Self-adaptive adjusting method and system for content refresh rate of LED display screen

By combining the image processing of the LED display and user monitoring data, the refresh rate is adaptively adjusted, which solves the problem that the refresh rate cannot adapt to different scenarios in traditional methods, and improves the visual experience and resource utilization efficiency.

CN119091795AInactive Publication Date: 2024-12-06VISION JUMP (SHENZHEN) TECH CO LTD
View PDF 0 Cites 10 Cited by

Patent Information

Application Number
CN202411520360.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-29
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional LED display refresh rate control methods cannot adapt to different usage scenarios and user interaction needs, resulting in excessive power consumption during static content, and may cause screen stuttering and flickering when changing content quickly.

Method used

By identifying the LED display image, detailed convolution reconstruction and dynamic area division, dynamic texture optical flow tracking, combining user eye monitoring and ambient light feature data, a dynamic visual frequency optimization model is built, and refresh rate is adaptively adjusted to optimize visual experience .

Benefits of technology

It realizes refresh rate optimization in different scenarios and user needs, reduces visual fatigue and flickering, improves user experience and viewing comfort, and improves display effect and resource utilization efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091795A_ABST
    Figure CN119091795A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of display screen refresh rate control, in particular to a self-adaptive adjusting method and system for the content refresh rate of an LED display screen. The method comprises the following steps: identifying an LED display screen image; detail convolution reconstruction is carried out on the LED display screen image, and dynamic region division is carried out, so that a dynamic region image and a static region image are generated; performing dynamic texture optical flow tracking on the dynamic region image, and performing optical flow change trend evolution so as to generate a dynamic optical flow trend feature map; performing dynamic gradual change smooth optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map, and constructing a dynamic gradual change refresh rate image; identifying a current user eyeball monitoring image based on a display screen camera; and performing user flicker frequency sensitivity analysis on the eyeball monitoring image of the current user to obtain a dynamic refresh rate change amplitude. According to the invention, the refresh rate of the display screen is adjusted adaptively and efficiently according to a real-time environment, and the visual experience of a user is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of display screen refresh rate control, and in particular to a method and system for adaptively adjusting the content refresh rate of an LED display screen. Background Art

[0002] With the rapid development of mobile devices and Internet of Things technology, LED displays have been widely used in various scenarios. In these applications, power management and power efficiency optimization are important issues, especially in battery-powered mobile devices. How to effectively reduce the energy consumption of LED displays becomes the key.

[0003] Traditional LED display refresh rate control methods usually set a fixed refresh rate to meet the visual needs of the human eye. However, this method cannot adapt to the changes in display content and user interaction needs in different usage scenarios. When the display content is relatively static, too high a refresh rate will cause unnecessary power consumption; and when the display content changes rapidly, too low a refresh rate will cause screen freezes and flickers. In order to solve this problem, a method that can adaptively adjust the refresh rate of LED displays is needed. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method and system for adaptively adjusting the content refresh rate of an LED display screen, so as to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a method for adaptively adjusting the content refresh rate of an LED display screen, comprising the following steps: Preferably, step S1 comprises the following steps: Step S1: Identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image, and perform dynamic area division to generate a dynamic area image and a static area image; Step S2: Tracking the dynamic texture optical flow of the dynamic region image and evolving the optical flow change trend to generate a dynamic optical flow trend feature map; Step S3: performing dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map to construct a dynamic gradient refresh rate image; Step S4: identifying the current user eyeball monitoring image based on the display screen camera; performing user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain a dynamic refresh rate change amplitude; Step S5: Acquire ambient light characteristic data of the LED display screen based on the sensor; perform ambient light distribution characteristic fitting on the ambient light characteristic data of the LED display screen to construct a light intensity scene characteristic map; Step S6: Based on the light intensity scene feature map and the dynamic refresh rate variation range, the dynamic visual frequency of the dynamic gradient refresh rate image is optimized to construct a dynamic visual frequency optimization model.

[0006] The present invention extracts important features in LED display screen images through detail convolution reconstruction and dynamic area division, provides high-quality data for subsequent processing, generates dynamic area images and static area images to distinguish static and dynamic elements in display content, and provides a basis for subsequent processing. Dynamic texture optical flow tracking captures the movement of dynamic areas in LED display screens, helps understand the dynamic changes of content, and generates dynamic optical flow trend feature maps that can provide dynamic information on changes in LED display content, providing a basis for subsequent adjustment of refresh rate. Dynamic gradient smoothing optimization improves the transition effect of display content, reduces flickering and discontinuity, and improves user experience. A dynamic gradient refresh rate image is constructed to adjust the refresh rate according to content changes to make the display smoother and more coherent. Through user flicker frequency sensitivity analysis, it is determined that the user is sensitive to the content. Sensitivity to refresh rate changes, adjusting the refresh rate to reduce visual fatigue and flicker, and obtaining the dynamic refresh rate change range can optimize the refresh rate of the LED display according to the user's perception, improve the user experience, and understand the lighting conditions around the LED display through the processing of ambient light feature data, provide a basis for subsequent adjustment of brightness and contrast, and construct a light intensity scene feature map to adapt to changes in display content and ambient light, improve display effect and comfort. Based on the light intensity scene characteristics and the user's sensitivity to the refresh rate, the dynamic visual frequency optimization model can intelligently adjust the refresh rate of the LED display to provide a better visual experience. Through dynamic visual frequency optimization, the refresh rate of the LED display is adjusted according to real-time conditions, reducing eye fatigue, improving viewing comfort, and allowing users to obtain better visual effects.

[0007] Preferably, step S1 comprises the following steps: Step S11: Identify the LED display screen image; Step S12: performing detail convolution reconstruction on the LED display screen image to construct a detail sharpened reconstructed image; Step S13: performing time-series frame segmentation processing on the detail-sharpened reconstructed image, thereby obtaining an image frame sequence; Step S14: performing continuous frame dynamic optical flow recognition on the image frame sequence to generate inter-frame dynamic optical flow data; Step S15: performing dynamic and static texture division according to the inter-frame dynamic optical flow data, thereby generating a dynamic region image and a static region image.

[0008] The present invention can accurately capture the display content by identifying the LED display screen image, provide accurate and complete data for subsequent processing, ensure that the subsequent processing is based on correct input, and improve the accuracy and reliability of the processing. The detail convolution reconstruction can extract tiny details in the display content, such as text, patterns, etc., to enhance the clarity and quality of the image. The reconstructed image accurately captures the subtle changes in the display content and provides richer information for subsequent processing. The time frame segmentation divides the image into a time frame sequence, captures the dynamic changes of the display content, and provides continuity data support. The segmentation process provides an ordered image sequence for subsequent dynamic optical flow recognition, making the analysis of the display content change more accurate and complete. Through continuous frame dynamic optical flow recognition, the system can accurately capture the motion trajectory and speed change of objects in the display content. The generation of dynamic optical flow data deeply analyzes the dynamic characteristics of the display content and provides an important reference for the subsequent adjustment of the refresh rate. The dynamic area division enables the system to clearly distinguish the dynamic part and the static part in the display content, and to process different areas in a targeted manner. The dynamic area image and the static area image are generated to dynamically adjust the content refresh rate of the LED display screen according to the user's visual needs, and provide a visual experience that better meets the user's expectations.

[0009] Preferably, the specific steps of step S12 are: Calculate the average brightness of the LED display screen image and extract the average brightness value of the display screen image; Based on the average brightness value of the display screen image, the dark area of ​​the LED display screen image is identified and the dark area of ​​the image is extracted; Performing local brightness enhancement processing on the dark area of ​​the image, thereby obtaining a brightness enhanced display screen image; Perform image noise detection on the brightness enhanced display screen image and mark multiple image noise points; Performing mean filtering to remove noise points in multiple images to generate a filtered and denoised display screen image; Perform multi-layer convolution decomposition processing on the filtered and denoised display screen image to obtain a multi-layer convolution image; Adaptively sharpen the multi-layer convolution images to obtain multiple detail-sharpened convolution images; The filtered denoised display screen image is deconvolved and reconstructed based on a plurality of detail-sharpened convolution images to construct a detail-sharpened reconstructed image.

