frame rate decreases to 30 fps→ 120 frames can only be rendered in 4 seconds; The actual brightness change time is extended to:

[0032] At this time, in order to ensure that the original slope is unchanged, the system will recalculate the change gradient of each frame (smaller) and insert gray scale compensation frames to fill the slow transition process. In summary, in the above example, the system can avoid brightness mutation stimulating the user's eyes through the above-mentioned manner, and also maintain the brightness change continuous and smooth in the case of abnormal frame rate fluctuation, so that the user's "eye protection strategy" can be intelligently executed.

[0033] In the embodiment, before the step S3 of identifying the image difference data of the current display content and the previous frame display content, the method further comprises: S301: Based on the data buffer area preset by the display screen, image data is extracted from the data buffer area, and based on the image data, an image frame time interval preset by the display screen is identified, wherein the image data specifically includes color values, gray scale brightness values and transparency of pixel points; S302: Determine whether the image frame time interval is synchronized with the screen refresh period; S303: If yes, analyze the display content of the image data, obtain the display parameters of the display screen according to the display content, and construct the frame rate change trend of the image data through the display parameters, wherein the display content specifically includes static images, video streams, fast scrolling texts and game pictures, and the display parameters specifically include frame rate, motion vector and change area distribution.

[0034] In this embodiment, the system extracts image data from a pre-set data buffer on the display screen. The image data specifically includes pixel color values, grayscale brightness values, and transparency. Based on different image data, the system identifies the pre-set image frame intervals on the display screen. Then, the system determines whether these image frame intervals are synchronized with the screen refresh cycle to execute corresponding steps. For example, if the system determines that the pre-set image frame intervals on the display screen cannot be synchronized with the screen refresh cycle, the system considers the image frames to arrive slowly, indicating insufficient image generation or transmission speed, which may lead to screen tearing, stuttering, or delay. The system will then "wait for synchronization" of the display frames, introducing a delay to ensure that each frame is aligned with the display refresh signal. To avoid screen tearing, multiple frames of image data are retained in the data buffer, allowing the next frame to be prepared in advance, improving smoothness and processing leeway. Furthermore, if the frame interval is not fixed, consistency can be improved by limiting rendering complexity, reducing image quality, and optimizing rendering. For example, when the system determines that the preset image frame interval of the display screen is synchronized with the screen refresh cycle, the system considers the image to be generated normally on the display screen. The system then analyzes the content of the image data, which may include static images, video streams, fast-scrolling text, and game footage. Based on the different display content, the system obtains the display parameters of the display screen, including frame rate, motion vectors, and distribution of changing areas. These display parameters are then used to... The system constructs a frame rate change trend for image data. By synchronizing the image frame interval with the display refresh cycle, it ensures continuous, tear-free, and stutter-free image playback on the screen, significantly improving the smoothness and consistency of the user's viewing experience. For example, when playing videos or scrolling subtitles, if the image frame interval is not synchronized with the refresh cycle, problems such as frame skipping or flickering may occur, affecting the viewing experience. The synchronization mechanism effectively eliminates such problems, ensuring stable visual output. Simultaneously, the system analyzes the display content type of the image data (such as static images, video streams, fast-scrolling text, or game footage) and dynamically adjusts display parameters such as frame rate, motion vectors, and changing areas based on type differences, making the image presentation more suitable for the actual scene requirements. For example, for fast-scrolling text, the focus can be on optimizing text clarity and scrolling smoothness, while game graphics need to improve response frame rate and motion smoothness. This dynamic adaptation mechanism improves the display's performance in various application scenarios and constructs the frame rate change trend of image data, which can help the system identify the dynamic characteristics of image content, such as whether there is violent movement, static screen or sudden change, so as to predict the possible display strategies (such as high refresh rate mode or low power mode) in advance. For example, when displaying static images for a long time, the refresh rate can be reduced to save energy, while when high-frequency actions are identified, the refresh rate is automatically increased to ensure smoothness. This trend modeling mechanism provides data support for intelligent display management and achieves dual optimization of display performance and power consumption.

[0035] Note that the system will analyze the display content of the image data, for example: I. Recognition and examples of static images, principle explanation: When the continuous image frames have little pixel change in a certain time (i.e. the frame difference tends to zero), the system considers the current display content as a static image, for example: the system continuously receives 10 frames of images, more than 95% of the pixel values have no significant change between each frame, i.e. it is determined as static; For example: when the user opens a landscape photo in the browser or views a poster, the display image does not move for a long time, and the system can judge it as a static image; II. Recognition and examples of video streams, principle explanation: During video playback, image frames change continuously and regularly, and the inter-frame pixel change shows continuity and motion vectors (such as the direction and speed of object movement), the system can detect the stability of these frame rates and the regularity of local change areas to determine it as a video stream; For example: the user plays a TV series on a video website, the system detects that the frame rate is 24fps, the image area moves smoothly over time, and the content switches continuously, which can be identified as a video stream; III. Recognition and examples of fast scrolling text and game pictures, 1. Fast scrolling text, principle explanation: When the system detects that there is a large area of vertical or horizontal continuous change in the screen, the change area is a high-contrast character pattern, and the position of the text area has a consistent directional movement, the system will determine it as fast scrolling text; For example: when the LED screen plays news or in the live broadcast of the pop-up, there is a continuous flow of text on the screen, and the system can identify it as fast scrolling text content; 2. Game pictures, principle explanation: Game pictures usually have high dynamicity, including complex changes of multiple independent areas, fast refresh rate, and frequent local motion vectors (such as character movement and scene switching), the system can identify it as a game picture by analyzing multi-area synchronous change + high frame rate fluctuation + nonlinear motion trajectory; For example: when the user runs a first-person shooter game, the screen presents a combination of rapid switching, perspective rotation, and multiple target motion, and the system can identify it as game content; In summary, through the above methods, the system not only accurately identifies the image content type, but also dynamically optimizes the display parameters (such as refresh rate matching, dynamic blur adjustment, backlight adjustment, etc.) based on the image content type, to achieve more efficient image adaptation and presentation experience.

[0036] In the embodiment, the step S3 of dynamically adjusting the content change frequency index of the display screen according to the pixel change rate further comprises: S31: based on the pre-divided grid area of the display screen, extracting the change pixel number proportion of the grid area; S32: judging whether the change pixel number proportion matches the pixel change rate; S33: if yes, constructing a corresponding local pixel change weight according to the change pixel number proportion, identifying the content complexity of the grid area, and dynamically adjusting the grid scale of the grid area, wherein the content complexity specifically includes dynamic content and static content, and the grid scale specifically includes coarse grid and fine grid.

[0037] In this embodiment, the system extracts the proportion of changed pixels in different grid areas based on the pre-divided grid areas of the display screen, and then determines whether the proportion of changed pixels matches the pixel change rate to perform corresponding steps; for example, when the system determines that the proportion of changed pixels in different grid areas cannot match the pixel change rate, the system considers that some areas should have continuous changes (such as video windows), but the number of changed pixels is much lower than the normal threshold, which may be caused by video decoding abnormalities, system rendering lag, playback source interruption, or freeze of false frames. The system extracts the areas with unmatched proportions of changed pixels, reevaluates their dynamic trends in multiple consecutive frames, combines the change rate curve with the state of the surrounding grid areas, and determines whether they are local dynamic images (such as marquee, loading bar), lag frames, frozen content, and false exceptions caused by misclassified grid areas. At the same time, the sensitivity of the “change rate matching” is adjusted adaptively according to the area type, such as allowing a small amount of change in static image areas, requiring high frame rate continuous change in game screen areas, allowing linear movement but maintaining content coherence in fast scrolling text areas, and confirming that a certain area does not meet the expected pixel change behavior of the current display type. The system can automatically switch the recognition mode (such as switching from “static image” to “video” mode), issue an exception warning or record a log, and use it for performance tuning or content recognition to optimize image processing strategies, such as reducing resource scheduling, resuming playback, refreshing buffer frames, etc. For example, when the system determines that the proportion of changed pixels in different grid areas can match the pixel change rate, the system considers that all grid areas of the display screen can change normally. The system constructs the corresponding local pixel change weight based on the proportion of changed pixels, identifies the content complexity of different grid areas, and the content complexity specifically includes dynamic content and static content. According to different content complexity, the grid scale of different grid areas is adjusted dynamically, and the grid scale specifically includes coarse grid and fine grid.The system can accurately determine which areas are in a dynamic state and which areas are relatively static content by identifying the proportion of the number of changed pixels in different grid areas and matching it with the preset pixel change rate. With this judgment mechanism, the system can avoid averaging the entire screen image, thereby realizing localized scheduling of computing resources and preferentially allocating more computing power to high dynamic areas (such as videos and game screens), significantly improving image processing efficiency. At the same time, based on the content complexity (dynamic or static content) of each grid area, the system can dynamically adjust the grid size. For example, fine grids can be applied to frequently changing areas to provide higher precision image analysis, while coarse grids can be used for static areas to reduce unnecessary computational load. This adaptive grid reconstruction mechanism effectively balances image analysis accuracy and computational cost, allowing the system to remain smooth and stable even in high frame rate scenarios or multi-task concurrent situations. Furthermore, after constructing local pixel change weights based on pixel changes, the system can provide targeted parameter references for subsequent image enhancement (such as dynamic noise reduction, content recognition, and frame interpolation). For example, for dynamic content areas, the system can improve frame rate estimation accuracy, while static areas can prioritize edge sharpening or color stabilization processing. This differentiated image processing approach driven by grid size and content complexity helps improve the overall visual perception of images and the accuracy of content recognition.

