LED display screen partitioning method

Abstract

The invention discloses an LED display screen partitioning method, which comprises the following steps: obtaining content information to be displayed, and performing feature extraction and classification on display content to obtain display feature parameters of different regions; for each partition, calculating an optimal display parameter by adopting a parameter optimization algorithm according to the display characteristic parameter of the partition; interpolation and transition processing are carried out on parameters of adjacent partitions at the boundary of the partitions, and abrupt and flicker during partition switching are avoided; through a historical data analysis and prediction algorithm, the display content in a period of time in the future is predicted, and partitioning and parameter optimization are carried out in advance; according to the characteristics and changes of the display content, the display effect is adaptively optimized, hardware resources are saved to the maximum extent while the display quality is guaranteed, and intelligent and fine management of the display content is achieved through comprehensive application of technologies such as dynamic partitioning, parameter optimization, smooth transition and predictive analysis.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to a method for partitioning an LED display screen. Background Art

[0002] In actual applications, LED displays often face the problem of complex and changeable display content. On the one hand, the display content may include different types such as text, images, and videos, and the size, position, and change frequency of the content are different. On the other hand, the hardware resources of LED displays, such as resolution, brightness, refresh rate and other parameters are limited.

[0003] How to achieve the best display effect for complex and changeable display content under limited hardware conditions is a technical problem that needs to be solved urgently. Specifically, different areas of display content have very different requirements for display quality and resources. For example, the text area is relatively static, while the video area changes frequently; the important information area requires higher display quality, while the background area is relatively unimportant; if a unified parameter setting is used for the entire display screen, it will inevitably lead to resource waste or poor display quality. If different parameters are used for different areas, an efficient and reasonable partitioning scheme is required to make full use of limited resources while avoiding frequent switching and additional overhead caused by too many partitions. The partitioning scheme also needs to be able to adapt to the dynamic changes of the display content and adjust the partition boundaries and attributes in time to ensure the overall display effect. This requires the partitioning algorithm to quickly and accurately analyze the characteristics of the display content and generate partition grids and parameters that match it. Summary of the invention

[0004] In order to solve the above-mentioned technical problems, the present invention provides a method for partitioning an LED display screen.

[0005] The technical solution of the present invention is achieved in this way:

[0006] A method for partitioning an LED display screen comprises the following steps:

[0007] S1. Obtaining content information to be displayed, extracting and classifying features of the displayed content, and obtaining display feature parameters of different areas;

[0008] S2. Adaptively partition the display screen according to display feature parameters and hardware resource parameters, dynamically generate a partition grid, and determine the position, size, and shape of each partition;

[0009] S3. For each partition, calculate the optimal display parameter according to its display characteristic parameter;

[0010] S4. During the display process, continuously monitor the changes in the displayed content to determine whether re-partitioning and parameter optimization are required. If significant changes are detected, partition adjustment and parameter update are triggered;

[0011] S5. At the partition boundary, interpolate and transition the parameters of adjacent partitions;

[0012] S6. Predict the display content in the future and optimize the partitions and parameters in advance.

[0013] Furthermore, the feature extraction and classification of the displayed content in S1 is specifically as follows:

[0014] Acquire multi-dimensional attribute information of the content to be displayed, including content type, size, location and change frequency, use content analysis algorithm to extract features of the display content, use clustering algorithm to classify the display content according to the extracted content features, and obtain content classification results of different display areas. For each display area, use decision tree algorithm to generate display feature parameters of the area according to the type, size, location and change frequency attributes of the display content in the area;

[0015] For display areas with high frequency of changes, more content attribute samples are obtained by increasing the sampling frequency to improve the accuracy of feature parameters. For display areas with complex content types and large feature differences, the accuracy of classification and parameter generation is improved by subdividing the content types and using multiple feature extraction algorithms. According to the spatial adjacency relationship of the display areas, the area division results are optimized through image segmentation algorithms to reduce regional fragmentation and improve the continuity of display feature parameters. The feature parameters of each display area are combined to generate complete display screen parameters to guide the generation and optimization of the display screen.

[0016] Furthermore, in S2, the display screen is adaptively partitioned, and a partition optimization algorithm is used to adaptively partition the display screen according to display feature parameters, hardware resource parameters, initial parameters and constraints, and the display screen is divided into multiple partitions. For each partition, a partition grid is dynamically generated according to the partition optimization algorithm, and the number of rows and columns of the partition grid and the position coordinates of each grid unit are determined. According to the number of rows and columns of the partition grid and the position coordinates of each grid unit, the position, size and shape of each partition are calculated to obtain a partition layout scheme;

[0017] Obtain the content to be displayed, and map it to the corresponding partition according to the partition layout plan to achieve adaptive display of the content in the partition. When the displayed content changes, re-obtain the display feature parameters and hardware resource parameters to determine whether the partition layout needs to be adjusted. If adjustment is required, re-execute the partition optimization algorithm to adjust the partition grid and partition layout in real time according to the changes in the displayed content and the dynamic changes in hardware resources.