[0010] The present invention quantifies the overall brightness level by calculating the average brightness value of the LED display screen image, providing a benchmark reference value for subsequent processing. Through the average brightness value, the overall brightness distribution of the display screen is understood, providing an important basis for subsequent dark area recognition. Dark area recognition can effectively locate the dark area in the LED display screen image, help extract the low-brightness area in the display content, identify the dark area and process these areas in a targeted manner to improve the visual quality and clarity of the overall image. Local brightness enhancement processing can highlight the details of the dark area in the display screen image, improve the visibility and clarity of these areas, enhance the brightness of the dark area to balance the brightness distribution of the overall image, making the display content clearer and easier to observe, detect and mark the noise in the image, accurately locate the existing noise and interference, and provide an accurate target for the subsequent denoising processing. Marking the noise can help the system identify abnormal points in the image , improve the accuracy and efficiency of subsequent processing, mean filtering effectively removes noise in the image, improves the clarity and quality of the image, the image after noise removal is more visible, and the user can observe the display content more clearly, multi-layer convolution decomposition extracts image features from different levels, and provides more information and more comprehensive feature expression for subsequent detail processing, layered processing better captures the detail information of the image, and provides a richer data basis for subsequent processing, adaptive detail sharpening processing effectively highlights the detail features of the image, enhances the clarity and layering of the image, detail sharpening improves the image's viewing and detail expression, and makes the display content more vivid and fascinating, through deconvolution reconstruction processing, reconstructs the image with detail sharpening effect, improves the image's detail expression and visual effect, and detail sharpening reconstructs the image to make the display content clearer and more realistic, and improves the user's visual experience and viewing quality.

[0011] Preferably, step S2 specifically comprises the following steps: Step S21: performing dynamic texture optical flow tracking on the dynamic area image to generate a dynamic texture optical flow trajectory; Step S22: Calculate the texture change frequency of the dynamic texture optical flow trajectory to extract the dynamic texture change frequency; Step S23: performing color type statistics on the dynamic area image to obtain the image color type; Step S24: performing color saturation change analysis based on the image color type to generate image color saturation change data; Step S25: Perform optical flow change trend evolution on the dynamic texture change frequency and image color saturation change data, thereby generating a dynamic optical flow trend feature map.

[0012] The present invention captures the motion information of the dynamic area in the image through dynamic texture optical flow tracking, provides key data for subsequent analysis, generates dynamic texture optical flow trajectory to help accurately describe the dynamic changes in the image, and provides a basis for subsequent processing. Texture change frequency calculation can quantify the change speed of the dynamic texture, help understand the characteristics of the dynamic texture in the image, extract the dynamic texture change frequency to evaluate the frequency of dynamic elements in the image, and provide an important reference for subsequent adjustment. Color type statistics help identify the color composition in the image, provide color information for subsequent analysis, obtain the image color type to understand the color composition of the image, and provide a basis for subsequent color analysis and adjustment. Color saturation change analysis can evaluate the change in the vividness of the image color, provide a basis for adjusting the image color, generate color saturation change data to understand the dynamic change of the image color, and provide guidance for subsequent color adjustment. The optical flow change trend evolution can comprehensively consider the dynamic texture change frequency and color saturation change data, depict the overall change trend of the image, generate a dynamic optical flow trend feature map to intuitively display the dynamic characteristics of the image, provide a basis for the adaptive adjustment of the refresh rate of the LED display content, and improve the user's visual experience and viewing quality.

[0013] Preferably, step S3 specifically comprises the following steps: Step S31: measuring the image refresh rate of the image frame sequence and extracting the current image dynamic refresh rate; Step S32: dynamically adjusting the dynamic refresh rate of the current image based on the dynamic optical flow trend feature map to obtain a dynamic optical flow adjusted refresh rate; Step S33: dynamically adjusting the refresh rate of the dynamic area image based on the dynamic optical flow to adjust the refresh rate, and constructing a dynamic refresh rate adjusted area image; Step S34: performing edge arc recognition on the dynamic region image and the static region image to extract region edge arcs; Step S35: performing a gradual frame rate interpolation analysis on the region edge arc, thereby obtaining a gradual frame rate interpolation transition zone; Step S36: using a gradual frame rate interpolation transition band to perform dynamic gradual smoothing optimization on the dynamic refresh rate adjustment area image and the static area image to construct a dynamic gradual refresh rate image.

[0014] The present invention can accurately evaluate the dynamic refresh frequency of the image through image refresh rate measurement, provide basic data for subsequent adjustment, extract the current image dynamic refresh rate to understand the refresh speed of the current display content, and provide a target value for subsequent adjustment. The dynamic optical flow trend feature map provides important information about image changes and can be used to dynamically adjust the refresh rate according to the image content. The dynamic refresh rate adjustment can be intelligently adjusted according to the dynamic nature of the image content, improve the display effect and reduce visual fatigue. The dynamic refresh rate adjustment can adjust the refresh rate according to the characteristics of the dynamic area image, improve the clarity and smoothness of the dynamic area, construct a dynamic refresh rate adjustment area image to highlight the dynamic elements, and enhance the user's attention to the dynamic content. , edge arc recognition extracts edge features of image areas, providing key clues for subsequent processing. Extracting regional edge arcs enhances the contours and details of the image, improving the visual experience and viewing experience. Gradient frame rate interpolation analysis can smooth the frame rate changes in the transition area and enhance the naturalness of the image transition effect. The gradient frame rate interpolation transition zone provides smooth animation effects and reduces visual discomfort during switching. Dynamic gradient smoothing optimization can combine frame rate interpolation technology to make image transitions smoother and more natural, improve visual experience, and construct dynamic gradient refresh rate images to make changes in displayed content more coherent, providing better visual effects for the adaptive adjustment of the content refresh rate of the LED display screen, and enhancing the user's viewing experience.

[0015] Preferably, the specific steps of step S4 are: Step S41: recognizing the current user's eye monitoring image based on the display screen camera; Step S42: performing user sight tracking on the current user eye monitoring image to extract the user sight track; Step S43: analyzing the gaze lingering area of ​​the user's gaze trajectory to generate a gaze lingering area; Step S44: Calculate the duration of the sight lingering area to obtain the duration of the user's sight lingering area; Step S45: Perform a user flicker frequency sensitivity analysis based on the duration of the user's gaze, so as to obtain the dynamic refresh rate change amplitude.

[0016] The present invention obtains the user's sight information in real time by identifying the current user's eyeball monitoring image, providing data support for subsequent analysis and adjustment, and obtains the user's visual image through the display screen camera, thereby realizing real-time monitoring and feedback of the user's behavior, providing a basis for personalized experience, user sight tracking to understand the user's focus and behavioral habits on the display screen content, providing a basis for content presentation and optimization, extracting the user's sight trajectory to reveal the user's attention focus area during the viewing process, and providing important data for subsequent analysis of user behavior, sight stay area analysis to determine the hot spot area where the user stays on the display screen, providing guidance for content adjustment and optimization, generating sight stay area to help the system better understand the user's interest points and habits, and providing personalized support for content recommendation and display, and the stay time calculation to quantify the user's attention to different areas, providing data support for content optimization and personalized recommendation, and obtaining the user's sight stay time to understand the user's reading speed and attention distribution, providing a basis for the improvement of content design and display methods, and user flicker frequency sensitivity analysis to adjust the refresh rate of the display screen according to the user's visual habits and reaction speed, improve user experience and comfort, and obtain the dynamic refresh rate change range to make the refresh of the displayed content more in line with the user's visual needs, effectively reduce visual fatigue and enhance the clarity of the content, and improve user satisfaction.

[0017] Preferably, the specific steps of step S45 are: Performing multi-time-point blink frequency calculation on the current user eyeball monitoring image to generate user blink frequency distribution data; Perform visual stay matching on the area where the sight stays according to the user's blink frequency distribution data, and extract the blink frequency of the area where the user's vision stays; Perform flicker frequency sensitivity analysis on the blink frequency of the user's visual stay area and the user's gaze stay duration to extract the user's flicker frequency sensitivity; Predicting image refresh rate adjustment constraints based on user flicker frequency sensitivity to generate image refresh rate adjustment constraint data; The refresh rate change range of the current image dynamic refresh rate is calculated according to the image refresh rate adjustment constraint data, thereby obtaining the dynamic refresh rate change range.

[0018] The present invention helps to understand the frequency distribution of user blinking through multi-time point blinking frequency calculation, provides data support for subsequent behavior analysis and adjustment, generates user blinking frequency distribution data to understand the user's blinking habits and frequency changes, and provides a basis for personalized experience and adjustment. Visual stay matching matches the user's blinking frequency with the sight stay area to understand the user's attention concentration in different areas, extracts the blinking frequency of the user's visual stay area to link the blinking behavior with the user's attention area, and provides a data basis for subsequent analysis of user behavior and experience. Flicker frequency sensitivity analysis combines blinking frequency and stay duration to identify the user's sensitivity to flicker frequency in different areas, provides a reference for adjusting the refresh rate, and provides a reference for adjusting the refresh rate. The user's flicker frequency sensitivity is taken to understand the user's adaptability to different refresh rates, and guidance is provided for personalized refresh rate adjustment and optimization. The image refresh rate adjustment constraint prediction is based on the user's flicker frequency sensitivity to predict the appropriate refresh rate range to ensure user viewing comfort. The image refresh rate adjustment constraint data is generated to set the upper and lower limits of the refresh rate to ensure the balance between the smoothness of the displayed content and the user experience. The refresh rate change amplitude calculation is based on the constraint data, and can intelligently adjust the refresh rate of the current image to make it more in line with the user's visual needs and comfort. The dynamic refresh rate change amplitude is obtained to optimize the refresh effect of the displayed content, improve the user's viewing experience and comfort, and realize the adaptive adjustment of the content refresh rate of the LED display.