[0038] It should be noted that the corresponding local pixel change weight is constructed based on the number of changed pixels, the content complexity of the grid area is identified, and the grid size of the grid area is dynamically adjusted. Specific examples are as follows: Suppose a 4K display screen (3840x2160 pixels) is pre-divided by the system into 12x12 grid areas (a total of 144 grids), with each grid area being 320x180 pixels in size. Step 1: Collect the proportion of the number of changed pixels. The system continuously collects two frames of images and calculates the number of changed pixels in each grid area: Grid X1 (top left corner of the screen): Among the 320x180 = 57,600 pixels, about 46,080 pixels have changed in brightness or color, with a change pixel proportion of about 80%; Grid X2 (center of the screen): In the same size area, only about 2,880 pixels have changed, with a change pixel proportion of about 5%; Grid X3 (bottom right corner of the screen): The change pixel proportion is about 40%; Step 2: Construct the local pixel change weight. The system maps the change pixel proportion to a weight value, with a weight range of 0-1: Grid X1 change proportion 80%, corresponding weight 0.9, indicating a high dynamic area; Grid X2 changes 5%, corresponding weight 0.1, represents low dynamic (almost static) area; Grid X3 changes 40%, corresponding weight 0.5, medium dynamic area; Step 3, identify content complexity, according to the weight, the system identifies: X1 area is dynamic content area (such as video playback area, animation picture, etc.); X2 area is static content area (such as LOGO, watermark, background picture); X3 area is medium complexity content area (such as local animation or part of dynamic elements); Step 4, dynamically adjust grid size, in order to optimize processing: X1 area (dynamic area): the system subdivides the grid into smaller subgrid, such as 3 × 3 subgrid, a total of 9 subareas, each subgrid is about 107 × 60 pixels; This can more finely monitor local pixel changes and capture subtle dynamics; X2 area (static area): merge with adjacent static grid into larger coarse grid, such as 3 grids into 1, reduce processing frequency, save computing resources; X3 area (medium area): maintain the original grid size, balance accuracy and performance; In summary, in the above example content, the system monitors the dynamic area in detail when playing video, can timely find local picture changes or abnormalities, ensure picture quality and response speed; For static area, reduce the computational load and save power consumption. Through the specific number of pixels and the size of the grid, the whole process from the change of pixel ratio to the weight, then to the identification of content complexity, and finally the dynamic adjustment of the grid size is described in detail.

[0039] In this embodiment, the step S5 of dynamically limiting the upper limit of the brightness change slope of the display screen and constructing the brightness change gradient of the display screen further comprises: S51: based on the continuous image frames of the display screen, extracting the brightness distribution statistical value of the continuous image frames, generating the brightness jump point in the preset time window, wherein the brightness distribution statistical value specifically includes average brightness, maximum brightness and minimum brightness; S52: determine whether the brightness jump point exceeds the preset mutation threshold; S53: if yes, according to the brightness jump point, identify the brightness mutation area corresponding to the continuous image frame, apply a preset exponential decay filter to the brightness mutation area for slow transition of brightness, real-time detect the adjacent area of the brightness mutation area, and dynamically limit the brightness change amplitude between the adjacent areas.

[0040] In the embodiment, the system extracts the luminance distribution statistical values of the continuous image frames collected in advance based on the display screen, the luminance distribution statistical values specifically include average luminance, maximum luminance and minimum luminance, generates the luminance jump points in the pre-set time window, and then the system judges whether the luminance jump points exceed the pre-set mutation threshold to execute corresponding steps; for example, when the system judges that the luminance jump points in the time window do not exceed the pre-set mutation threshold, the system considers that the luminance of the current display screen is stable in the time window, there is no obvious luminance mutation, the system keeps the current luminance adjustment mechanism unchanged, if there is an automatic luminance adjustment function, the current luminance level can be maintained to avoid excessive adjustment causing visual jump, and when there is no jump in continuous multiple time windows, the image frame sampling frequency can be appropriately reduced to save system resources, and the window segment is marked as a "luminance stable period"; for example, when the system judges that the luminance jump points in the time window exceed the pre-set mutation threshold, the system considers that the luminance of the current display screen is unstable in the time window, there is luminance mutation, the system identifies the luminance mutation regions of the continuous image frames according to different luminance mutation points, applies a pre-set exponential decay filter to perform luminance slow transition on the luminance mutation regions, and dynamically limits the luminance change amplitude between adjacent regions of different luminance mutation regions; when the system identifies the luminance mutation points in the time window and performs dynamic processing, the sudden luminance change of the local region in the picture can be effectively smoothed, and the dazzling effect or visual stimulation caused by sudden light and dark switching can be avoided. The slow transition mechanism based on the exponential decay filter helps to build a softer and more natural picture performance, thereby reducing the visual fatigue of the user in a long-time watching process and improving the watching comfort. At the same time, by smoothing the adjustment of the luminance mutation region and dynamically limiting the luminance change amplitude between adjacent regions, the obvious boundary between high-light and low-light regions can be avoided, thereby reducing the interference of image artifacts or halo effect on the content, which not only improves the clarity of image details in the transition region, but also improves the overall image recognition, especially suitable for scenes playing dynamic images, advertisement videos or high-contrast content. Moreover, the real-time detection and dynamic processing of the system on the luminance mutation belong to an intelligent luminance adjustment strategy, which can adaptively adjust the output luminance according to the content change, thereby avoiding unnecessary luminance increase, reducing the backlight driving pressure and power consumption fluctuation. The luminance adjustment logic is only triggered and processed in the mutation region, which ensures the effective allocation of computing resources and response efficiency. The strategy helps to prolong the service life of the display device and improve the adaptive performance in complex environments.

[0041] It should be noted that, according to the luminance jump point, the luminance mutation region corresponding to the continuous image frame is identified, a preset exponential decay filter is applied to the luminance mutation region for luminance slow transition, the adjacent region of the luminance mutation region is detected in real time, and the luminance change amplitude between the adjacent regions is dynamically limited. The specific example is as follows: Suppose that the outdoor LED advertisement large screen is playing content switching, and the content is switched from a "night scene picture" to a "daylight strong light picture", the steps are simulated as follows: Step 1, identify the luminance mutation region, the original picture is a night scene (the average luminance is about 40 / 255), and the switching is a daytime street scene (the average luminance is about 200 / 255). The system detects that the luminance jump value is 160, which is far higher than the set mutation threshold (for example, the threshold is 80). The luminance value of the sun region in the street scene picture is even up to 250. The system marks the high-light region (for example, a central 200x200 pixel region) as a "luminance mutation region"; Step 2, apply an exponential decay filter to the mutation region, set the exponential decay factor a = 0.7, the current frame luminance = a x last frame luminance + (1-a) x current target luminance, the initial frame luminance is 40, and the target luminance is 250: The first frame transition: 0.7x40+0.3x250≈109; The second frame: 0.7x109+0.3x250≈156; The third frame: 0.7x156+0.3x250≈184; ... until close to 250, so the luminance gradually rises and will not flash white; Step 3, detect the adjacent region and limit the luminance change amplitude. If the luminance of the region around the mutation region still maintains about 40, to avoid the luminance boundary being too abrupt, the system will judge whether the luminance change of the adjacent region is too small or too large. If the luminance difference between the adjacent region and the mutation region exceeds the set limit (for example, it cannot exceed 60), the luminance of the adjacent region is automatically slowly raised (or adapted according to the background), so that the picture transition is smooth; As described above, in the example content described above, the high-light picture of the display screen will not be displayed instantly, preventing stimulation to the human eye, especially in the night viewing scene, effectively improving the comfort level and enhancing the natural transition effect of the image. The luminance is smoothly transitioned from low to high, and there is no obvious "luminance fault" between the surrounding region and the mutation region. The picture is more natural and has no light spots, improving the image quality stability of the LED large screen or display system. In the case of fast content switching, advertisement replacement, weather scene transformation, etc., the system can intelligently control the large amplitude luminance jump to ensure continuous visual experience.