[0018] Furthermore, the step of calculating the optimal display parameters in S3 includes:

[0019] S301, obtaining characteristic parameters of each display partition and using them as input of a parameter optimization algorithm;

[0020] S302, establishing a mapping relationship between display parameters and quality according to a preset display quality evaluation model, and obtaining an objective function for parameter optimization;

[0021] S303, using a particle swarm optimization algorithm, taking the characteristic parameters of the display partition as constraints, minimizing the objective function, and searching for the optimal display parameter combination;

[0022] S304, during the optimization process, the particle swarm is continuously updated iteratively by dynamically adjusting the speed and position of the particles until a preset convergence condition is met or a preset maximum number of iterations is reached;

[0023] S305, selecting a particle with the smallest objective function value from the converged particle swarm, and setting the corresponding display parameter as the optimal parameter setting for the partition;

[0024] S306, repeating steps S303 to S305 for all display partitions until an optimal display parameter combination for each partition is obtained;

[0025] S307: Apply the optimized display parameters to the display control of each partition.

[0026] Furthermore, during the display process, S4 continuously monitors changes in the displayed content and determines whether re-partitioning and parameter optimization are required, which specifically includes:

[0027] Acquire the current display content of the display device, divide the display content into multiple display partitions, extract display features for each display partition, compare the currently extracted display features with preset historical display features, and determine the degree of change between the current display content and the historical display content through a feature similarity judgment algorithm;

[0028] If the degree of change exceeds the preset significant change threshold, the redivision of the display partitions is triggered, and the number and size of the display partitions are dynamically adjusted using an adaptive partitioning algorithm based on the changing characteristics of the displayed content;

[0029] For each re-divided display partition, the updated display features are extracted, and the display parameters of each display partition, including brightness, contrast, and color saturation, are determined through a feature analysis algorithm. According to the change range of the display parameters, a progressive parameter optimization algorithm is used to gradually adjust the display parameters to the target values ​​within multiple display cycles to avoid sudden changes in the display effect. During the display parameter adjustment process, the changes in the display content are continuously monitored;

[0030] If significant changes are detected during the adjustment process, the current parameter adjustment process is interrupted and the display partition division and parameter optimization are re-performed;

[0031] If the display content still changes significantly after the display parameter adjustment is completed, the significant change threshold is reduced and the sensitivity of change detection is increased to respond to changes in the display content in a timely manner. At the same time, the current display features are updated to the historical display feature library as a reference for subsequent change detection.

[0032] Furthermore, in S5, the parameters of adjacent partitions are interpolated and transitioned at the partition boundaries, specifically:

[0033] Obtain the location coordinate information of the partition boundary, determine the adjacent partitions that need to be smoothly transitioned, perform transition processing on the parameters of the adjacent partitions, calculate the interpolation coefficients according to the location coordinates of the partition boundary, perform weighted averaging on the parameters of the adjacent partitions according to the interpolation coefficients, obtain the parameter values ​​of the transition area, and achieve smooth transition of the parameters;

[0034] When switching between partitions, determine whether the current display area is in the transition area. If it is in the transition area, use the parameters of the transition area for display. Render and display the image according to the parameter values ​​of the transition area to avoid abruptness and flickering when switching between partitions and improve display continuity.

[0035] Adopting adaptive weight adjustment algorithm, the interpolation coefficient and the range of transition area are dynamically adjusted according to the size of transition area and the frequency of partition switching;

[0036] Through the iterative optimization algorithm, the interpolation coefficients and parameters of the transition area are continuously optimized to minimize the visual differences when switching between partitions and improve the consistency of display.

[0037] Furthermore, the prediction of the display content in a future period of time in S6 and the optimization of the partition and parameters in advance specifically include:

[0038] The acquired historical display content data is cleaned and preprocessed to remove invalid and abnormal data to obtain a normalized historical data set. According to the characteristics of the historical data set, appropriate time series prediction models such as ARIMA and LSTM are selected to build a prediction model. The model is trained and optimized using methods such as cross-validation to improve the accuracy of the prediction.