[0019] Preferably, the specific steps of step S5 are: Step S51: extracting the optimal adjustable refresh rate variation range based on the visual performance evaluation value of each variation range; Step S52: Calculate the light intensity of the ambient light characteristic data of the LED display screen to generate an ambient light intensity value; Step S53: performing light intensity spatial distribution analysis on the ambient light intensity value to generate ambient light intensity spatial distribution data; Step S54: performing light intensity uniformity evolution on the ambient light intensity spatial distribution data to extract ambient light intensity uniformity features; Step S55: performing light intensity fluctuation identification on the ambient light intensity value to generate light intensity fluctuation data; Step S56: fitting the ambient light distribution characteristics to the ambient light intensity uniformity characteristics and the light intensity fluctuation data to construct a light intensity scene characteristic map.

[0020] The present invention can determine the best refresh rate adjustment range according to the visual performance evaluation value by extracting the optimal adjustable refresh rate variation range, thereby improving the user's visual experience and comfort, optimizing the refresh rate variation range to make the refresh of the LED display content smoother and adapt to the user's needs, and improving the clarity and coherence of the content, calculating the ambient light intensity value to understand the lighting conditions of the environment in which the LED display is located, providing data support for subsequent ambient light adaptive adjustment, generating ambient light intensity values ​​to adjust the brightness and contrast of the displayed content according to the ambient light intensity, and improving the user's viewing experience in different lighting environments, and the light intensity spatial distribution analysis reveals the spatial characteristics of the ambient light around the LED display, providing basic data for ambient light adaptive adjustment, generating ambient light intensity spatial distribution data to adjust the brightness and color of the displayed content according to the lighting conditions at different locations, and improving the user experience. Consistency, light intensity uniformity evolution evaluates the uniformity of ambient light around the LED display screen, provides a reference for adjusting the brightness and contrast of the displayed content, extracts the uniformity characteristics of ambient light intensity to ensure the balanced brightness distribution of the displayed content in different areas, and improves the viewing comfort and clarity of users. Light intensity fluctuation recognition discovers the fluctuation of ambient light around the LED display screen, and provides a basis for dynamically adjusting the brightness and contrast of the displayed content. It generates light intensity fluctuation data to adjust the displayed content according to the changes in ambient light, maintains the stability and continuity of the picture, and improves the user's viewing experience. Ambient light distribution feature fitting combines light intensity uniformity and fluctuation data to construct a feature map of the ambient light around the LED display screen, which provides guidance for dynamically adjusting the displayed content. It constructs a light intensity scene feature map to adjust the brightness and contrast of the displayed content according to different ambient light conditions, and improves the stability and comfort of the user's viewing experience.

[0021] Preferably, the specific steps of step S6 are: Step S61: performing scene adaptive frequency adjustment on the dynamic gradient refresh rate image based on the light intensity scene feature map to generate a dynamic frequency adaptive image; Step S62: dividing a plurality of adjustable refresh rate change ranges according to the dynamic refresh rate change range; Step S63: performing iterative adjustment simulation on the dynamic frequency adaptive image based on multiple adjustable refresh rate change ranges to generate multiple change range simulated images; Step S64: performing user visual performance evaluation on a plurality of variation amplitude simulated images to obtain a visual performance evaluation value of each variation amplitude; Step S65: extracting the optimal adjustable refresh rate change range based on the visual performance evaluation value of each change range; Step S66: Utilize the optimal adjustable refresh rate variation range to perform dynamic visual frequency optimization on the dynamic frequency adaptive image, and construct a dynamic visual frequency optimization model.

[0022] The present invention can dynamically adjust the refresh rate according to the light intensity scene characteristic map through scene adaptive frequency adjustment, so that the displayed content maintains clarity and stability in different environments, generates dynamic frequency adaptive images, can optimize the refresh rate according to the ambient light conditions, improve the comfort and visual effect of the user's viewing experience, divide a plurality of adjustable refresh rate variation amplitudes, can optimize the refresh rate adjustment range of the LED display screen according to the dynamic refresh rate change requirements, improve the display effect, the division of the adjustable refresh rate variation amplitude meets the user's requirements for refresh rate changes in different scenarios, improves the adaptability and flexibility of the LED display screen, iterative adjustment simulation can optimize the image according to a plurality of adjustable refresh rate variation amplitudes, improves the continuity and stability of the displayed content, generates a plurality of variation amplitude simulation images to evaluate the influence of different refresh rate changes on the user's visual experience, provides a reference for the final optimization, and the user's visual Visual performance evaluation quantifies the visual effects of simulated images with different change amplitudes, providing an objective basis for selecting the optimal refresh rate change amplitude. The visual performance evaluation value of each change amplitude is obtained to determine the optimal refresh rate change amplitude to improve the user's visual experience and comfort. Extracting the optimal adjustable refresh rate change amplitude can determine the optimal refresh rate adjustment range according to the user's visual performance evaluation value, thereby improving the user's visual experience and comfort. The extraction of the optimal adjustable refresh rate change amplitude optimizes the refresh frequency of the displayed content to make it more in line with the user's visual needs and habits. The dynamic visual frequency optimization model can adjust the refresh frequency of the displayed content according to the optimal adjustable refresh rate change amplitude, thereby improving the consistency and comfort of the user's viewing experience. The construction of a dynamic visual frequency optimization model can realize intelligent adjustment of the refresh rate according to user needs and ambient light conditions, thereby improving the visual performance and user experience of the LED display.

[0023] In this specification, a system for adaptively adjusting the content refresh rate of an LED display screen is provided, which is used to execute the method for adaptively adjusting the content refresh rate of an LED display screen as described above, including: The area division module is used to identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image and perform dynamic area division to generate dynamic area images and static area images; The optical flow trend module is used to track the dynamic texture optical flow of dynamic area images and evolve the optical flow change trend to generate a dynamic optical flow trend feature map; The dynamic gradient module is used to perform dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map, and construct a dynamic gradient refresh rate image; A visual frequency sensitive module is used to identify the current user eyeball monitoring image based on the display screen camera; perform user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain the dynamic refresh rate change amplitude; The ambient light feature module is used to obtain the ambient light feature data of the LED display screen based on the sensor; perform ambient light distribution feature fitting on the ambient light feature data of the LED display screen to construct a light intensity scene feature map; The dynamic visual frequency optimization module is used to optimize the dynamic visual frequency of the dynamic gradient refresh rate image based on the light intensity scene feature map and the dynamic refresh rate change amplitude, and to build a dynamic visual frequency optimization model.

[0024] The present invention identifies the LED display screen image and performs detail convolution reconstruction to extract the subtle features of the displayed content, providing high-quality input data for subsequent processing. The dynamic area division can divide the image into a dynamic area and a static area according to the changes in the displayed content, providing differentiation for subsequent processing and optimizing the processing effect. The dynamic texture optical flow tracking captures the trend and change of the movement of objects in the image, providing dynamic information for subsequent processing. The generation of a dynamic optical flow trend feature map can reflect the change of the optical flow in the image, providing important information about the dynamics of the image for subsequent processing. The dynamic gradient smoothing optimization can perform gradient processing on the image according to the dynamic optical flow trend feature map, improve the continuity and smoothness of the displayed content, construct a dynamic gradient refresh rate image, adjust the refresh rate according to the dynamics of the image content, improve the comfort and visual effect of the user's viewing experience, and analyze the user's flicker frequency. Rate sensitivity can determine the dynamic refresh rate change range according to the user's perception of the refresh rate change, provide a personalized visual experience, and obtain the dynamic refresh rate change range to adjust the refresh frequency of the displayed content according to the user's visual needs to improve the user's viewing comfort. The fitted light intensity scene feature map can construct an ambient light distribution map according to the characteristics of the ambient light around the LED display screen, provide ambient light information for subsequent adjustments, and obtain ambient light feature data to optimize the displayed content according to the lighting conditions to improve the display effect and viewing experience. Dynamic visual frequency optimization based on the light intensity scene feature map and the dynamic refresh rate change range can adjust the refresh rate according to ambient light conditions and user needs, provide a personalized visual experience, and construct a dynamic visual frequency optimization model to comprehensively adjust the refresh rate of the LED display screen according to different factors, so that users can get a more comfortable and high-quality visual experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of the steps of a method for adaptively adjusting the content refresh rate of an LED display screen according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0026] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0027] The present application example provides a method and system for adaptively adjusting the content refresh rate of an LED display screen. The execution subject of the method and system for adaptively adjusting the content refresh rate of an LED display screen includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as a general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0028] See also Figures 1 to 4 The present invention provides a method for adaptively adjusting the content refresh rate of an LED display screen, and the method for adaptively adjusting the content refresh rate of an LED display screen comprises the following steps: Step S1: Identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image, and perform dynamic area division to generate a dynamic area image and a static area image; Step S2: Tracking the dynamic texture optical flow of the dynamic region image and evolving the optical flow change trend to generate a dynamic optical flow trend feature map; Step S3: performing dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map to construct a dynamic gradient refresh rate image; Step S4: identifying the current user eyeball monitoring image based on the display screen camera; performing user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain a dynamic refresh rate change amplitude; Step S5: Acquire ambient light characteristic data of the LED display screen based on the sensor; perform ambient light distribution characteristic fitting on the ambient light characteristic data of the LED display screen to construct a light intensity scene characteristic map; Step S6: Based on the light intensity scene feature map and the dynamic refresh rate variation range, the dynamic visual frequency of the dynamic gradient refresh rate image is optimized to construct a dynamic visual frequency optimization model.