[0042] In the embodiment, in the step S2 of judging whether the image content refresh speed matches the preset backlight refresh response speed, further comprising: S21: based on the continuous image frames of the display screen, extracting pixel differences between the continuous image frames, and constructing a dynamic heat map of the display screen according to the pixel differences; S22: judging whether the dynamic heat map can be mapped to physical display screen coordinates of the display screen; S23: if yes, identifying high-frequency irregular jitter of the dynamic heat map, marking the high-frequency irregular jitter as an abnormal frame of the display screen, demarcating a corresponding to-be-reconstructed region according to the abnormal frame, and constructing a candidate frame block of the to-be-reconstructed region through a preset pixel texture similarity.

[0043] In the present embodiment, the system extracts pixel differences between consecutive image frames of the display screen based on the consecutive image frames, constructs a dynamic heat map of the display screen according to different pixel differences, and then judges whether the dynamic heat map can be mapped to the physical display screen coordinates of the display screen to perform corresponding steps. For example, when the system determines that the dynamic heat map of the display screen cannot be mapped to the physical display screen coordinates of the display screen, the system considers that the coordinate system / resolution / geometry relationship used by the heat map is inconsistent with the actual coordinate system of the physical screen, resulting in that the heat map cannot find the correct position on the physical screen. The system re-establishes the mapping relationship between the physical coordinates and the heat map by reading the real-time resolution, pixel ratio, rotation angle and splicing layout of the current display screen, and introduces feature point matching for geometric correction when necessary. At the same time, the display timing should be synchronized to ensure that the heat map data is consistent with the frame time of the physical screen content, so as to restore the accurate correspondence between the heat map and the screen. For example, when the system determines that the dynamic heat map of the display screen can be mapped to the physical display screen coordinates of the display screen, the system considers that the heat map is consistent with the actual coordinate system of the physical screen. The system identifies high-frequency irregular jitter of the dynamic heat map, marks the high-frequency irregular jitter as abnormal frames of the display screen, and delimits the corresponding to-be-reconstructed region of the display screen according to the abnormal frames. When the heat map can be accurately mapped to the physical display screen coordinates, the system can accurately detect high-frequency irregular jitter signals by using the consistency, and identify the high-frequency irregular jitter signals as abnormal frames. Such jitter is often caused by hardware interface interference, drive delay, cache frame error or external electromagnetic interference. By marking these abnormal frames, the system can avoid that they are regarded as normal pictures to participate in display, thereby reducing the situation of picture flicker, jitter or misplacement, significantly improving the stability and consistency of the picture. At the same time, through the marking of the abnormal frames, the system can accurately delimit the corresponding to-be-reconstructed region, instead of globally repairing the whole picture. Such localized processing mode can reduce the interference of irrelevant regions, reduce the complexity of reconstruction operation, avoid secondary damage to the normal region picture quality, and ensure that the reconstruction process is more efficient, reduces the delay, improves the response speed of the system in the dynamic content scene, and searches for candidate frame blocks from historical frames based on the preset pixel texture similarity algorithm to replace or supplement the pixel data of the abnormal region after determining the to-be-reconstructed region. The texture similarity can comprehensively consider brightness distribution, edge features, color gradient and other information, so as to ensure that the selected candidate frame is highly matched with the current picture in details, color and texture. In this way, not only the abnormal region can be smoothly repaired, but also the natural transition and visual consistency of the whole picture can be maintained, thereby providing users with a smoother and more realistic viewing experience.

[0044] It should be noted that the high-frequency irregular jitter of the dynamic heat zone map is identified, the high-frequency irregular jitter is marked as an abnormal frame of the display screen, a corresponding to-be-reconstructed region is demarcated according to the abnormal frame, a candidate frame block of the to-be-reconstructed region is constructed through a preset pixel texture similarity, and a specific example is as follows: Display: 1920x1080, frame rate 60 fps, block division: 32x32 pixel blocks (1024 pixels per block), about 6480 blocks on the full screen, time window W=30W=30 frames (0.5 s) for spectrum analysis and jitter determination; Determination threshold: high-frequency energy ratio threshold , spectrum entropy threshold , block instantaneous difference threshold (gray scale difference), frame abnormality proportion threshold (10% block abnormality, that is, frame abnormality); Suppose that a channel advertisement is being played, and local abnormal jitter occurs in the t-th frame at a certain moment, the following is processed step by step, Step 1, first calculate the frame difference and construct the block time sequence, for consecutive frames , first calculate the block-level average frame difference sequence of each frame; for a block b, the block average frame difference of each frame is: For example (take 30 items of sequence of a block b, unit: gray scale):

[0045] Note that the 6th and 10th items are abnormal pulses (120, 130); Step 2, spectrum analysis - calculate high-frequency energy ratio (HF) and spectrum entropy (H), for Discrete Fourier transform (FFT) is performed to obtain the power spectrum For the sake of example, it is assumed that the statistical result after FFT is: Total energy

[0046] High-frequency (>8hz) energy

[0047] Spectrum entropy calculation (normalized spectrum entropy) gives

[0048] Therefore, it is determined that: ; Step 3, instantaneous difference detection and abnormal block marking, take the instantaneous block difference of the current frame (n=0) , compare the threshold , obviously 120>12, so the block is marked as an abnormal block in the current frame, and the same process is performed on all blocks in the frame, and it is assumed that a total of 820 abnormal blocks are detected (about 12.65% of 6480), because So the current frame is determined as an abnormal frame; Step 4, the ROI is obtained by merging the connected blocks, and a padding 1 block is added to accommodate the edge effect. Assuming that three larger connected regions are found, the largest one is composed of 48 adjacent abnormal blocks, and the ROI after padding contains 64 blocks (pixels: 64x1024=65,536 pixels), which is the key area that needs to be reconstructed; Step 5, search for candidate blocks in the history non-abnormal frames (texture similarity), search for candidates in the non-abnormal frames in the last 120 frames (2 seconds) on the time axis according to the position or by coarse optical flow displacement estimation. For each block in the ROI, R (fixed position or moved to the corresponding position in the history frame by optical flow estimation), calculate the similarity; The similarity is calculated using SSIM (Structural Similarity Index), and the candidate blocks and the target block are calculated as follows:

[0049] For example, for a block R in a ROI, the SSIMs of three candidates C1, C2, and C3 found in the history frames are 0.93, 0.87, and 0.62 respectively, and the SSIM threshold is set to 0.85, so C1 (0.93) and C2 (0.87) are selected as the main candidates (K=2); Step 6: motion compensation (if there is displacement) and alignment, if the scene has motion (camera / picture motion), first calculate the optical flow to warp the candidate blocks to the current frame position, For example, the displacement vector of C1 is calculated using Lucas-Kanade optical flow , and the displacement of C2 is Affine interpolation or bilinear interpolation is performed on C1 and C2 for alignment; Step 7: candidate fusion and final reconstruction, weighted fusion is performed using similarity as weight (normalized weight):

[0050] The fused pixel value is:

[0051] If there is a discontinuity at the edge after fusion, local median filtering or edge-preserving filtering (such as bilateral filtering) is used to avoid blurring, and the reconstructed block is replaced back to the ROI area of the current frame to complete the local repair; Step 8, verification and fallback mechanism, after reconstruction, calculate frame difference and SSIM of ROI compared with surrounding frames: if SSIM of reconstructed frame compared with previous and next frame is improved to >0.9, and frame difference is reduced to <T_d, then consider the repair successful, if no suitable candidate (for example, all candidate SSIM <0.7), then the system activates the degradation strategy, using the median of past N frames or the result of temporal low-pass filtering as the reconstruction candidate, or marking as needing manual / offline analysis; In summary, in the above example, the system uses high-frequency jitter to determine the spectrum (HF ratio) and spectral entropy, which can distinguish between narrow-band periodic noise (low entropy) and wide-band irregular jitter (high entropy); This criterion is commonly used and reliable in the field of signal processing; Block-based processing (32x32) ensures detection accuracy while considering computational efficiency, SSIM is used for block-level texture similarity search to better reflect structure and perceptual quality, which is more consistent with visual perception than pure MSE, motion compensation (optical flow / block matching) ensures that the candidates extracted from historical frames are aligned in space and time, thereby avoiding reconstruction artifacts, and the fusion weight is allocated according to similarity to balance the information of the most recent high-quality frame and the slightly older frame, taking into account details and stability.

[0052] In the embodiment, in the step S4 of judging whether the content change frequency index meets the display setting of the display screen by the user, further comprising: S41: identifying a scene type of the display screen based on a display mode preset by the user for the display screen, wherein the display mode specifically includes a power saving mode, a standard mode, and a high performance mode, and the scene type specifically includes a reading scene, a video scene, and a game scene; S42: judging whether the display mode matches the refresh requirement of the scene type; S43: if not, constructing a corresponding pixel motion area according to the pixel difference between consecutive image frames, generating a displacement path of each pixel point through the pixel motion area, collecting a motion discontinuous area of the pixel motion area according to the displacement path, detecting a corresponding missing pixel from the motion discontinuous area, scaling the displacement path by a vector offset distance according to a proportion to obtain a target position of each pixel required for frame interpolation, and combining pixels of the missing pixel as an interpolation content between original frames.