[0039] Use the trained prediction model to predict the display content in the future to obtain the predicted content data set, and perform confidence analysis on the prediction results to determine the reliability of the prediction. According to the predicted content data set, analyze the characteristics and distribution of the content, use clustering algorithms to group the content, obtain content clusters with different characteristics, and determine the priority of different clusters according to the size and access frequency of the content clusters. Design corresponding storage partitions and index strategies for different content clusters, load the predicted content data into the corresponding partitions in advance, and optimize index parameters to improve data retrieval and access efficiency;

[0040] When actually displaying content, quickly locate the corresponding data partition and index according to the content characteristics of the request, directly obtain the pre-processed content data, reduce the overhead of real-time calculation and retrieval, and improve the response speed;

[0041] Continuously monitor the operating status and performance indicators of the system, collect user access logs and feedback data, regularly evaluate and optimize the prediction model and partitioning strategy, continuously improve the prediction accuracy and service quality, and ensure the stable operation of the system.

[0042] Further,

[0043] Further,

[0044] Further,

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention extracts and classifies the display content through a content analysis algorithm to obtain display feature parameters of different areas. According to these parameters and hardware resource conditions, a partition optimization algorithm is used to dynamically partition the display screen, and the optimal display parameters are calculated for each partition. During the display process, content changes are continuously monitored and adjusted in real time. In order to avoid the abrupt feeling when switching partitions, a smooth transition algorithm is used at the partition boundary. In addition, through historical data analysis and prediction algorithms, future display content is predicted, and partitions and parameters are optimized in advance. The present invention can adaptively optimize the display effect according to the characteristics and changes of the display content, and save hardware resources to the maximum extent while ensuring the display quality. Through the comprehensive application of technologies such as dynamic partitioning, parameter optimization, smooth transition and prediction analysis, the present invention realizes intelligent and refined management of display content, and improves display performance and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a process of a LED display screen partitioning method of the present invention;

[0048] Figure 2 It is a schematic diagram of the process of calculating the optimal display parameters in S3 of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example 1

[0051] like Figure 1-2 As shown, a method for partitioning an LED display screen includes the following steps:

[0052] S1. Obtaining content information to be displayed, extracting and classifying features of the displayed content, and obtaining display feature parameters of different areas;

[0053] S2. Adaptively partition the display screen according to display feature parameters and hardware resource parameters, dynamically generate a partition grid, and determine the position, size, and shape of each partition;

[0054] S3. For each partition, calculate the optimal display parameter according to its display characteristic parameter;

[0055] S4. During the display process, continuously monitor the changes in the displayed content to determine whether re-partitioning and parameter optimization are required. If significant changes are detected, partition adjustment and parameter update are triggered;

[0056] S5. At the partition boundary, interpolate and transition the parameters of adjacent partitions to avoid abruptness and flickering when switching partitions;

[0057] S6. Predict the display content in the future and optimize the partitions and parameters in advance.

[0058] Furthermore, the feature extraction and classification of the displayed content in S1 is specifically as follows:

[0059] Obtain multi-dimensional attribute information of the content to be displayed, including content type, size, location and change frequency, and use content analysis algorithm to extract features of the displayed content. Based on the extracted content features, the content analysis algorithm can extract features such as moving targets, text areas, color distribution, etc. in the picture, laying the foundation for subsequent processing;

[0060] A clustering algorithm is used to classify the display content to obtain the content classification results of different display areas. For each display area, a decision tree algorithm is used to generate the display feature parameters of the area according to the type, size, location and change frequency attributes of the display content in the area. The application of the clustering algorithm can divide the display content into different areas.

[0061] For display areas with high frequency of changes, more content attribute samples are obtained by increasing the sampling frequency to improve the accuracy of feature parameters. For display areas with complex content types and large feature differences, the accuracy of classification and parameter generation is improved by subdividing the content types and using multiple feature extraction algorithms. Specifically:

[0062] For display areas with high frequency of changes, increasing the sampling frequency is an effective way to improve the accuracy of feature parameters. The optimal display parameters for action scenes and static scenes may be very different. By increasing the sampling frequency of action scenes, picture changes can be captured more accurately, thereby dynamically adjusting parameters such as color saturation and motion compensation to provide a smoother viewing experience.

[0063] For display areas with complex content types and large feature differences, more sophisticated processing is required. By subdividing the content types, different feature extraction algorithms can be used for different elements. For example, for images containing multiple elements such as text, charts, and pictures, optical character recognition (OCR) technology can be used to extract text information from text areas, while edge detection algorithms can be used to extract contour features from charts. The collaborative approach of multiple algorithms can significantly improve the classification accuracy and the accuracy of parameter generation.