[0029] The present invention extracts important features in LED display screen images through detail convolution reconstruction and dynamic area division, provides high-quality data for subsequent processing, generates dynamic area images and static area images to distinguish static and dynamic elements in display content, and provides a basis for subsequent processing. Dynamic texture optical flow tracking captures the movement of dynamic areas in LED display screens, helps understand the dynamic changes of content, and generates dynamic optical flow trend feature maps that can provide dynamic information on changes in LED display content, providing a basis for subsequent adjustment of refresh rate. Dynamic gradient smoothing optimization improves the transition effect of display content, reduces flickering and discontinuity, and improves user experience. A dynamic gradient refresh rate image is constructed to adjust the refresh rate according to content changes to make the display smoother and more coherent. Through user flicker frequency sensitivity analysis, it is determined that the user is sensitive to the content. Sensitivity to refresh rate changes, adjusting the refresh rate to reduce visual fatigue and flicker, and obtaining the dynamic refresh rate change range can optimize the refresh rate of the LED display according to the user's perception, improve the user experience, and understand the lighting conditions around the LED display through the processing of ambient light feature data, provide a basis for subsequent adjustment of brightness and contrast, and construct a light intensity scene feature map to adapt to changes in display content and ambient light, improve display effect and comfort. Based on the light intensity scene characteristics and the user's sensitivity to the refresh rate, the dynamic visual frequency optimization model can intelligently adjust the refresh rate of the LED display to provide a better visual experience. Through dynamic visual frequency optimization, the refresh rate of the LED display is adjusted according to real-time conditions, reducing eye fatigue, improving viewing comfort, and allowing users to obtain better visual effects.

[0030] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for adaptively adjusting the content refresh rate of an LED display screen according to the present invention. In this example, the steps of the method for adaptively adjusting the content refresh rate of an LED display screen include: Step S1: Identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image, and perform dynamic area division to generate a dynamic area image and a static area image; In this example, a high-resolution camera (e.g., 1080p or higher) is used to capture the LED display image in real time. Ensure that the device has autofocus and good lighting conditions. Preprocess the captured image, including: applying a Gaussian blur with a convolution kernel parameter set to (5, 5). Convert the image to grayscale to simplify processing.

[0031] import cv2; image = cv2.imread('led_display.jpg'); gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY); denoised_image = cv2.GaussianBlur(gray_image, (5, 5), 0); Use convolutional neural networks (CNNs) for image recognition. Train the model to recognize specific characters or graphics, using Keras or PyTorch frameworks. Example parameters: Input image size: 128x128, Convolution layer settings: 32 3x3 convolution kernels, activation function is ReLU, build a convolutional neural network (CNN) model specifically for detail reconstruction. The model contains multiple convolutional layers and upsampling layers.

[0032] Network structure example: Input layer: 128x128 image, Convolutional layer: 64 3x3 convolutional layers, activation function is ReLU, upsampling layer: 2x2 upsampling layer, output layer: 128x128 reconstructed image, use high-quality LED display image dataset for model training, the goal is to improve image detail and clarity. Input the identified LED display image into the trained convolutional model to generate a detailed sharpened reconstructed image. Use optical flow or background subtraction to identify dynamic areas. Use background subtractors in OpenCV (such as MOG2) to achieve this.

[0033] bg_subtractor = cv2.createBackgroundSubtractorMOG2(); fg_mask = bg_subtractor.apply(reconstructed_image); Threshold the foreground mask to extract the dynamic region. Set the threshold parameter to 150. Use contour detection methods (such as cv2.findContours) to extract the dynamic and static regions. Set the minimum contour area to 500 pixels. Save the dynamic and static region images to files or in memory for further processing.

[0034] Step S2: Tracking the dynamic texture optical flow of the dynamic region image and evolving the optical flow change trend to generate a dynamic optical flow trend feature map; In this embodiment, the dynamic area images obtained from step S1 are used as input. These images should be processed foreground masks containing dynamic content. Image resolution: 1920x1080, frame rate: 30 fps, select a suitable optical flow algorithm for dynamic texture tracking. Commonly used algorithms include: Lucas-Kanade method: suitable for small-scale motion, fast speed. Farneback method: suitable for large-scale motion, providing dense optical flow. Use OpenCV to implement optical flow calculation. The following is a sample code using the Lucas-Kanade method: import cv2; import numpy as np; # Read consecutive frames; prev_frame = cv2.imread('dynamic_frame1.jpg'); next_frame = cv2.imread('dynamic_frame2.jpg'); # Convert the frame to grayscale; prev_gray = cv2.cvtColor(prev_frame, cv2.COLOR_BGR2GRAY); next_gray = cv2.cvtColor(next_frame, cv2.COLOR_BGR2GRAY); # Calculate optical flow; flow = cv2.calcOpticalFlowFarneback(prev_gray, next_gray, None, 0.5,3, 15, 3, 5, 1.2, 0); Decompose the optical flow vector into speed and direction for subsequent analysis. Calculate the magnitude and direction of the optical flow.

[0035] flow_magnitude, flow_angle = cv2.cartToPolar(flow[..., 0], flow[...,1]); By calculating the optical flow changes between consecutive frames, the trend of optical flow changes is extracted. Use moving average or standard deviation to smooth the optical flow data. Window size: 5 frames (for moving average); Generate a dynamic optical flow trend feature map based on the trend of optical flow changes. This is achieved by visualizing the optical flow amplitude, usually represented by a heat map.

[0036] import matplotlib.pyplot as plt; plt.imshow(smoothed_magnitude.reshape(prev_gray.shape), cmap='hot'); plt.colorbar(); plt.title('Dynamic Optical Flow Trend Feature Map'); plt.savefig('dynamic_flow_trend_map.jpg'); The generated dynamic optical flow trend feature map is stored as a file to facilitate subsequent processing and analysis.

[0037] Step S3: performing dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map to construct a dynamic gradient refresh rate image; In this embodiment, according to the dynamic optical flow trend feature map, the image is divided into two parts: a dynamic area and a static area. The dynamic area refers to an area with a large optical flow change trend, and the static area refers to an area with a small optical flow change. A threshold value of an optical flow trend feature value is set, and pixels above the threshold belong to the dynamic area, and pixels below the threshold belong to the static area. For the dynamic area, the refresh rate is optimized by a gradual smoothing method. Specifically, a bilateral filter is used to calculate the gradual smoothing refresh rate of the pixel according to the optical flow trend feature value of the pixel and the feature value of the neighboring pixels. The larger the optical flow feature value, the more violent the movement of the area, and the higher the refresh rate is given; conversely, the smaller the optical flow feature value, the lower the refresh rate is given. This makes the dynamic area have a higher refresh rate, while the static area has a lower refresh rate, thereby saving system resources. For the static area, a uniform low refresh rate is used. Because these areas change less, there is no need to refresh frequently, and only a low refresh frequency needs to be maintained. The following parameters are set: dynamic area optical flow trend eigenvalue threshold: 0.5, dynamic area bilateral filter window size: 9x9, dynamic area maximum refresh rate: 120Hz, static area refresh rate: 30Hz. Through such gradient smoothing optimization, a dynamic gradient refresh rate image is constructed. The dynamic area has a higher refresh rate and the static area has a lower refresh rate, which greatly improves the resource utilization efficiency of the system and provides better visual input for subsequent charging control.

[0038] Step S4: identifying the current user eyeball monitoring image based on the display screen camera; performing user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain a dynamic refresh rate change amplitude; In this embodiment, it is necessary to use the camera on the shared charging cabinet to collect the user's visual image in real time. A camera with a resolution of 1920x1080 and a frame rate of 30fps is used here. The collected visual image is processed to extract the user's eye area features. Using the method of face detection and key point positioning, the user's eye area is first located, and then the eye area is further analyzed. By analyzing the continuous image sequence of the eye area, the flicker frequency of the user's eyes is detected. Here, the optical flow method is used to observe the brightness changes in the eye area to obtain the user's flicker frequency. With the user's flicker frequency information, the user's perception threshold for different refresh rates is calculated based on this. For example, assuming that the user's flicker frequency is between 10-15Hz, it is inferred that the user will have a significant flickering feeling for a refresh rate below 60Hz. Therefore, it is necessary to control the refresh rate of the charging cabinet to above 60Hz to meet the user's visual experience needs. Camera resolution: 1920x1080, camera frame rate: 30fps, eye region size: 120x80 pixels, flicker frequency detection threshold: 10-15Hz, minimum refresh rate requirement: 60Hz. Through such user visual analysis, the change range of the dynamic refresh rate is obtained, which provides an important human-computer interaction basis for subsequent charging control and ensures that users have a comfortable visual experience when using the charging cabinet.