[0053] In the embodiment, the system identifies the scene type of the display screen based on the display mode preset by the user for the display screen, the display mode specifically includes the power saving mode, the standard mode and the high performance mode, the scene type specifically includes the reading scene, the video scene and the game scene, and then the system judges whether the display mode matches the refresh requirement of different scene types to execute corresponding steps; for example, when the system determines that the display mode preset by the user for the display screen can match the refresh requirement of different scene types, the system considers that the current display mode and the refresh performance requirement of the scene are in the optimal or better state, no additional mode switching or refresh parameter adjustment is needed, the system does not switch, the delay and visual discomfort caused by mode change are reduced, the power consumption fluctuation caused by frequent adjustment is avoided, and if it is detected that part of the parameters meet the requirement but have room for improvement (such as the brightness is slightly lower than the optimal value of the scene), the individual parameters can be dynamically fine-tuned without changing the mode, for example, in the video mode, the color temperature is slightly adjusted to adapt to the night viewing, and the scene type and user interaction are continuously monitored, when it is detected that the scene change trend is obvious (such as the video playback ends and returns to reading), the mode switching suggestion or the corresponding mode configuration is prepared in advance; for example, when the system determines that the display mode preset by the user for the display screen cannot match the refresh requirement of different scene types, the system considers that the current display mode and the refresh performance of the scene are not in the optimal state, the system constructs the corresponding pixel motion area according to the pixel difference between the continuous image frames, generates the displacement path of each pixel point through different pixel motion areas, collects the motion non-continuous area of the pixel motion area according to the displacement path, detects the corresponding missing pixels from the motion non-continuous area, performs vector offset distance scaling on the displacement path in proportion, obtains the target position of each pixel required for the frame interpolation, and takes each pixel target position as the frame interpolation content between the original frames to combine the missing pixels.When the display mode does not match the scene refresh requirements, some high-speed changes or large displacement pictures may appear to be stuck or torn. By constructing a pixel motion area based on pixel differences and generating a displacement path for each pixel, the continuous motion trajectory of the pixel on the time axis can be accurately restored. The missing pixel detection and interpolation in the motion discontinuity area can effectively reduce the time gap between frames and improve the smoothness of the picture. Fast switching scenes (such as high-speed moving shots in games or motion pictures in videos) are smoother. Motion discontinuity areas often have the largest differences and the most severe detail loss in the original frame and the target frame. By scaling the vector offset distance of the displacement path, the target position of each missing pixel can be accurately calculated during the interpolation process, ensuring that the interpolation content is consistent with the actual physical motion trend. This pixel-level accurate calculation can significantly reduce motion artifacts, ghosting, and blurring in conventional interpolation algorithms, ensuring the restoration of dynamic details. Compared to directly increasing the refresh rate to meet the scene requirements, this interpolation method can achieve high refresh rate visual effects at a lower refresh rate, meeting the visual experience requirements of high-speed motion scenes while avoiding the energy consumption increase caused by frequent switching of display modes or continuous operation of high-performance modes. It is suitable for use in power-sensitive scenarios such as mobile devices and portable display devices.

[0054] It should be noted that, according to the pixel difference between consecutive image frames, a corresponding pixel motion area is constructed. The pixel motion area is used to generate a displacement path for each pixel point. According to the displacement path, a motion discontinuity area of the pixel motion area is collected. The missing pixels are detected from the motion discontinuity area. The displacement path is scaled by vector offset distance to obtain the target position of each pixel required for interpolation. The target position of each pixel is used as the interpolation content between the original frames. The missing pixels are combined by pixel, and the specific example is as follows: Suppose there are two images with a resolution of 5x5 pixels and a frame rate of 30fps. A frame is inserted between them (increased to 60fps). 1. Pixel difference calculation. Assuming that the RGB values of position (3, 3) in the previous and next two frames are (100, 100, 100) and (130, 130, 130) respectively, first calculate the pixel difference between the two consecutive frames (or multiple frames). The formula is:

[0055] wherein, represents the brightness or color value of the current frame at position (x, y), represents the pixel difference value. The system sets a threshold , and the pixel points greater than the threshold are considered to have changed significantly, and a "pixel motion area" is constructed according to the spatial distribution of these changed pixels. The difference is:

[0056] If the threshold = 20, this pixel is marked as a motion pixel. Repeat the process, assuming we find (2,3), (3,3), (4,3) continuous changes, forming a motion region; 2, displacement path calculation, in the motion region, calculate the motion vector (u, v) of each pixel by optical flow method (such as Horn-Schunck or Lucas-Kanade algorithm) to get the displacement path of the pixel in time, for example, the displacement path can be expressed by the formula:

[0057] In this way, the motion trajectory of each pixel can be continuously tracked, and through optical flow analysis, we get: (2,3) displacement: (+2,0) (3,3) displacement: (+2,0) (4,3) displacement: (+2,0) This means that they move 2 pixels to the right in 1 / 30 seconds; 3, motion discontinuity detection, on the motion path, if the displacement direction or speed of adjacent pixels in a certain period of time changes suddenly, and there is no corresponding pixel value in the target frame at this position, it is marked as "motion discontinuity area", in these areas, the position of the missing pixel needs to be generated by interpolation, by checking the position of the pixels in the target frame in this motion area, it is found that (4,3) will exceed the boundary or overlap in the next frame, which belongs to the non-continuous area and needs to be interpolated; 4, vector scaling, in order to generate an interpolated frame, the system will scale the displacement vector by time. For example, if the original frame interval is Δt, and we need to generate an intermediate frame (Δt / 2), the scaling formula is:

[0058] In this way, the target position of the pixel at the intermediate time point can be obtained, and when the intermediate frame is inserted, the displacement is scaled by half:

[0059] That is, each pixel moves 1 pixel to the right in the intermediate frame; 5, interpolation frame generation, mapping the scaled displacement path to the intermediate frame coordinate system, selecting the corresponding pixel value from the original frame (or using bilinear interpolation, pixel texture matching, etc.) to fill the missing pixel position, forming a complete interpolation frame picture; for (3, 3) pixels, the intermediate frame position=(3+1, 3)=(4, 3), fill (4, 3) with the pixel value of (3, 3) from the original frame; the missing (2, 3) position is synthesized from the adjacent pixels by texture similarity; the final intermediate frame picture is motion continuous, no obvious gap, and the visual effect is smooth.

[0060] In the embodiment, in the step S1 of obtaining the image content refresh speed of the current display content based on the inter-frame variation of the display screen pre-acquired for the current display content, further comprising: S11: based on the preset frequency of the display screen, intercepting the frame data structure of the current display content, wherein the frame data structure specifically includes a pixel matrix, brightness information and a color channel; S12: judging whether the frame data structure matches the image hot area preset by the display screen; S13: if yes, obtaining the pixel motion speed of the current display content, and dynamically adjusting the refresh period of the image hot area according to the pixel motion speed.

[0061] In this embodiment, the system intercepts the frame data structure of the current display content based on the preset frequency of the display screen, which specifically includes the pixel matrix, brightness information, and color channel, and then determines whether these frame data structures match the image hot area preset by the display screen to execute corresponding steps. For example, when the system determines that the frame data structure of the current display content cannot match the image hot area preset by the display screen, it is considered that the pixel distribution, brightness information, or color channel characteristics in the current frame deviate from the predefined key area mode, meaning that the display content is inconsistent with the expected hot area layout, such as hot spot area position offset, size change, or no preset content, the system will perform secondary analysis on the current frame data, reposition the potential high-attention area, and simultaneously find the candidate area close to the preset hot area characteristics through methods such as pixel difference detection, brightness gradient analysis, and color histogram matching, generate its similarity score, and if the similarity of the candidate area is higher than the set threshold, the system will dynamically update the hot area mapping relationship to align the new hot area position with the actual content, if no area meeting the conditions is found, a warning is triggered or switched to the default display mode to ensure the stability of display quality and user experience. For example, when the system determines that the frame data structure of the current display content can match the image hot area preset by the display screen, it is considered that the pixel distribution, brightness information, or color channel characteristics in the current frame are consistent with the predefined key area mode, the system will obtain the pixel motion speed of the current display content, and dynamically adjust the refresh cycle of the image hot area according to different pixel motion speeds. When the system determines that the frame data structure of the current display content can match the image hot area preset by the display screen, it means that the key area (such as important visual focus or high dynamic change area) in the current display picture is highly consistent with the predefined hot area layout. This consistency ensures the accurate identification of the system on the display content, so that the subsequent dynamic refresh strategy based on the hot area can accurately act on the area that needs most attention, thereby improving the overall display efficiency and visual experience. Meanwhile, based on the matched hot area, the system further obtains the motion speed of each pixel in the current display content, dynamically adjusts the refresh cycle of the corresponding hot area by analyzing the size and distribution of the pixel motion speed, specifically, for the area with fast motion speed, the system will shorten the refresh cycle to increase the refresh rate, ensuring the smoothness and clarity of the motion picture, while for the area with slow motion or static, the refresh cycle can be appropriately prolonged to reduce unnecessary refresh frequency, thereby saving power consumption and reducing the heat generation of the display screen. Moreover, this mechanism of adjusting the hot area refresh cycle based on the pixel motion speed helps to realize "regional intelligent refresh", effectively balancing the display performance and energy consumption, especially suitable for application scenarios with rich dynamic content and frequent local changes (such as video playback, game pictures), through precise hot area refresh control, it not only ensures the high-quality display of key visual areas, but also avoids the energy waste caused by high-frequency refresh of the whole screen, improving the endurance and user experience of the device.