[0064] According to the spatial adjacency relationship of the display area, the region division result is optimized by the image segmentation algorithm, the regional fragmentation is reduced, and the continuity of the display feature parameters is improved. Specifically, since adjacent display elements may be incorrectly divided into multiple fragmented regions, these adjacent regions can be effectively merged by applying image segmentation algorithms such as region growing or watershed to form a more coherent display unit, thereby improving the continuity of the display feature parameters and the overall visual effect; the feature parameters of each display area are combined to generate a complete display screen parameter, which is used to guide the generation and optimization of the display screen;

[0065] Preferably, in a multi-screen collaborative working environment, the display characteristics of different devices are different, such as laptops, tablets, and projectors. By comprehensively considering the color space, resolution, refresh rate and other parameters of each device, a set of globally optimal display solutions can be generated;

[0066] For example, when sharing content across devices, color mapping and scaling can be automatically adjusted to ensure that the content presents consistent visual effects on different devices. The display optimization method based on multi-dimensional attribute analysis and multi-algorithm collaboration can improve display quality and adapt to the needs of different scenarios and devices.

[0067] Through continuous learning and optimization, we can continuously improve the user's visual experience and provide strong technical support for various display applications.

[0068] Furthermore, in S2, the display screen is adaptively partitioned, and a partition optimization algorithm is used according to display feature parameters, hardware resource parameters, initial parameters and constraints, wherein the feature parameters include resolution, color depth, refresh rate, etc., and the hardware resource parameters involve GPU performance, memory capacity, etc. These parameters directly affect the initial settings and constraints of the partition optimization algorithm. The core of the partition optimization algorithm is to dynamically divide the screen according to display content characteristics and hardware resources;

[0069] Adaptively partition the display screen into multiple partitions. For each partition, dynamically generate a partition grid according to a partition optimization algorithm, determine the number of rows and columns of the partition grid, and the position coordinates of each grid unit. According to the number of rows and columns of the partition grid and the position coordinates of each grid unit, calculate the position, size and shape of each partition to obtain a partition layout plan.

[0070] Among them, dynamically generating partition grids is a key step in achieving refined display control. The regional grids are divided according to different display contents. Each grid unit can accommodate different types of content, such as titles, text, pictures or advertisements. The number of rows and columns and unit coordinates of the grid will be adjusted in real time according to the device orientation and user preferences.

[0071] The calculation of the partition layout scheme needs to comprehensively consider the grid structure and content characteristics. For example, on the email page, it can be dynamically adjusted according to the user's reading habits and the number of emails;

[0072] Obtain the content to be displayed, and map the content to be displayed to the corresponding partition according to the partition layout scheme to realize adaptive display of the content in the partition. When the displayed content changes, re-obtain the display feature parameters and hardware resource parameters to determine whether the partition layout needs to be adjusted. If adjustment is required, re-execute the partition optimization algorithm to adjust the partition grid and partition layout in real time according to the changes in the displayed content and the dynamic changes in hardware resources.

[0073] Mapping content to partitions is a key step in achieving adaptive display. The following uses video as an example. In a video streaming platform, video content with different resolutions and aspect ratios needs to be mapped to predefined partitions.

[0074] Partition adjustment when display content changes is the key to maintaining a good user experience. In multitasking, when users switch from full-screen video playback to split-screen working mode, it is necessary to re-evaluate hardware resource allocation and display characteristics. For example, GPU resources originally used for video decoding may need to be reallocated to document processing and browser rendering, and the partition layout also needs to be adjusted from a single large partition to multiple small partitions. Each partition requires an independent optimization strategy.

[0075] Real-time adjustment of partition grids and layouts is an effective way to cope with dynamic content and changes in hardware resources. Taking the process of running a game as an example, the rendering quality and resolution need to be dynamically adjusted according to the complexity of the scene.

[0076] Furthermore, the step of calculating the optimal display parameters in S3 includes:

[0077] S301, obtaining characteristic parameters of each display partition and using them as input of a parameter optimization algorithm;

[0078] The characteristic parameters of each display partition include area, shape, position and other information, which will directly affect the subsequent optimization process;

[0079] S302, establishing a mapping relationship between display parameters and quality according to a preset display quality evaluation model, and obtaining an objective function for parameter optimization;

[0080] Taking mobile devices as an example, the establishment of a display quality evaluation model can take into account factors such as brightness, contrast, and color saturation. Through user surveys and expert evaluations, a comprehensive scoring model can be established to quantify these factors into scores of 0-100;

[0081] S303, using a particle swarm optimization algorithm, taking the characteristic parameters of the display partition as constraints, minimizing the objective function, and searching for the optimal display parameter combination;

[0082] Assuming that the display parameters of the content area need to be optimized, brightness, contrast and saturation can be used as three dimensions. Each particle represents a set of parameter combinations, such as (brightness 70%, contrast 80%, saturation 60%). Initially, 50 particles are randomly generated and distributed in the parameter space. By iteratively updating the position and speed of the particles, the algorithm can find the most power-saving parameter combination while ensuring display quality.