[0039] Step S5: Acquire ambient light characteristic data of the LED display screen based on the sensor; perform ambient light distribution characteristic fitting on the ambient light characteristic data of the LED display screen to construct a light intensity scene characteristic map; In this embodiment, a high-precision ambient light sensor (such as TSL2561 or BH1750) is used to measure the ambient light intensity in front of the LED display. These sensors can capture light intensity data in real time and output it in Lux. Parameter example: sensor sensitivity: 0.01 Lux, measurement range: 0.1 Lux to 64000 Lux, install the sensor in front of the LED display to ensure that it can accurately measure the ambient light. Set periodic data collection, for example, collect light intensity data every 500 milliseconds. Use a microcontroller such as Arduino or Raspberry Pi for data collection and processing.

[0040] import time; import smbus; # Initialize the sensor; bus = smbus.SMBus(1); sensor_address = 0x39 # I2C address of TSL2561; def read_light(): # Read the light intensity value; data = bus.read_i2c_block_data(sensor_address, 0x8C, 2); return (data[1]<<8) | data[0]# Return the light intensity value; light_data = []; for _ in range(100):# Collect data 100 times; light_data.append(read_light()); time.sleep(0.5); Process the collected ambient light data, including denoising and normalization. Use median filtering or moving average method to smooth the data. Fit the smoothed data to a suitable model. Common models include Gaussian distribution or polynomial fitting. Use SciPy's curve_fit function for fitting. Generate a light intensity scene feature map based on the fitted parameters. Use Matplotlib library for visualization.

[0041] import matplotlib.pyplot as plt; plt.figure(figsize=(10, 6)); plt.plot(x_data, smoothed_data, label='Smoothed Data'); plt.plot(x_data, gaussian(x_data, *popt), label='Fitted Gaussian',linestyle='--'); plt.title('Light Intensity Distribution'); plt.xlabel('Sample Index'); plt.ylabel('Light Intensity (Lux)'); plt.legend(); plt.savefig('light_intensity_distribution.png').

[0042] Step S6: Based on the light intensity scene feature map and the dynamic refresh rate variation range, the dynamic visual frequency of the dynamic gradient refresh rate image is optimized to construct a dynamic visual frequency optimization model.

[0043] In this embodiment, a dynamic visual frequency optimization model is set according to the light intensity scene characteristic map and the dynamic refresh rate variation range. The model will adjust the refresh rate of the display screen in real time according to the changes in ambient light and dynamic content to improve visual comfort and energy saving. Calculate the relationship between the light intensity scene characteristic map and the dynamic refresh rate. Use linear regression or polynomial regression methods to build a model to find out the impact of light intensity changes on the refresh rate. Calculate the optimal dynamic refresh rate under different light intensity conditions according to the model. Use the fitted model to predict the optimal refresh rate at a given light intensity value. Adjust the actual display frequency of the dynamic gradient refresh rate image according to the ambient light intensity and the dynamic refresh rate variation range. Use interpolation methods to smooth the transition to avoid visual discomfort caused by sudden changes.

[0044] def interpolate_refresh_rate(current_rate, target_rate, alpha=0.1): return current_rate + alpha * (target_rate - current_rate); current_refresh_rate = 30 # initial value; for target_rate in optimal_refresh_rates: current_refresh_rate = interpolate_refresh_rate(current_refresh_rate,target_rate); # Update the refresh rate of the display; update_display_refresh_rate(current_refresh_rate); The final dynamic visual frequency optimization model parameters are saved for subsequent use and adjustment.

[0045] Based on the real-time collected ambient light data, the refresh rate of the LED display is dynamically adjusted to ensure the best visual experience under different ambient light conditions.

[0046] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Identify the LED display screen image; Step S12: performing detail convolution reconstruction on the LED display screen image to construct a detail sharpened reconstructed image; Step S13: performing time-series frame segmentation processing on the detail-sharpened reconstructed image, thereby obtaining an image frame sequence; Step S14: performing continuous frame dynamic optical flow recognition on the image frame sequence to generate inter-frame dynamic optical flow data; Step S15: performing dynamic and static texture division according to the inter-frame dynamic optical flow data, thereby generating a dynamic region image and a static region image.

[0047] In this embodiment, a high-resolution camera is used to capture images of the LED display screen in real time to ensure that the images are clear and have appropriate lighting conditions. The captured images are preprocessed, including denoising, color correction and image enhancement. Image processing algorithms (such as Gaussian blur and histogram equalization) are used to improve image quality. Convolutional neural networks (CNNs) or other machine learning models are applied to extract features from the images and identify display content (such as charging status, remaining power, etc.). The model is trained to distinguish different display information and provide location information. A high-pass filter (such as a Laplacian operator) is used to perform a convolution operation on the identified image to extract detail features and enhance edges and textures in the image. The extracted details are synthesized with the original image. A weighted sum method is used to construct a detail-sharpened reconstructed image so that the final image maintains the original information while enhancing the visibility of the details. The detail-sharpened reconstructed image is saved as an intermediate result for subsequent processing. Image frames are extracted from the real-time video stream to ensure that each frame is arranged in chronological order. The detail-sharpened reconstructed image is divided into multiple image frames at a specified time interval. It is set to extract 30 frames per second. The rules of the frames are used to analyze the temporal changes. The image frame sequences obtained by segmentation are stored in a structured format (such as an array or a list) to prepare for subsequent analysis. The optical flow algorithm (such as the Lucas-Kanade method or the Horn-Schunck method) is used to analyze the continuous image frames, calculate the motion vector of each pixel, identify the direction and speed of the object's movement, extract dynamic features from the calculated optical flow data, identify the intensity and direction of movement, and provide data support for the subsequent dynamic area division. The clustering algorithm (such as K-means or MeanShift) is used to analyze the optical flow data to identify the dynamic area and the static area. The dynamic area is identified according to the rate and direction of change of the optical flow data. Usually, the optical flow intensity of the dynamic area is higher. The appropriate threshold is set to filter out unimportant dynamic information to ensure that only significant dynamic areas are retained. According to the stability of the optical flow data, the static area is identified. The optical flow changes in these areas are small. The images of the dynamic area and the static area are saved or marked for subsequent use or analysis. In this embodiment, the specific steps of step S12 are: Calculate the average brightness of the LED display screen image and extract the average brightness value of the display screen image; Based on the average brightness value of the display screen image, the dark area of ​​the LED display screen image is identified and the dark area of ​​the image is extracted; Performing local brightness enhancement processing on the dark area of ​​the image, thereby obtaining a brightness enhanced display screen image; Perform image noise detection on the brightness enhanced display screen image and mark multiple image noise points; Performing mean filtering to remove noise points in multiple images to generate a filtered and denoised display screen image; Perform multi-layer convolution decomposition processing on the filtered and denoised display screen image to obtain a multi-layer convolution image; Adaptively sharpen the multi-layer convolution images to obtain multiple detail-sharpened convolution images; The filtered denoised display screen image is deconvolved and reconstructed based on a plurality of detail-sharpened convolution images to construct a detail-sharpened reconstructed image.

[0048] In this embodiment, an image processing library (such as OpenCV or PIL) is used to load an LED display screen image, and the loaded image is converted into a grayscale mode to simplify the brightness calculation. Each pixel of the grayscale image is traversed, the brightness value of each pixel is calculated, and the average value of all pixels is obtained. A threshold is set according to the calculated average brightness value, which is usually a certain proportion of the average brightness (such as 70%). The grayscale image is traversed, pixels below the threshold are identified and marked as dark areas, and a binary image is generated based on the marked dark areas. The regional information of the dark areas is extracted, and a local contrast enhancement algorithm (such as CLAHE, local histogram equalization) is used to process the dark areas of the image. The brightness of the extracted dark areas is enhanced to make the display effect of these areas better. The enhanced area is synthesized with the original image to generate a brightness enhanced display screen image, and the characteristics of the noise (such as brightness abnormality, noise size) are set. Generally, the local brightness is significantly higher than the surrounding pixels. The convolution kernel is used to slide on the brightness enhanced display screen image to detect potential noise points, mark the detected noise points, and generate a marked image for subsequent processing. The image is processed using a mean filter or a Gaussian filter algorithm to smooth the noise points. The mean filter is applied to the marked noise point area, the mean of the surrounding pixels is calculated and the original noise point pixel value is replaced. The display screen image after filtering is output to reduce noise interference. A multi-layer convolution kernel (such as Gabor) is designed. Filters, Laplace operators, etc.) are used to extract features of different scales, multi-layer convolution is applied to the filtered denoising display screen image to generate a multi-layer convolution image, the result of each layer of convolution is saved for subsequent processing, an adaptive detail sharpening algorithm is selected, and convolution images of different levels are processed by adjusting the enhancement factor, and the multi-layer convolution images are sharpened in detail to highlight important features and edges in the image, and multiple convolution images after detail sharpening are output. According to the detail sharpened convolution image, a deconvolution algorithm is designed to restore a higher quality image, the detail sharpened convolution image is input into the deconvolution model, and a detail sharpened image is reconstructed to generate a final detail sharpened reconstructed image for subsequent analysis and display.