[0062] Reference is made to the accompanying drawings Figure 2 For the backlight dynamic refresh system based on content change frequency in an embodiment of the present application, comprising: The acquisition module 10 is configured to acquire an image content refresh speed of the current display content based on an inter-frame change amount of the current display content pre-acquired by the display screen; The judgment module 20 is configured to judge whether the image content refresh speed matches a preset backlight refresh response speed; The execution module 30 is configured to, if not, identify image difference data between the current display content and a previous frame display content, apply a preset frame difference method to calculate a pixel change rate within a unit time, and dynamically adjust a content change frequency index of the display screen according to the pixel change rate, wherein the content change frequency index specifically includes low dynamic, medium dynamic and high frequency dynamic; The second judgment module 40 is configured to judge whether the content change frequency index meets a display setting of the display screen by a user; The second execution module 50 is configured to, if not, adjust a PWM frequency and a duty cycle of a backlight source of the display screen according to a preset mapping strategy of the display screen by the user, detect a frame rate fluctuation of the display screen, add a preset intermediate compensation frame of gray scale when the backlight source of the display screen changes, dynamically limit an upper limit of a brightness change slope of the display screen, and construct a brightness change gradient of the display screen, wherein the mapping strategy specifically includes a brightness strategy, a picture quality strategy and a perception priority strategy.

[0063] In the embodiment, the acquisition module 10 acquires the image content refresh speed of the current display content based on the interframe variation quantity of the current display content pre-acquired by the display screen, and then the judgment module 20 judges whether the image content refresh speed matches the backlight refresh response speed set in advance to execute corresponding steps; for example, when the system determines that the image content refresh speed of the current display content can match the backlight refresh response speed set in advance, the system considers that the interframe image variation frequency of the current display content is within the range that the backlight source can respond in time, and there is no overshoot or delay phenomenon, the system keeps the current backlight PWM modulation strategy and refresh parameter, and does not perform additional frame insertion or gamma adjustment, at the same time, since the synchronization is good, the "energy saving compensation mode" can be started, the local backlight weakening is started for the image static area, the frame rate limiting is started for the area with low refresh intensity, the backlight power output is controlled, the energy consumption and heat are reduced, and the inertia tracking strategy is executed for the refresh direction and speed of the local hot area to avoid frequent brightness jumps caused by slight changes; for example, when the system determines that the image content refresh speed of the current display content cannot match the backlight refresh response speed set in advance, the execution module 30 considers that the interframe image variation frequency of the current display content is not within the response range of the backlight source, the system identifies the image difference data of the current display content and the previous frame display content, applies the frame difference method set in advance to calculate the pixel variation rate in a unit time, dynamically adjusts the content variation frequency index according to different pixel variation rates, and the content variation frequency index specifically includes low dynamic, medium dynamic and high frequency dynamic; the system can realize intelligent identification of the dynamic level of the display content by judging whether the current image content refresh speed matches the backlight response speed and adjusting the content variation frequency index (such as low dynamic, medium dynamic and high frequency dynamic) according to the interframe image difference, this classification can effectively optimize the backlight adjustment strategy, make the change of the display content more coordinated with the response of the backlight, and further improve the overall picture smoothness, definition and transition naturalness, thereby avoiding visual fatigue problems such as trailing and flickering caused by unmatched refresh; when the system identifies that the image variation is low dynamic or static, the backlight refresh frequency can be reduced to reduce unnecessary power consumption; and under high frequency dynamic variation, the backlight response speed is improved to ensure that the brightness is updated synchronously with the image content, and to avoid ghosting or brightness lag, the scheme can flexibly adjust the backlight strategy according to the dynamic level of the content to balance energy saving and performance, and is particularly suitable for scenes with high energy efficiency requirements such as mobile devices and smart televisions, and the pixel variation rate in a unit time is calculated by the frame difference method to establish the content variation frequency index, the system has stronger image perception and analysis capability, it can not only judge the variation trend at the content level, but also feed back the judgment results to the backlight source in real time to realize the content-driven display optimization mechanism, this mechanism improves the intelligence of the system, and provides a structural basis for further combining AI image recognition, scene analysis and other functions.Then the second judging module 40 judges whether the content change frequency index meets the user's display settings of the display screen to perform corresponding steps; for example, when the system determines that the content change frequency index of the display screen meets the user's display settings of the display screen, the system considers that the current display state has reached the user's expected requirements for display clarity, smoothness, response speed or energy saving mode, etc. The system detects the display mode of the display screen. If the user sets the high refresh, dynamic clarity mode, the system maintains a higher content update rate and backlight synchronization. If the user sets the energy saving or eye protection mode, the system ensures a lower change frequency to avoid the power consumption and visual fatigue caused by frequent changes, while keeping the current refresh and backlight parameters stable operation, entering the balance state of power consumption and experience, suspending unnecessary refresh adjustment calculation, reducing resource consumption, and recording the current configuration as the recommended reference parameter for fast switching configuration under similar content. For example, when the system determines that the content change frequency index of the display screen does not meet the user's display settings of the display screen, the second execution module 50 considers that the current display state cannot meet the user's expected requirements for display clarity, smoothness, response speed or energy saving mode, etc. The system adjusts the PWM frequency and duty cycle of the display screen backlight source according to the user's pre-set mapping strategy for the display screen, which specifically includes brightness strategy, image quality strategy and perception priority strategy. The system detects the frame rate fluctuation of the display screen, adds the pre-set intermediate compensation frame when the display screen backlight source changes, dynamically limits the upper limit of the brightness change slope of the display screen, and constructs the brightness change gradient of the display screen.The system dynamically adjusts the PWM frequency and duty cycle of the backlight by detecting the change frequency index of the display content and combining the preset brightness strategy and image quality strategy, which helps to reduce the brightness instability problem caused by frame rate fluctuation. When the system identifies frame rate fluctuation, it inserts a gray-scale intermediate compensation frame and limits the upper limit of the brightness change slope, which can effectively smooth the brightness transition process and avoid the flicker or tearing feeling caused by sudden brightness changes, thereby enhancing display stability and improving overall image quality performance. At the same time, through the perception priority strategy, the system can dynamically match the display effect according to the user's preference for clarity, response speed or energy saving in different scenarios. For example, in reading mode, it tends to reduce frame rate and brightness change rate to reduce visual fatigue, while in video playback or game mode, it increases the refresh response to ensure smoothness. This strategy ensures that the display effect always matches the user's current use intention, optimizes visual perception and operation feedback, and when the system determines that the current display state does not meet the user's energy saving requirements, it can automatically enter the energy efficiency optimization path, such as reducing the backlight PWM frequency, reducing unnecessary dynamic brightness adjustment or suppressing high-frequency frame rate fluctuations, thereby reducing power consumption. In addition, by constructing the brightness change gradient and limiting the upper limit of the brightness change slope, it avoids wasting power due to frequent brightness jumps, which helps to balance the overall system energy consumption and prolong the battery life of mobile devices such as mobile phones and tablets.

[0064] In the embodiment, further comprising: The identification module is configured to extract image data from the data buffer based on the preset data buffer of the display screen, and identify an image frame time interval of the display screen according to the image data, wherein the image data specifically includes color values, gray scale brightness values and transparencies of pixel points. The third judgment module is configured to judge whether the image frame time interval is synchronized with a screen refresh cycle. The third execution module is configured to, if yes, analyze display content of the image data, obtain display parameters of the display screen according to the display content, and construct a frame rate change trend of the image data through the display parameters, wherein the display content specifically includes static images, video streams, fast scrolling texts and game pictures, and the display parameters specifically include frame rates, motion vectors and change area distributions.