[0083] S304, during the optimization process, the particle swarm is continuously updated iteratively by dynamically adjusting the speed and position of the particles until a preset convergence condition is met or a preset maximum number of iterations is reached;

[0084] During the optimization process, if it is found that particles in a certain area perform well, the algorithm will increase the search density in this area. For example, if it is found that particles with brightness between 60% and 70% generally perform well, more particles will gather in this range in subsequent iterations. This strategy can speed up convergence and improve optimization efficiency. The setting of convergence conditions requires a balance between optimization effect and computational cost. The maximum number of iterations can be set to 100, or convergence is considered when the improvement of the optimal solution for 10 consecutive iterations is less than 1%.

[0085] S305, selecting a particle with the smallest objective function value from the converged particle swarm, and setting the corresponding display parameter as the optimal parameter setting for the partition;

[0086] S306, repeating steps S303 to S305 for all display partitions until an optimal display parameter combination for each partition is obtained;

[0087] S307: Apply the optimized display parameters to the display control of each partition.

[0088] Furthermore, during the display process, S4 continuously monitors changes in the displayed content and determines whether re-partitioning and parameter optimization are required, which specifically includes:

[0089] Acquire the current display content of the display device, divide the display content into multiple display partitions, extract display features for each display partition, compare the currently extracted display features with preset historical display features, and determine the degree of change between the current display content and the historical display content through a feature similarity judgment algorithm;

[0090] Specifically, the current content of the display device is acquired in real time by sampling the data of the frame buffer. After the content is acquired, the display area is divided into multiple partitions, and clustering is performed based on the similarity of the content;

[0091] The display features of each area will be extracted as the feature vector of the area. In order to determine the degree of change in the displayed content, the currently extracted features are compared with the stored historical features, and algorithms such as cosine similarity are used to calculate the similarity between feature vectors;

[0092] If the degree of change exceeds the preset significant change threshold, the redivision of the display partition is triggered. According to the changing characteristics of the displayed content, the adaptive partitioning algorithm is used to dynamically adjust the number and size of the display partitions. Assuming that the significant change threshold is set to 0.8, when the similarity is lower than this value, redivision will be triggered, and the adaptive partitioning algorithm will take effect when a significant change is detected. For example, when the user switches from the text reading interface to the video playback interface, the partitioning strategy will be adjusted from a multi-region division based on text to a large-region division based on video images, ensuring that the partitioning strategy always matches the current display content;

[0093] For each re-divided display partition, the updated display features are extracted, and the display parameters of each display partition, including brightness, contrast, and color saturation, are determined through a feature analysis algorithm. According to the change range of the display parameters, a progressive parameter optimization algorithm is used to gradually adjust the display parameters to the target values ​​within multiple display cycles to avoid sudden changes in the display effect. During the display parameter adjustment process, the changes in the display content are continuously monitored;

[0094] If significant changes are detected during the adjustment process, the current parameter adjustment process is interrupted and the display partition division and parameter optimization are re-performed;

[0095] If the display content still changes significantly after the display parameter adjustment is completed, the significant change threshold is reduced to increase the sensitivity of change detection so as to respond to the change of the display content in a timely manner. At the same time, the current display features are updated to the historical display feature library as a reference for subsequent change detection;

[0096] In this step, for each newly divided area, the display features will be re-extracted and the optimal display parameters will be determined accordingly. In the video playback area, the color saturation and contrast may be increased to enhance the picture effect; while in the surrounding control button area, the brightness may be reduced to save energy. In order to avoid sudden changes in the display effect, a progressive parameter optimization algorithm is used. For example, when the brightness of a certain area needs to be increased from 50% to 80%, it may be gradually adjusted within 5 display cycles, increasing by about 6% each time. This smooth transition not only protects the user's visual experience, but also reduces the instantaneous load of the hardware;

[0097] During the parameter adjustment process, the display content changes are continuously monitored. If it is detected that the user has switched applications during the adjustment, the current adjustment process will be immediately interrupted, and the partition and parameter optimization will be re-performed to ensure that the display effect is always consistent with the content. If after completing a round of parameter adjustment, it is found that the display content is still changing frequently, it will automatically lower the significant change threshold, such as from 0.8 to 0.7. This can increase the sensitivity to subtle changes and better adapt to dynamic content; at the same time, the current display features will be updated to the historical feature library for subsequent change detection, which can not only improve the display quality, but also effectively save energy consumption; through continuous learning and adaptive adjustment, it can continuously optimize its performance and provide users with a smarter and more efficient display experience.