[0049] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing dynamic texture optical flow tracking on the dynamic area image to generate a dynamic texture optical flow trajectory; Step S22: Calculate the texture change frequency of the dynamic texture optical flow trajectory to extract the dynamic texture change frequency; Step S23: performing color type statistics on the dynamic area image to obtain the image color type; Step S24: performing color saturation change analysis based on the image color type to generate image color saturation change data; Step S25: Perform optical flow change trend evolution on the dynamic texture change frequency and image color saturation change data, thereby generating a dynamic optical flow trend feature map.

[0050] In this embodiment, an optical flow tracking algorithm (such as the Lucas-Kanade method or Dense OpticalFlow) is used to track texture changes in dynamic areas, the dynamic area image is processed, the texture area with obvious movement is extracted, and a dynamic texture optical flow trajectory is generated. The dynamic texture optical flow trajectory is analyzed, and the change frequency of each trajectory is calculated. The frequency of texture change is determined by counting the number of moving points in each time period. For example, the number of optical flow trajectories passing through a fixed point per unit time is calculated, and the calculated dynamic texture change frequency is organized into a data table including a timestamp and a corresponding frequency value. A color space conversion (such as RGB to HSV) is applied to the dynamic area image to better analyze color features, the number of pixels of different colors in the image is counted, and the color types are identified and classified. K-means clustering is used to identify the main colors. The color types and their corresponding pixel numbers are saved as structured data for subsequent analysis. The saturation channel in the color space extracts the color saturation value, compares the color saturation at different time points, calculates the rate of change of saturation, organizes the calculated color saturation change data into a table containing the timestamp and the corresponding saturation change rate, integrates the dynamic texture change frequency and the color saturation change data into the same data set, uses time series analysis methods (such as sliding average, linear regression) to analyze the integrated data, identifies the trend of optical flow changes, visualizes the analysis results, and generates a dynamic optical flow trend feature map to highlight the relationship between dynamic texture changes and color saturation changes.

[0051] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: measuring the image refresh rate of the image frame sequence and extracting the current image dynamic refresh rate; Step S32: dynamically adjusting the dynamic refresh rate of the current image based on the dynamic optical flow trend feature map to obtain a dynamic optical flow adjusted refresh rate; Step S33: dynamically adjusting the refresh rate of the dynamic area image based on the dynamic optical flow to adjust the refresh rate, and constructing a dynamic refresh rate adjusted area image; Step S34: performing edge arc recognition on the dynamic region image and the static region image to extract region edge arcs; Step S35: performing a gradual frame rate interpolation analysis on the region edge arc, thereby obtaining a gradual frame rate interpolation transition zone; Step S36: using a gradual frame rate interpolation transition band to perform dynamic gradual smoothing optimization on the dynamic refresh rate adjustment area image and the static area image to construct a dynamic gradual refresh rate image.

[0052] In this embodiment, continuous image frames are obtained from the image frame sequence to ensure that each frame has an accurate timestamp. The refresh rate of the image frame is calculated according to the timestamp, G= , where G is the dynamic refresh rate, A is the number of image frames, and b is the time interval between each frame. The number of frames and the time interval of each time period are recorded to calculate the dynamic refresh rate of the current image. The motion characteristics of the image are analyzed through the dynamic optical flow trend feature map to identify the changes in the dynamic area of ​​the image. According to the intensity and changes of the dynamic optical flow, the appropriate dynamic refresh rate is determined. If the motion intensity is high, the refresh rate is increased to ensure smooth images. If the motion intensity is low, the refresh rate is reduced to save resources. The new dynamic optical flow adjustment refresh rate is calculated and saved as an adjustment parameter for subsequent use. According to the calculated dynamic optical flow adjustment refresh rate, the dynamic area image is matched and adjusted so that the image refresh rate meets the new dynamic requirements. Interpolation technology (such as motion compensation interpolation) is used to insert additional frames into the dynamic area image to achieve smooth visual transitions. The dynamic refresh rate adjustment area image is constructed to ensure that the image remains smooth and clear in dynamic scenes. Edge detection algorithms (such as Canny) are applied to the dynamic area image and the static area image. Edge detection) identifies the edge of the image, uses methods such as Hough transform to extract the arc features of the edge from the edge image, identifies important contours and shapes, and stores the identified edge arc data for subsequent processing and analysis, selects a suitable interpolation algorithm (such as linear interpolation, spline interpolation, etc.) to analyze the edge arc, applies the interpolation algorithm to the extracted edge arc, generates an interpolation transition zone with a gradient frame rate, ensures that the edge changes smoothly between different frames, uses the interpolation transition zone to process the dynamic refresh rate adjustment area image and the static area image, ensures that the image remains smooth under dynamic changes, and eliminates the flickering and discontinuity in the picture by dynamically gradient processing the image, making the overall visual effect smoother, and constructing a dynamic gradient refresh rate image to ensure that the image performance is more natural and comfortable in different dynamic scenes.

[0053] In this embodiment, step S4 includes the following steps: Step S41: recognizing the current user's eye monitoring image based on the display screen camera; Step S42: performing user sight tracking on the current user eye monitoring image to extract the user sight track; Step S43: analyzing the gaze lingering area of ​​the user's gaze trajectory to generate a gaze lingering area; Step S44: Calculate the duration of the sight lingering area to obtain the duration of the user's sight lingering area; Step S45: Perform a user flicker frequency sensitivity analysis based on the duration of the user's gaze, so as to obtain the dynamic refresh rate change amplitude.

[0054] In this embodiment, a camera installed on the display screen is used to capture the user's visual image in real time, ensuring that the camera can clearly focus on the user's face, and preprocessing the captured image, including denoising, color correction and contrast enhancement, to improve the accuracy of subsequent analysis. Face recognition technology is applied to identify the facial features of the current user to ensure that the user's visual image is successfully captured. An eye tracking algorithm (such as an eye tracking-based technology) is used to determine the movement of the line of sight by analyzing the user's eye position and direction, and the eye position, openness and pupil center are extracted from the user's visual image. The direction and movement trajectory of the line of sight are determined by these features, and the user's line of sight trajectory is recorded in real time, including the starting point, path and end point of the line of sight movement, to generate line of sight trajectory data, and by analyzing the line of sight trajectory data, the point where the user stays for a long time in a specific area is identified, the line of sight stay area is determined, and a threshold value of the stay time is set ( For example, if the user's gaze stays in the area for more than a certain number of milliseconds, the system can distinguish between effective gaze stay areas and short glances, and record the identified gaze stay areas as data, including the location of the stay area and the corresponding stay time. For each gaze stay area, the user's stay time in the area is calculated and recorded. The stay time data is organized into a structured format for subsequent analysis, such as forming a table containing stay areas and corresponding durations. The calculated user gaze stay time is saved as a file or database record for subsequent use and analysis. The user's flicker frequency is calculated based on the user's stay time and gaze trajectory. The user's flickering behavior in a specific area is analyzed to understand the user's sensitivity to screen content. According to the flickering frequency, the user's sensitivity to dynamic refresh rate changes is evaluated, and dynamic refresh rate change amplitude data is generated. The analysis results are recorded as structured data, including the user's flickering frequency and the corresponding dynamic refresh rate change amplitude.

[0055] In this embodiment, the specific steps of step S45 are: Performing multi-time-point blink frequency calculation on the current user eyeball monitoring image to generate user blink frequency distribution data; Perform visual stay matching on the area where the sight stays according to the user's blink frequency distribution data, and extract the blink frequency of the area where the user's vision stays; Perform flicker frequency sensitivity analysis on the blink frequency of the user's visual stay area and the user's gaze stay duration to extract the user's flicker frequency sensitivity; Predicting image refresh rate adjustment constraints based on user flicker frequency sensitivity to generate image refresh rate adjustment constraint data; The refresh rate change range of the current image dynamic refresh rate is calculated according to the image refresh rate adjustment constraint data, thereby obtaining the dynamic refresh rate change range.