[0065] In the embodiment, the system extracts image data from the data buffer pre-set by the display screen, which specifically includes color value, gray scale brightness value and transparency of the pixel points, identifies the image frame time interval pre-set by the display screen according to different image data, and then judges whether the image frame time interval is synchronized with the screen refresh cycle to execute corresponding steps. For example, when the system determines that the image frame time interval pre-set by the display screen cannot be synchronized with the screen refresh cycle, the system considers that the image frame arrives slowly, which indicates that the image generation or transmission speed is not enough, which may cause tearing, lag or delay of the picture. The system will "wait for synchronization" for the display frame, introduces delay to ensure that each frame is aligned with the display refresh signal to avoid tearing, and at the same time, multiple image data are reserved in the data buffer to enable the next frame to be prepared in advance, thereby improving the smoothness and processing allowance. If the frame time interval is not fixed, consistency can be improved by limiting rendering complexity, reducing image quality and optimizing rendering. For example, when the system determines that the image frame time interval pre-set by the display screen can be synchronized with the screen refresh cycle, the system considers that the image can be normally generated on the display screen. The system analyzes what the display content of the image data is, which specifically includes static image, video stream, fast scrolling text and game picture. According to different display content, the display parameters of the display screen are obtained, which specifically include frame rate, motion vector and change area distribution. Through these display parameters, the frame rate change trend of the image data is constructed. The system can ensure that the image is continuously played on the display screen without tearing and lagging by judging the synchronization of the image frame time interval and the display screen refresh cycle, which significantly improves the smoothness and consistency of the user's viewing experience. For example, when playing a video or scrolling subtitles, if the image frame interval is not synchronized with the refresh cycle, picture frame skipping or flickering may occur, which affects the viewing effect. Through the synchronization mechanism, such problems can be effectively eliminated to ensure stable output of the visual effect. At the same time, the system analyzes the display content type of the image data (such as static image, video stream, fast scrolling text or game picture), dynamically adjusts the display parameters such as frame rate, motion vector and change area based on the type difference, so that the image presentation is more in line with the actual scene requirements. For example, for fast scrolling text, the clarity and smoothness of the text can be optimized, while for game pictures, the response frame rate and motion coherence need to be improved. The dynamic adaptation mechanism improves the performance of the display screen in various application scenarios, and the frame rate change trend of the image data can help the system to identify the dynamic characteristics of the image content, such as whether there is intense motion, picture stillness or mutation, so as to judge in advance the display strategy that may be needed (such as high refresh mode or low power mode). For example, when displaying static images for a long time, the refresh rate can be reduced to save energy, while when high-frequency motion is identified, the refresh rate is automatically increased to ensure smoothness. This trend modeling mechanism provides data support for intelligent display management to realize dual optimization of display performance and power consumption.

[0066] In the embodiment, the execution module further comprises: An extraction unit is configured to extract a proportion of changed pixel quantity of the grid area based on the grid area pre-divided by the display screen; A judgment unit is configured to judge whether the proportion of changed pixel quantity matches the pixel change rate; An execution unit is configured to, if yes, construct a corresponding local pixel change weight according to the proportion of changed pixel quantity, identify a content complexity of the grid area, and dynamically adjust a grid scale of the grid area, wherein the content complexity specifically includes dynamic content and static content, and the grid scale specifically includes a coarse grid and a fine grid.

[0067] In this embodiment, the system extracts the proportion of changed pixels in different grid areas based on the pre-divided grid areas of the display screen, and then determines whether the proportion of changed pixels matches the pixel change rate to perform corresponding steps; for example, when the system determines that the proportion of changed pixels in different grid areas cannot match the pixel change rate, the system considers that some areas should have continuous changes (such as video windows), but the number of changed pixels is much lower than the normal threshold, which may be caused by video decoding abnormalities, system rendering lag, playback source interruption, or freeze of false frames. The system extracts the areas with unmatched proportions of changed pixels, reevaluates their dynamic trends in multiple consecutive frames, combines the change rate curve with the state of the surrounding grid areas, and determines whether they are local dynamic images (such as marquee, loading bar), lag frames, frozen content, and false exceptions caused by misclassified grid areas. At the same time, the sensitivity of the “change rate matching” is adjusted adaptively according to the area type, such as allowing a small amount of change in static image areas, requiring high frame rate continuous change in game screen areas, allowing linear movement but maintaining content coherence in fast scrolling text areas, and confirming that a certain area does not meet the expected pixel change behavior of the current display type. The system can automatically switch the recognition mode (such as switching from “static image” to “video” mode), issue an exception warning or record a log, and use it for performance tuning or content recognition to optimize image processing strategies, such as reducing resource scheduling, resuming playback, refreshing buffer frames, etc. For example, when the system determines that the proportion of changed pixels in different grid areas can match the pixel change rate, the system considers that all grid areas of the display screen can change normally. The system constructs the corresponding local pixel change weight based on the proportion of changed pixels, identifies the content complexity of different grid areas, and the content complexity specifically includes dynamic content and static content. According to different content complexity, the grid scale of different grid areas is adjusted dynamically, and the grid scale specifically includes coarse grid and fine grid.The system can accurately determine which areas are in a dynamic change state and which areas are relatively static content by identifying the proportion of the number of changed pixels in different grid areas and matching it with a preset pixel change rate. With this judgment mechanism, the system can avoid averaging the entire screen image, thereby realizing localized scheduling of computing resources and preferentially allocating more computing power to high dynamic areas (such as videos and game screens), significantly improving image processing efficiency. At the same time, according to the content complexity (dynamic content or static content) of each grid area, the system can dynamically adjust the grid size. For example, fine grids can be applied to frequently changing areas to provide higher precision image analysis, and coarse grids can be used for static areas to reduce unnecessary computational load. This adaptive grid reconstruction mechanism effectively balances image analysis accuracy and computational cost, allowing the system to remain smooth and stable even in high frame rate scenarios or multi-task concurrent situations. Furthermore, after the system constructs a local pixel change weight based on pixel changes, it can provide targeted parameter references for subsequent image enhancement (such as dynamic noise reduction, content recognition, and frame interpolation). For example, the system can improve frame rate estimation accuracy for dynamic content areas, while static areas can preferentially perform edge sharpening or color stabilization processing. This differentiated image processing approach driven by grid size and content complexity helps improve the overall visual perception of images and the accuracy of content recognition.

[0068] In the embodiment, the second execution module further includes: A generation unit configured to extract a brightness distribution statistical value of the continuous image frames based on the continuous image frames of the display screen, and generate a brightness jump point within a preset time window, wherein the brightness distribution statistical value specifically includes an average brightness, a maximum brightness, and a minimum brightness. A second judgment unit configured to judge whether the brightness jump point exceeds a preset mutation threshold. A second execution unit configured to, if so, identify a brightness mutation area corresponding to the continuous image frames according to the brightness jump point, apply a preset exponential decay filter to perform a slow transition of brightness on the brightness mutation area, and dynamically limit a brightness change amplitude between adjacent areas of the brightness mutation area.

[0069] In the embodiment, the system extracts the luminance distribution statistical values of the continuous image frames based on the continuous image frames collected in advance by the display screen, the luminance distribution statistical values specifically include average luminance, maximum luminance and minimum luminance, generates the luminance jump points in the time window set in advance, and then the system judges whether the luminance jump points exceed the mutation threshold set in advance to execute corresponding steps; for example, when the system judges that the luminance jump points in the time window do not exceed the mutation threshold set in advance, the system considers that the luminance change of the current display screen in the time window is stable, there is no obvious luminance mutation, the system keeps the current luminance adjustment mechanism unchanged, if there is an automatic luminance adjustment function, the current luminance level can be maintained to avoid excessive adjustment causing visual jump, and when there is no jump in continuous time windows, the image frame sampling frequency can be appropriately reduced to save system resources, and the window segment is marked as a "luminance stable period"; for example, when the system judges that the luminance jump points in the time window exceed the mutation threshold set in advance, the system considers that the luminance change of the current display screen in the time window is unstable, there is luminance mutation, the system identifies the luminance mutation regions of the continuous image frames according to different luminance mutation points, applies the exponential decay filter set in advance to perform luminance slow transition on the luminance mutation regions, and dynamically limits the luminance change amplitude between adjacent regions of different luminance mutation regions; when the system identifies the luminance mutation points in the time window and performs dynamic processing, the sudden luminance change of the local region in the picture can be effectively smoothed, the dazzling effect or visual stimulation caused by sudden light and dark switching can be avoided, the slow transition mechanism based on the exponential decay filter is helpful to build a picture performance that is softer and more natural in transition, thereby reducing the visual fatigue of the user in a long watching process and improving the watching comfort, and through the smooth adjustment of the luminance mutation regions and the dynamic limitation of the luminance change amplitude between adjacent regions, the obvious boundary between high-light and low-light regions can be avoided, thereby reducing the interference of image artifacts or halo effect on the content, which not only improves the clarity of image details in the transition region, but also improves the overall image recognition, especially suitable for scenes playing dynamic images, advertisement videos or high-contrast content, and the real-time detection and dynamic processing of the system on the luminance mutation belong to an intelligent luminance adjustment strategy, which can adaptively adjust the output luminance according to the content change, thereby avoiding unnecessary luminance increase, reducing backlight driving pressure and power consumption fluctuation, the luminance adjustment logic is only triggered and processed in the mutation region, which ensures the effective allocation of computing resources and response efficiency, and the strategy is helpful to prolong the service life of the display device and improve the adaptive performance in complex environments.