[0098] Furthermore, in S5, the parameters of adjacent partitions are interpolated and transitioned at the partition boundaries, specifically:

[0099] Obtain the location coordinate information of the partition boundary, determine the adjacent partitions that need to be smoothly transitioned, perform transition processing on the parameters of the adjacent partitions, calculate the interpolation coefficients according to the location coordinates of the partition boundary, perform weighted averaging on the parameters of the adjacent partitions according to the interpolation coefficients, obtain the parameter values ​​of the transition area, and achieve smooth transition of the parameters;

[0100] Interpolation algorithms are used for processing. Common interpolation methods include linear interpolation, bilinear interpolation, and cubic spline interpolation. In this step, a linear interpolation algorithm is used. Taking a smartphone screen as an example, assuming that the brightness parameter of the status bar is 80% and the content area is 60%, a 10-pixel-wide transition area will be created between the two partitions. In this area, the brightness will gradually transition from 80% to 60% to avoid abrupt brightness jumps.

[0101] The calculation of the interpolation coefficient is based on the position coordinates of the partition boundary. In the above example, the interpolation coefficient of each pixel in the transition area can be expressed as: (current pixel position - transition area starting position) / transition area width. The pixels close to the status bar will be closer to 80% brightness, while the pixels close to the content area will be closer to 60% brightness.

[0102] According to the calculated interpolation coefficients, the parameters of adjacent partitions are weighted averaged. In addition to brightness, other display parameters such as contrast and color temperature can also achieve smooth transition through similar methods. This processing method is not only suitable for static images, but can also effectively reduce the visual discontinuity between partitions in dynamic scenes such as video playback.

[0103] When switching between partitions, determine whether the current display area is in the transition area. If it is in the transition area, use the parameters of the transition area for display. Render and display the image according to the parameter values ​​of the transition area to avoid abruptness and flickering when switching between partitions and improve display continuity.

[0104] In the actual display process, it is determined whether the currently rendered pixel is located in the transition area. If so, the calculated transition parameters are used for display. This method can effectively avoid the abruptness and flickering when switching partitions, greatly improving the display continuity and viewing comfort.

[0105] Adopting adaptive weight adjustment algorithm, the interpolation coefficient and the range of transition area are dynamically adjusted according to the size of transition area and the frequency of partition switching;

[0106] Through iterative optimization algorithms, the interpolation coefficients and transition area parameters are continuously optimized to minimize the visual differences when switching between partitions and improve display consistency.

[0107] In this step, in order to further optimize the display effect, an adaptive weight adjustment algorithm is used to dynamically adjust the interpolation coefficient and the range of the transition area according to the size of the transition area and the frequency of partition switching; for example, in a fast-scrolling web page, the width of the transition area may be increased to ensure a smoother visual experience; on the contrary, when a static image is displayed, the transition area may be reduced to maintain the clarity of the image;

[0108] Through iterative optimization algorithms, we continuously optimize the interpolation coefficients and parameters of the transition area, including collecting user feedback and analyzing eye movement data. For example, we may find that users are more sensitive to brightness changes when viewing certain types of content (such as reading long texts). Based on this finding, we will use wider transition areas and smoother interpolation curves in the display of such content, which not only improves display quality but also effectively saves energy consumption.

[0109] Through fine parameter control and smooth transition, the overall power consumption of the screen can be reduced while ensuring an excellent visual experience.

[0110] Furthermore, the prediction of the display content in a future period of time in S6 and the optimization of the partition and parameters in advance specifically include:

[0111] The acquired historical display content data is cleaned and preprocessed to remove invalid and abnormal data to obtain a normalized historical data set. According to the characteristics of the historical data set, appropriate time series prediction models such as ARIMA and LSTM are selected to build a prediction model. The model is trained and optimized using methods such as cross-validation to improve the accuracy of the prediction.

[0112] During the model training process, the cross-validation method is used to evaluate the model performance. Assuming that there is one year of historical data, the data of the first 11 months can be used for training, and the data of the last month can be used for testing. By adjusting the model parameters, such as the number of hidden layers of LSTM, the model is continuously optimized until a satisfactory prediction accuracy is achieved on the test set;

[0113] Use the trained prediction model to predict the display content in the future to obtain the predicted content data set, and perform confidence analysis on the prediction results to determine the reliability of the prediction. According to the predicted content data set, analyze the characteristics and distribution of the content, use clustering algorithms to group the content, obtain content clusters with different characteristics, and determine the priority of different clusters according to the size and access frequency of the content clusters. Design corresponding storage partitions and index strategies for different content clusters, load the predicted content data into the corresponding partitions in advance, and optimize index parameters to improve data retrieval and access efficiency;

[0114] When actually displaying content, quickly locate the corresponding data partition and index according to the content characteristics of the request, directly obtain the pre-processed content data, reduce the overhead of real-time calculation and retrieval, and improve the response speed;

[0115] Continuously monitor the operating status and performance indicators of the system, collect user access logs and feedback data, regularly evaluate and optimize the prediction model and partitioning strategy, continuously improve the prediction accuracy and service quality, and ensure the stable operation of the system.