[0056] In this embodiment, continuous user visual images are captured from the display screen camera to ensure that data is obtained at multiple time points, eye tracking or computer vision technology (such as Haar cascade classifier or deep learning model) is applied to detect the user's blinking behavior in real time, the number of blinks at each time point is recorded, the blink frequency is calculated, the results are organized into a time series, and the user's blink frequency distribution data is generated to determine the user's stay information in a specific area, the calculated blink frequency is matched with the corresponding sight stay area, the blink frequency of each stay area is extracted, and the flicker frequency sensitivity is calculated in combination with the user's blink frequency and sight stay duration in each visual stay area. The calculated flicker frequency sensitivity is organized into a table containing the sensitivity value of each stay area, and a suitable model (such as linear regression or machine learning algorithm) is selected to predict the constraints of image refresh rate adjustment. The user's flicker frequency sensitivity data is used, combined with the refresh requirements of dynamic scenes, to calculate a reasonable refresh rate adjustment constraint, the adjustment constraint = basic refresh rate ± k × flicker frequency sensitivity, the refresh rate change amplitude is calculated according to the adjustment constraint data: the change amplitude = current dynamic refresh rate - adjustment constraint, the change amplitude is recorded, and it is ensured that the change amplitude is within a reasonable range.

[0057] In this embodiment, the specific steps of step S5 are: Step S51: extracting the optimal adjustable refresh rate variation range based on the visual performance evaluation value of each variation range; Step S52: Calculate the light intensity of the ambient light characteristic data of the LED display screen to generate an ambient light intensity value; Step S53: performing light intensity spatial distribution analysis on the ambient light intensity value to generate ambient light intensity spatial distribution data; Step S54: performing light intensity uniformity evolution on the ambient light intensity spatial distribution data to extract ambient light intensity uniformity features; Step S55: performing light intensity fluctuation identification on the ambient light intensity value to generate light intensity fluctuation data; Step S56: fitting the ambient light distribution characteristics to the ambient light intensity uniformity characteristics and the light intensity fluctuation data to construct a light intensity scene characteristic map.

[0058] In this embodiment, visual performance evaluation values ​​of each change range are collected. These evaluation values ​​are obtained through user feedback or experimental data, reflecting the impact of different refresh rates on the user's visual experience. The performance of each change range is evaluated through statistical analysis (such as mean, variance, and maximum value). An optimization algorithm (such as genetic algorithm or particle swarm optimization) is used to extract the optimal adjustable refresh rate change range, and an objective function is determined, such as maximizing user comfort or minimizing visual fatigue. The extracted optimal adjustable refresh rate change range is saved in the data record for subsequent use, and an optical sensor or camera is used to collect LED The lighting characteristic data of the environment around the display screen, including information such as light intensity, light source type and position, is calculated through sensor data, and the ambient light intensity value is usually in lux. The ambient light intensity formula is used in combination with the output of the sensor for calculation. Based on the collected ambient light intensity values, a spatial distribution model is established, and interpolation methods (such as Kriging interpolation) are used to estimate the light intensity at different locations. A light intensity spatial distribution map is generated to show the light intensity distribution at different locations, so as to intuitively understand the distribution characteristics of ambient light, select appropriate methods (such as standard deviation, coefficient of variation) to evaluate the uniformity of light intensity distribution, and calculate the light intensity by analyzing the light intensity spatial distribution data. A strong uniformity index is used to determine whether the lighting is uniform, identify local areas of abnormal light intensity, select a suitable fluctuation recognition algorithm (such as wavelet transform or Fourier transform) to analyze changes in light intensity values, identify the amplitude and frequency of light intensity fluctuations by analyzing the light intensity time series data, generate light intensity fluctuation data, and establish an ambient light distribution feature fitting model based on the light intensity uniformity characteristics and light intensity fluctuation data. Linear regression or nonlinear fitting methods are used to generate a light intensity scene feature map through the fitting model to display the overall characteristics of the ambient light distribution, including uniformity and fluctuations. The generated light intensity scene feature map is saved as an image file or data record for subsequent reference and use.

[0059] In this embodiment, the specific steps of step S6 are: Step S61: performing scene adaptive frequency adjustment on the dynamic gradient refresh rate image based on the light intensity scene feature map to generate a dynamic frequency adaptive image; Step S62: dividing a plurality of adjustable refresh rate change ranges according to the dynamic refresh rate change range; Step S63: performing iterative adjustment simulation on the dynamic frequency adaptive image based on multiple adjustable refresh rate change ranges to generate multiple change range simulated images; Step S64: performing user visual performance evaluation on a plurality of variation amplitude simulated images to obtain a visual performance evaluation value of each variation amplitude; Step S65: extracting the optimal adjustable refresh rate change range based on the visual performance evaluation value of each change range; Step S66: Optimize the dynamic visual frequency of the dynamic frequency adaptive image using the optimal adjustable refresh rate variation range, and construct a dynamic visual frequency optimization model.

[0060] In this embodiment, the light intensity scene feature map generated previously is used to analyze the uniformity and fluctuation characteristics of light intensity in different areas in order to determine a suitable refresh rate. An adaptive frequency adjustment strategy is formulated according to the light intensity distribution. For example, in areas with uniform light intensity, a lower refresh rate is selected; in areas with large light intensity fluctuations, a higher refresh rate is selected. The adjustment strategy is applied to the dynamic gradient refresh rate image to generate a new dynamic frequency adaptive image to ensure that the image can maintain a good visual effect in different scenes. Based on the dynamic refresh rate variation amplitude data obtained, its range and distribution characteristics are analyzed. According to the distribution of the variation amplitude, multiple adjustable refresh rate variation amplitudes are set. For example, the range is evenly divided into several intervals, each interval represents an adjustable amplitude, and the multiple adjustable refresh rate variation amplitudes obtained by division are saved as structured data for subsequent use. An iterative adjustment simulation framework is determined, and the initial image is set as a dynamic frequency adaptive image. Stepwise adjust each adjustable refresh rate change amplitude to generate multiple simulated images with different change amplitudes, use interpolation methods or motion compensation algorithms to maintain image continuity, save the generated multiple simulated images with different change amplitudes as a data set for subsequent visual performance evaluation, determine the user's visual performance evaluation criteria, including visual comfort, clarity, and smoothness, collect feedback through user experiments, let users select the image with the best visual performance from the simulated images with different change amplitudes, organize the visual performance evaluation value of each change amplitude into a scoring table, record user feedback and scoring data, perform statistical analysis on the user's visual performance evaluation value, and identify the best performing change amplitude. Calculate the average evaluation value of each change range, determine the adjustable refresh rate change range with the highest score as the optimal value, save the extracted optimal adjustable refresh rate change range as a data record for subsequent dynamic visual frequency optimization, build a dynamic visual frequency optimization model based on the optimal adjustable refresh rate change range, define the optimization goals and constraints, optimize the dynamic frequency adaptive image by applying the optimal refresh rate, adjust the image refresh frequency to achieve the best visual effect in different scenarios, verify the optimized image effect, confirm the optimization effect through user feedback and automatic evaluation indicators, and save the final dynamic visual frequency optimization model as a document or software module.

[0061] In this embodiment, a system for adaptively adjusting the content refresh rate of an LED display screen is provided, which is used to execute the method for adaptively adjusting the content refresh rate of an LED display screen as described above, including: The area division module is used to identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image and perform dynamic area division to generate dynamic area images and static area images; The optical flow trend module is used to track the dynamic texture optical flow of dynamic area images and evolve the optical flow change trend to generate a dynamic optical flow trend feature map; The dynamic gradient module is used to perform dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map, and construct a dynamic gradient refresh rate image; A visual frequency sensitive module is used to identify the current user eyeball monitoring image based on the display screen camera; perform user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain the dynamic refresh rate change amplitude; The ambient light feature module is used to obtain the ambient light feature data of the LED display screen based on the sensor; perform ambient light distribution feature fitting on the ambient light feature data of the LED display screen to construct a light intensity scene feature map; The dynamic visual frequency optimization module is used to optimize the dynamic visual frequency of the dynamic gradient refresh rate image based on the light intensity scene feature map and the dynamic refresh rate change amplitude, and to build a dynamic visual frequency optimization model.

[0062] The present invention identifies the LED display screen image and performs detail convolution reconstruction to extract the subtle features of the displayed content, providing high-quality input data for subsequent processing. The dynamic area division can divide the image into a dynamic area and a static area according to the changes in the displayed content, providing differentiation for subsequent processing and optimizing the processing effect. The dynamic texture optical flow tracking captures the trend and change of the movement of objects in the image, providing dynamic information for subsequent processing. The generation of a dynamic optical flow trend feature map can reflect the change of the optical flow in the image, providing important information about the dynamics of the image for subsequent processing. The dynamic gradient smoothing optimization can perform gradient processing on the image according to the dynamic optical flow trend feature map, improve the continuity and smoothness of the displayed content, construct a dynamic gradient refresh rate image, adjust the refresh rate according to the dynamics of the image content, improve the comfort and visual effect of the user's viewing experience, and analyze the user's flicker frequency. Rate sensitivity can determine the dynamic refresh rate change range according to the user's perception of the refresh rate change, provide a personalized visual experience, and obtain the dynamic refresh rate change range to adjust the refresh frequency of the displayed content according to the user's visual needs to improve the user's viewing comfort. The fitted light intensity scene feature map can construct an ambient light distribution map according to the characteristics of the ambient light around the LED display screen, provide ambient light information for subsequent adjustments, and obtain ambient light feature data to optimize the displayed content according to the lighting conditions to improve the display effect and viewing experience. Dynamic visual frequency optimization based on the light intensity scene feature map and the dynamic refresh rate change range can adjust the refresh rate according to ambient light conditions and user needs, provide a personalized visual experience, and construct a dynamic visual frequency optimization model to comprehensively adjust the refresh rate of the LED display screen according to different factors, so that users can get a more comfortable and high-quality visual experience.