[0070] In the embodiment, the judging module further includes: a constructing unit configured to extract pixel differences between consecutive image frames of the display screen based on the consecutive image frames, and construct a dynamic heat map of the display screen according to the pixel differences; a third judging unit configured to judge whether the dynamic heat map can be mapped to physical display screen coordinates of the display screen; a third executing unit configured to, if yes, identify high-frequency irregular jitter of the dynamic heat map, mark the high-frequency irregular jitter as an abnormal frame of the display screen, demarcate a corresponding to-be-reconstructed region according to the abnormal frame, and construct a candidate frame block of the to-be-reconstructed region by using a preset pixel texture similarity.

[0071] In the embodiment, the system extracts pixel differences between consecutive image frames of the display screen based on the consecutive image frames, constructs a dynamic heat map of the display screen according to different pixel differences, and then judges whether the dynamic heat map can be mapped to physical display screen coordinates of the display screen to perform corresponding steps. For example, when the system determines that the dynamic heat map of the display screen cannot be mapped to the physical display screen coordinates of the display screen, the system considers that the coordinate system / resolution / geometry of the heat map is inconsistent with the actual coordinate system of the physical screen, which leads to the fact that the heat map cannot find the correct position on the physical screen. The system re-establishes the mapping relationship between the physical coordinates and the heat map by reading real-time resolution, pixel ratio, rotation angle and splicing layout and other parameters of the current display screen, and introduces feature point matching for geometric correction when necessary. At the same time, the display timing should be synchronized to ensure that the heat map data is consistent with the frame time of the physical screen content, so as to restore the accurate correspondence between the heat map and the screen. For example, when the system determines that the dynamic heat map of the display screen can be mapped to the physical display screen coordinates of the display screen, the system considers that the heat map is consistent with the actual coordinate system of the physical screen. The system identifies high-frequency irregular jitter of the dynamic heat map, marks the high-frequency irregular jitter as abnormal frames of the display screen, and delimits a to-be-reconstructed region of the display screen according to the abnormal frames. The system constructs candidate frame blocks of the to-be-reconstructed region by using pixel texture similarity that is preset in advance. After the heat map can be accurately mapped to the physical display screen coordinates, the system can accurately detect high-frequency irregular jitter signals by using the consistency, and identify the high-frequency irregular jitter signals as abnormal frames. The jitter is often caused by hardware interface interference, drive delay, cache frame error or external electromagnetic interference. By marking the abnormal frames, the system can avoid that the abnormal frames are regarded as normal pictures to participate in display, so as to reduce picture flicker, jitter or misplacement, significantly improve the stability and consistency of the picture, and accurately delimit the to-be-reconstructed region by marking the abnormal frames, instead of globally repairing the whole picture. The localized processing manner can reduce interference of irrelevant regions, reduce complexity of reconstruction operation, avoid secondary damage to normal region picture quality, accurately position the local region to ensure that the reconstruction process is more efficient, reduce delay, improve response speed of the system in a dynamic content scene, and search candidate frame blocks from historical frames based on the preset pixel texture similarity algorithm to replace or supplement pixel data of the abnormal region. The texture similarity can comprehensively consider brightness distribution, edge feature, color gradient and other information, so as to ensure that the selected candidate frame is highly matched with the current picture in detail, color and texture. In this way, the abnormal region can be smoothly repaired, and the natural transition and visual consistency of the whole picture can be maintained, so as to provide a smoother and more realistic viewing experience for a user.

[0072] In the embodiment, the second judging module further includes: An identifying unit is configured to identify a scene type of the display screen based on a display mode preset by the user for the display screen, wherein the display mode specifically includes a power saving mode, a standard mode and a high performance mode, and the scene type specifically includes a reading scene, a video scene and a game scene; A fourth judging unit is configured to judge whether the display mode matches a refresh requirement of the scene type; A fourth executing unit is configured to, if not, construct a corresponding pixel motion area according to a pixel difference between continuous image frames, generate a displacement path of each pixel point through the pixel motion area, collect a motion discontinuous area of the pixel motion area according to the displacement path, detect a corresponding missing pixel from the motion discontinuous area, perform vector offset distance scaling on the displacement path in proportion, obtain a target position of each pixel required for frame insertion, and perform pixel combination on the missing pixel by taking the target position of each pixel as frame insertion content between original frames.

[0073] In the embodiment, the system identifies the scene type of the display screen based on the display mode preset by the user for the display screen, the display mode specifically includes the power saving mode, the standard mode and the high performance mode, the scene type specifically includes the reading scene, the video scene and the game scene, and then the system judges whether the display mode matches the refresh requirement of different scene types to execute corresponding steps; for example, when the system determines that the display mode preset by the user for the display screen can match the refresh requirement of different scene types, the system considers that the current display mode and the refresh performance requirement of the scene are in the optimal or better state, no additional mode switching or refresh parameter adjustment is needed, the system does not switch, the delay and visual discomfort caused by mode change are reduced, the power consumption fluctuation caused by frequent adjustment is avoided, and if it is detected that part of the parameters meet the requirement but have room for improvement (such as the brightness is slightly lower than the optimal value of the scene), the individual parameters can be dynamically fine-tuned without changing the mode, for example, in the video mode, the color temperature is slightly adjusted to adapt to the night viewing, and the scene type and user interaction are continuously monitored, when it is detected that the scene change trend is obvious (such as the video playback ends and returns to reading), the mode switching suggestion or the corresponding mode configuration is prepared in advance; for example, when the system determines that the display mode preset by the user for the display screen cannot match the refresh requirement of different scene types, the system considers that the current display mode and the refresh performance of the scene are not in the optimal state, the system constructs the corresponding pixel motion area according to the pixel difference between the continuous image frames, generates the displacement path of each pixel point through different pixel motion areas, collects the motion non-continuous area of the pixel motion area according to the displacement path, detects the corresponding missing pixels from the motion non-continuous area, performs vector offset distance scaling on the displacement path in proportion, obtains the target position of each pixel required for the frame interpolation, and takes each pixel target position as the frame interpolation content between the original frames to combine the missing pixels.When the display mode does not match the scene refresh requirement, part of the high-speed change or large displacement picture may appear lag or tearing, by constructing the pixel motion area based on the pixel difference, and generating the displacement path of each pixel, the continuous motion trajectory of the pixel on the time axis can be accurately restored, the missing pixel detection and interpolation in the motion discontinuous area can effectively reduce the time gap between frames, thereby significantly improving the picture smoothness, making the fast switching scene (such as game high-speed moving lens or video motion picture) more smooth, and the motion discontinuous area is often the part with the largest difference and the most serious detail loss in the original frame and the target frame, by performing vector offset distance scaling on the displacement path, the target position of each missing pixel can be accurately calculated in the frame interpolation process, thereby ensuring that the frame interpolation content is consistent with the actual physical motion trend, this pixel-level accurate calculation can significantly reduce the motion artifacts, ghosting and blur problems in the conventional frame interpolation algorithm, ensure the restoration degree of dynamic details, and compared with directly improving the refresh rate to meet the scene requirement, this picture optimization strategy through frame interpolation can obtain high refresh rate visual effect under the display mode with low refresh rate, which meets the visual experience requirement of high-speed motion scene, avoids the energy consumption increase caused by frequent switching of display mode or continuous running of high-performance mode, and is suitable for use in power-sensitive scenes such as mobile devices and portable display devices.

[0074] In the embodiment, the acquisition module further includes: The intercepting unit is configured to intercept a frame data structure of the current display content based on a preset frequency of the display screen, where the frame data structure specifically includes a pixel matrix, brightness information, and a color channel. The fifth judging unit is configured to judge whether the frame data structure matches a preset image hot area of the display screen. The fifth executing unit is configured to, if yes, acquire a pixel motion speed of the current display content, and dynamically adjust a refresh cycle of the image hot area according to the pixel motion speed.