[0116] In this step, the content that users may browse in the next week can be predicted. The prediction results may show that users tend to read news on weekday mornings, but prefer to watch entertainment videos on weekends. The confidence analysis of these prediction results shows that, for example, the prediction confidence of weekday behavior patterns may be as high as 90%, while the prediction confidence of weekend behavior may be only 70%. Based on the prediction results, the clustering algorithm K-means is used to group the content. The content clusters that may be formed include "morning news", "work-related documents", "entertainment videos", etc. The priority is determined according to the size of the cluster and the access frequency. For example, "morning news" may be given the highest priority; for different content clusters, the corresponding storage strategy is designed, for example, "morning news" is stored in the cache, and "entertainment videos" are stored in a slower but larger storage device;

[0117] At the same time, we optimize index parameters, such as establishing a multi-dimensional index based on time and subject for news content to improve retrieval efficiency. In the actual display process, when users turn on their devices in the morning, they can quickly locate the "Morning News" partition and directly load the pre-processed news content, which greatly reduces the loading time. This predictive loading not only improves the response speed, but also optimizes the use of network bandwidth, because most of the content has been pre-loaded during non-peak hours.

[0118] Continuously monitor the running status, collect performance indicators such as page loading time and user stay time, and analyze user feedback behavior. Use this data to regularly optimize the prediction model and partitioning strategy. For example, if it is found that users begin to frequently visit a new topic, the prediction model will be quickly adjusted to include related content in the high-priority loading range. Through this dynamic optimization mechanism, the prediction accuracy and service quality are continuously improved. In actual applications, this set of technologies can reduce content loading time by more than 50%, significantly improve user experience, and effectively reduce server load and network pressure.

[0119] The specific embodiments of the invention are described in detail above, but they are only examples. The present invention is not limited to the specific embodiments described above. Those skilled in the art should understand that the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for partitioning an LED display screen, characterized in that: The following steps are involved: S1. Obtaining content information to be displayed, extracting and classifying features of the displayed content, and obtaining display feature parameters of different areas; S2. Adaptively partition the display screen according to display feature parameters and hardware resource parameters, dynamically generate a partition grid, and determine the position, size, and shape of each partition; S3. For each partition, calculate the optimal display parameter according to its display characteristic parameter; S4. During the display process, continuously monitor the changes in the displayed content to determine whether re-partitioning and parameter optimization are required. If significant changes are detected, partition adjustment and parameter update are triggered; S5. At the partition boundary, interpolate and transition the parameters of adjacent partitions; S6. Predict the display content in the future and optimize the partitions and parameters in advance.

2. A method for partitioning an LED display screen according to claim 1, characterized in that: In S1, feature extraction and classification of the displayed content are performed, specifically: The multi-dimensional attribute information of the content to be displayed is obtained, including the content type, size, position and change frequency. The content analysis algorithm is used to extract the features of the display content. According to the extracted content features, the clustering algorithm is used to classify the display content to obtain the content classification results of different display areas. For each display area, the decision tree algorithm is used to generate the display feature parameters of the area according to the type, size, position and change frequency attributes of the display content in the area.

3. A method for partitioning an LED display screen according to claim 2, characterized in that: The feature extraction and classification of the displayed content in S1 further includes: For display areas with high frequency of changes, more content attribute samples are obtained by increasing the sampling frequency to improve the accuracy of feature parameters. For display areas with complex content types and large feature differences, the accuracy of classification and parameter generation is improved by subdividing the content types and using multiple feature extraction algorithms. According to the spatial adjacency relationship of the display areas, the area division results are optimized through image segmentation algorithms to reduce regional fragmentation and improve the continuity of display feature parameters. The feature parameters of each display area are combined to generate complete display screen parameters to guide the generation and optimization of the display screen.

4. A method for partitioning an LED display screen according to claim 1, characterized in that: In S2, the display screen is adaptively partitioned, and a partition optimization algorithm is used to adaptively partition the display screen according to display feature parameters, hardware resource parameters, initial parameters and constraints, and the display screen is divided into multiple partitions. For each partition, a partition grid is dynamically generated according to the partition optimization algorithm, and the number of rows and columns of the partition grid and the position coordinates of each grid unit are determined. According to the number of rows and columns of the partition grid and the position coordinates of each grid unit, the position, size and shape of each partition are calculated to obtain a partition layout scheme; Obtain the content to be displayed, and map it to the corresponding partition according to the partition layout plan to achieve adaptive display of the content in the partition. When the displayed content changes, re-obtain the display feature parameters and hardware resource parameters to determine whether the partition layout needs to be adjusted. If adjustment is required, re-execute the partition optimization algorithm to adjust the partition grid and partition layout in real time according to the changes in the displayed content and the dynamic changes in hardware resources.