[0063] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0064] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for adaptively adjusting the content refresh rate of an LED display screen, characterized in that: The following steps are involved: Step S1: Identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image, and perform dynamic area division to generate a dynamic area image and a static area image; Step S2: Tracking the dynamic texture optical flow of the dynamic region image and evolving the optical flow change trend to generate a dynamic optical flow trend feature map; Step S3: performing dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map to construct a dynamic gradient refresh rate image; Step S4: identifying the current user eyeball monitoring image based on the display screen camera; performing user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain a dynamic refresh rate change amplitude; Step S5: Acquire ambient light characteristic data of the LED display screen based on the sensor; Fit the ambient light distribution characteristics of the ambient light characteristic data of the LED display screen to construct a light intensity scene characteristic map; Step S6: Based on the light intensity scene feature map and the dynamic refresh rate variation range, the dynamic visual frequency of the dynamic gradient refresh rate image is optimized to construct a dynamic visual frequency optimization model.

2. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Identify the LED display screen image; Step S12: performing detail convolution reconstruction on the LED display screen image to construct a detail sharpened reconstructed image; Step S13: performing time-series frame segmentation processing on the detail-sharpened reconstructed image, thereby obtaining an image frame sequence; Step S14: performing continuous frame dynamic optical flow recognition on the image frame sequence to generate inter-frame dynamic optical flow data; Step S15: performing dynamic and static texture division according to the inter-frame dynamic optical flow data, thereby generating a dynamic region image and a static region image.

3. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 2, characterized in that: The specific steps of step S12 are: Calculate the average brightness of the LED display screen image and extract the average brightness value of the display screen image; Based on the average brightness value of the display screen image, the dark area of ​​the LED display screen image is identified and the dark area of ​​the image is extracted; Performing local brightness enhancement processing on the dark area of ​​the image, thereby obtaining a brightness enhanced display screen image; Perform image noise detection on the brightness enhanced display screen image and mark multiple image noise points; Performing mean filtering to remove noise points in multiple images to generate a filtered and denoised display screen image; Perform multi-layer convolution decomposition processing on the filtered and denoised display screen image to obtain a multi-layer convolution image; Adaptively sharpen the multi-layer convolution images to obtain multiple detail-sharpened convolution images; The filtered denoised display screen image is deconvolved and reconstructed based on a plurality of detail-sharpened convolution images to construct a detail-sharpened reconstructed image.

4. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing dynamic texture optical flow tracking on the dynamic area image to generate a dynamic texture optical flow trajectory; Step S22: Calculate the texture change frequency of the dynamic texture optical flow trajectory to extract the dynamic texture change frequency; Step S23: performing color type statistics on the dynamic area image to obtain the image color type; Step S24: performing color saturation change analysis based on the image color type to generate image color saturation change data; Step S25: Perform optical flow change trend evolution on the dynamic texture change frequency and image color saturation change data, thereby generating a dynamic optical flow trend feature map.

5. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: measuring the image refresh rate of the image frame sequence and extracting the current image dynamic refresh rate; Step S32: dynamically adjusting the dynamic refresh rate of the current image based on the dynamic optical flow trend feature map to obtain a dynamic optical flow adjusted refresh rate; Step S33: dynamically adjusting the refresh rate of the dynamic area image based on the dynamic optical flow to adjust the refresh rate, and constructing a dynamic refresh rate adjusted area image; Step S34: performing edge arc recognition on the dynamic region image and the static region image to extract region edge arcs; Step S35: performing a gradual frame rate interpolation analysis on the region edge arc, thereby obtaining a gradual frame rate interpolation transition zone; Step S36: using a gradual frame rate interpolation transition band to perform dynamic gradual smoothing optimization on the dynamic refresh rate adjustment area image and the static area image to construct a dynamic gradual refresh rate image.

6. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: recognizing the current user's eye monitoring image based on the display screen camera; Step S42: performing user sight tracking on the current user eye monitoring image to extract the user sight track; Step S43: analyzing the gaze lingering area of ​​the user's gaze trajectory to generate a gaze lingering area; Step S44: Calculate the duration of the sight lingering area to obtain the duration of the user's sight lingering area; Step S45: Perform a user flicker frequency sensitivity analysis based on the duration of the user's gaze, so as to obtain the dynamic refresh rate change amplitude.

7. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 6, characterized in that: The specific steps of step S45 are: Performing multi-time-point blink frequency calculation on the current user eyeball monitoring image to generate user blink frequency distribution data; Perform visual stay matching on the area where the sight stays according to the user's blink frequency distribution data, and extract the blink frequency of the area where the user's vision stays; Perform flicker frequency sensitivity analysis on the blink frequency of the user's visual stay area and the user's gaze stay duration to extract the user's flicker frequency sensitivity; Predicting image refresh rate adjustment constraints based on user flicker frequency sensitivity to generate image refresh rate adjustment constraint data; The refresh rate change range of the current image dynamic refresh rate is calculated according to the image refresh rate adjustment constraint data, thereby obtaining the dynamic refresh rate change range.

8. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: extracting the optimal adjustable refresh rate variation range based on the visual performance evaluation value of each variation range; Step S52: Calculate the light intensity of the ambient light characteristic data of the LED display screen to generate an ambient light intensity value; Step S53: performing light intensity spatial distribution analysis on the ambient light intensity value to generate ambient light intensity spatial distribution data; Step S54: performing light intensity uniformity evolution on the ambient light intensity spatial distribution data to extract ambient light intensity uniformity features; Step S55: performing light intensity fluctuation identification on the ambient light intensity value to generate light intensity fluctuation data; Step S56: fitting the ambient light distribution characteristics to the ambient light intensity uniformity characteristics and the light intensity fluctuation data to construct a light intensity scene characteristic map.

9. The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing scene adaptive frequency adjustment on the dynamic gradient refresh rate image based on the light intensity scene feature map to generate a dynamic frequency adaptive image; Step S62: dividing a plurality of adjustable refresh rate change ranges according to the dynamic refresh rate change range; Step S63: performing iterative adjustment simulation on the dynamic frequency adaptive image based on multiple adjustable refresh rate change ranges to generate multiple change range simulated images; Step S64: performing user visual performance evaluation on a plurality of variation amplitude simulated images to obtain a visual performance evaluation value of each variation amplitude; Step S65: extracting the optimal adjustable refresh rate change range based on the visual performance evaluation value of each change range; Step S66: Utilize the optimal adjustable refresh rate variation range to perform dynamic visual frequency optimization on the dynamic frequency adaptive image, and construct a dynamic visual frequency optimization model.

10. An adaptive adjustment system for content refresh rate of LED display screen, characterized in that: The method for adaptively adjusting the content refresh rate of an LED display screen according to claim 1 comprises: The area division module is used to identify the LED display screen image; perform detail convolution reconstruction on the LED display screen image and perform dynamic area division to generate dynamic area images and static area images; The optical flow trend module is used to track the dynamic texture optical flow of dynamic area images and evolve the optical flow change trend to generate a dynamic optical flow trend feature map; The dynamic gradient module is used to perform dynamic gradient smoothing optimization on the dynamic area image and the static area image according to the dynamic optical flow trend feature map, and construct a dynamic gradient refresh rate image; A visual frequency sensitive module is used to identify the current user eyeball monitoring image based on the display screen camera; perform user flicker frequency sensitivity analysis on the current user eyeball monitoring image to obtain the dynamic refresh rate change amplitude; The ambient light feature module is used to obtain the ambient light feature data of the LED display screen based on the sensor; perform ambient light distribution feature fitting on the ambient light feature data of the LED display screen to construct a light intensity scene feature map; The dynamic visual frequency optimization module is used to optimize the dynamic visual frequency of the dynamic gradient refresh rate image based on the light intensity scene feature map and the dynamic refresh rate change amplitude, and to build a dynamic visual frequency optimization model.

Citation Information

Cited By

  • Energy-saving control method and device for LED special-shaped screen

    CN119811277A

  • Display performance optimization method and system of OLED screen and computer readable storage medium

    CN120319172A

  • Modeling low-power-consumption display driving method based on multi-frequency linkage

    CN120472855A

  • Advertising board content safety monitoring method and system for expressway

    CN120656176A

  • Billboard content security monitoring method and system for highways

    CN120656176B