[0075] In this embodiment, the system intercepts the frame data structure of the current display content based on the preset frequency of the display screen, which specifically includes the pixel matrix, brightness information, and color channel, and then determines whether these frame data structures match the image hot area preset by the display screen to execute corresponding steps. For example, when the system determines that the frame data structure of the current display content cannot match the image hot area preset by the display screen, it is considered that the pixel distribution, brightness information, or color channel characteristics in the current frame deviate from the predefined key area mode, meaning that the display content is inconsistent with the expected hot area layout, such as hot spot area position offset, size change, or no preset content, the system will perform secondary analysis on the current frame data, reposition the potential high-attention area, and simultaneously find the candidate area close to the preset hot area characteristics through methods such as pixel difference detection, brightness gradient analysis, and color histogram matching, generate its similarity score, and if the similarity of the candidate area is higher than the set threshold, the system will dynamically update the hot area mapping relationship to align the new hot area position with the actual content, and if no region meeting the conditions is found, a warning will be triggered or switched to the default display mode to ensure the stability of display quality and user experience. For example, when the system determines that the frame data structure of the current display content can match the image hot area preset by the display screen, it is considered that the pixel distribution, brightness information, or color channel characteristics in the current frame are consistent with the predefined key area mode, and the system will obtain the pixel motion speed of the current display content and dynamically adjust the refresh cycle of the image hot area according to different pixel motion speeds. When the system determines that the frame data structure of the current display content can match the image hot area preset by the display screen, it means that the key area (such as important visual focus or high dynamic change area) in the current display picture is highly consistent with the predefined hot area layout. This consistency ensures the accurate recognition of the system to the display content, so that the subsequent dynamic refresh strategy based on the hot area can accurately act on the area that needs most attention, thereby improving the overall display efficiency and visual experience. Meanwhile, based on the matched hot area, the system further obtains the motion speed of each pixel in the current display content, dynamically adjusts the refresh cycle of the corresponding hot area by analyzing the size and distribution of the pixel motion speed, specifically, for the area with fast motion speed, the system will shorten the refresh cycle to increase the refresh rate and ensure the smoothness and clarity of the motion picture, while for the area with slow motion or static, the refresh cycle can be appropriately prolonged to reduce unnecessary refresh frequency, thereby saving power consumption and reducing heat generation of the display screen. Moreover, this mechanism of adjusting the hot area refresh cycle based on the pixel motion speed helps to realize "regional intelligent refresh", effectively balances the display performance and energy consumption, and is particularly suitable for application scenarios with rich dynamic content and frequent local changes (such as video playback and game pictures). Through precise hot area refresh control, the high-quality display of key visual areas is ensured, and the energy waste caused by high-frequency refresh of the entire screen is avoided, thereby improving the endurance and user experience of the device.

[0076] While embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A backlight dynamic refresh method based on content change frequency, characterized in that, Includes the following steps: Based on the inter-frame changes of the display screen pre-collected on the currently displayed content, the image content refresh rate of the currently displayed content is obtained; Determine whether the image content refresh rate matches the preset backlight refresh response rate; If not, the image difference data between the currently displayed content and the content displayed in the previous frame is identified, and the pixel change rate per unit time is calculated using a preset frame difference method. Based on the pixel change rate, the content change frequency index of the display screen is dynamically adjusted. Specifically, the content change frequency index includes low dynamic, medium dynamic and high dynamic. Determine whether the content change frequency index matches the user's display settings for the display screen; If not, the PWM frequency and duty cycle of the display backlight are adjusted according to the user's preset mapping strategy for the display screen. The frame rate fluctuation of the display screen is detected. Based on the frame rate fluctuation, when the display backlight changes, a preset grayscale intermediate compensation frame is added to dynamically limit the upper limit of the brightness change slope of the display screen and construct the brightness change gradient of the display screen. The mapping strategy specifically includes a brightness strategy, an image quality strategy, and a perception priority strategy.

2. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, Before the step of identifying the image difference data between the currently displayed content and the content displayed in the previous frame, the method further includes: Based on the preset data buffer of the display screen, image data is extracted from the data buffer, and the preset image frame time interval of the display screen is identified according to the image data. The image data specifically includes the color value, grayscale brightness value and transparency of the pixel. Determine whether the image frame time interval is synchronized with the screen refresh cycle; If so, the display content of the image data is parsed, and the display parameters of the display screen are obtained based on the display content. The frame rate change trend of the image data is constructed through the display parameters. The display content specifically includes static images, video streams, fast-scrolling text, and game screens, and the display parameters specifically include frame rate, motion vector, and distribution of change areas.

3. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, The step of dynamically adjusting the content change frequency index of the display screen based on the pixel change rate further includes: Based on the pre-divided grid area of ​​the display screen, extract the percentage of the number of changing pixels in the grid area; Determine whether the percentage of changed pixels matches the pixel change rate; If so, then based on the proportion of the number of changed pixels, a corresponding local pixel change weight is constructed, the content complexity of the grid region is identified, and the grid scale of the grid region is dynamically adjusted. The content complexity specifically includes dynamic content and static content, and the grid scale specifically includes coarse grid and fine grid.

4. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, The step of dynamically limiting the upper limit of the slope of the brightness change of the display screen and constructing the brightness change gradient of the display screen further includes: Based on the continuous image frames of the display screen, the brightness distribution statistics of the continuous image frames are extracted to generate brightness jump points within a preset time window. Specifically, the brightness distribution statistics include average brightness, maximum brightness, and minimum brightness. Determine whether the brightness jump point exceeds a preset abrupt change threshold; If so, based on the brightness jump point, identify the brightness change region corresponding to the consecutive image frames, apply a preset exponential decay filter to slowly transition the brightness of the brightness change region, detect the adjacent regions of the brightness change region in real time, and dynamically limit the brightness change amplitude between the adjacent regions.

5. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, The step of determining whether the image content refresh rate matches the preset backlight refresh response rate further includes: Based on the continuous image frames of the display screen, the pixel differences between the continuous image frames are extracted, and a dynamic heat map of the display screen is constructed according to the pixel differences; Determine whether the dynamic heat map can be mapped to the physical display screen coordinates of the display screen; If possible, identify the high-frequency irregular jitter of the dynamic heat map, mark the high-frequency irregular jitter as abnormal frames of the display screen, delineate the corresponding region to be reconstructed based on the abnormal frame, and construct candidate frame blocks of the region to be reconstructed through preset pixel texture similarity.

6. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, The step of determining whether the content change frequency index meets the user's display settings for the display screen further includes: Based on the user's preset display mode for the display screen, the scene type of the display screen is identified, wherein the display mode specifically includes power saving mode, standard mode and high performance mode, and the scene type specifically includes reading scene, video scene and game scene; Determine whether the display mode matches the refresh requirements of the scene type; If not, then based on the pixel differences between consecutive image frames, a corresponding pixel motion region is constructed. Through the pixel motion region, a displacement path for each pixel is generated. Based on the displacement path, the motion discontinuity area of ​​the pixel motion region is collected. The corresponding missing pixels are detected from the motion discontinuity area. The displacement path is scaled proportionally by vector offset distance to obtain the target position of each pixel required for frame interpolation. The target position of each pixel is used as the interpolation content between the original frames, and the missing pixels are combined.

7. The backlight dynamic refresh method based on content change frequency according to claim 1, characterized in that, The step of obtaining the image content refresh rate of the currently displayed content based on the inter-frame change amount pre-collected by the display screen on the currently displayed content further includes: Based on the preset frequency of the display screen, the frame data structure of the currently displayed content is extracted, wherein the frame data structure specifically includes a pixel matrix, brightness information and color channels; Determine whether the frame data structure matches the preset image hotspot of the display screen; If so, the pixel motion speed of the currently displayed content is obtained, and the refresh cycle of the image hotspot is dynamically adjusted according to the pixel motion speed.

8. A backlight dynamic refresh system based on content change frequency, characterized in that, include: The acquisition module is used to acquire the image content refresh rate of the currently displayed content based on the inter-frame change amount pre-collected by the display screen. The judgment module is used to determine whether the image content refresh rate matches the preset backlight refresh response rate; The execution module is used to identify the image difference data between the currently displayed content and the content displayed in the previous frame if no, calculate the pixel change rate per unit time using a preset frame difference method, and dynamically adjust the content change frequency index of the display screen according to the pixel change rate. The content change frequency index specifically includes low dynamic, medium dynamic and high dynamic. The second judgment module is used to determine whether the content change frequency index meets the user's display settings for the display screen; The second execution module is used to, if not conforming, adjust the PWM frequency and duty cycle of the display backlight according to the user's preset mapping strategy for the display, detect the frame rate fluctuation of the display, and, through the frame rate fluctuation, add a preset grayscale intermediate compensation frame when the display backlight changes, dynamically limit the upper limit of the brightness change slope of the display, and construct the brightness change gradient of the display. The mapping strategy specifically includes a brightness strategy, an image quality strategy, and a perception priority strategy.

9. The backlight dynamic refresh system based on content change frequency according to claim 8, characterized in that, Also includes: The recognition module is used to extract image data from the data buffer preset by the display screen, and to recognize the preset image frame time interval of the display screen according to the image data, wherein the image data specifically includes the color value, grayscale brightness value and transparency of the pixel; The third judgment module is used to determine whether the image frame time interval is synchronized with the screen refresh cycle; The third execution module is used to parse the display content of the image data if the condition is met, obtain the display parameters of the display screen based on the display content, and construct the frame rate change trend of the image data through the display parameters. The display content specifically includes static images, video streams, fast-scrolling text, and game screens, and the display parameters specifically include frame rate, motion vector, and distribution of changing areas.

10. The backlight dynamic refresh system based on content change frequency according to claim 8, characterized in that, The execution module further includes: The extraction unit is used to extract the percentage of changing pixel counts in a pre-divided grid region of the display screen. The judgment unit is used to determine whether the percentage of the number of changed pixels matches the pixel change rate; An execution unit is configured to, if so, construct a corresponding local pixel change weight based on the proportion of the number of changed pixels, identify the content complexity of the grid region, and dynamically adjust the grid scale of the grid region, wherein the content complexity specifically includes dynamic content and static content, and the grid scale specifically includes coarse grid and fine grid.

Images