5. A method for partitioning an LED display screen according to claim 1, characterized in that: The step of calculating the optimal display parameters in S3 includes: S301, obtaining characteristic parameters of each display partition and using them as input of a parameter optimization algorithm; S302, establishing a mapping relationship between display parameters and quality according to a preset display quality evaluation model, and obtaining an objective function for parameter optimization; S303, using a particle swarm optimization algorithm, taking the characteristic parameters of the display partition as constraints, minimizing the objective function, and searching for the optimal display parameter combination; S304, during the optimization process, the particle swarm is continuously updated iteratively by dynamically adjusting the speed and position of the particles until a preset convergence condition is met or a preset maximum number of iterations is reached; S305, selecting a particle with the smallest objective function value from the converged particle swarm, and setting the corresponding display parameter as the optimal parameter setting for the partition; S306, repeating steps S303 to S305 for all display partitions until an optimal display parameter combination for each partition is obtained; S307: Apply the optimized display parameters to the display control of each partition.

6. A method for partitioning an LED display screen according to claim 1, characterized in that: The process of continuously monitoring the change of the displayed content during the display process in S4 is as follows: obtaining the current displayed content of the display device, dividing the displayed content into a plurality of display partitions, extracting display features for each display partition, comparing the currently extracted display features with preset historical display features, and determining the degree of change between the current displayed content and the historical displayed content through a feature similarity judgment algorithm; If the degree of change exceeds the preset significant change threshold, the redivision of the display partitions is triggered. According to the changing characteristics of the displayed content, the number and size of the display partitions are dynamically adjusted using an adaptive partitioning algorithm.

7. A method for partitioning an LED display screen according to claim 6, characterized in that: The determination of whether re-partitioning and parameter optimization are required in S4 is specifically as follows: For each re-divided display partition, the updated display features are extracted, and the display parameters of each display partition are determined by the feature analysis algorithm. According to the change range of the display parameters, the progressive parameter optimization algorithm is used to gradually adjust the display parameters to the target values ​​within multiple display cycles. During the display parameter adjustment process, the changes in the display content are continuously monitored. If significant changes are detected during the adjustment process, the current parameter adjustment process is interrupted and the display partition division and parameter optimization are re-performed; If the display content still changes significantly after the display parameter adjustment is completed, the significant change threshold is reduced to increase the sensitivity of change detection. At the same time, the current display features are updated to the historical display feature library as a reference for subsequent change detection.

8. A method for partitioning an LED display screen according to claim 1, characterized in that: In S5, the parameters of adjacent partitions are interpolated and transitioned at the partition boundaries, specifically: Obtain the location coordinate information of the partition boundary, determine the adjacent partitions that need to be smoothly transitioned, perform transition processing on the parameters of the adjacent partitions, calculate the interpolation coefficients based on the location coordinates of the partition boundary, and perform weighted averaging on the parameters of the adjacent partitions based on the interpolation coefficients to obtain the parameter values ​​of the transition area and achieve smooth transition of the parameters.

9. A method for partitioning an LED display screen according to claim 8, characterized in that: The step of interpolating and transitioning the parameters of adjacent partitions at the partition boundary also includes: When switching between partitions, determine whether the current display area is in the transition area. If it is in the transition area, use the parameters of the transition area for display. Render and display the image according to the parameter values ​​of the transition area to avoid abruptness and flickering when switching between partitions and improve display continuity. Adopting adaptive weight adjustment algorithm, the interpolation coefficient and the range of transition area are dynamically adjusted according to the size of transition area and the frequency of partition switching; Through the iterative optimization algorithm, the interpolation coefficients and parameters of the transition area are continuously optimized to minimize the visual differences when switching between partitions and improve the consistency of display.

10. The LED display screen partitioning method according to claim 1, characterized in that: The prediction of the display content in the future and the optimization of the partition and parameters in advance in S6 specifically include: The acquired historical display content data is cleaned and preprocessed to remove invalid and abnormal data to obtain a normalized historical data set. According to the characteristics of the historical data set, appropriate time series prediction models, including ARIMA and LSTM, are selected to build a prediction model, and the cross-validation method is used to train and optimize the model. Use the trained prediction model to predict the display content in the future to obtain the predicted content data set, and perform confidence analysis on the prediction results to determine the reliability of the prediction. According to the predicted content data set, analyze the characteristics and distribution of the content, use the clustering algorithm to group the content, obtain content clusters with different characteristics, and determine the priority of different clusters according to the size and access frequency of the content clusters. Design corresponding storage partitions and index strategies for different content clusters, load the predicted content data into the corresponding partitions in advance, and optimize the index parameters; When actually displaying the content, quickly locate the corresponding data partition and index according to the content characteristics of the request, and directly obtain the pre-processed content data; Continuously monitor the operating status and performance indicators, collect user access logs and feedback data, and regularly evaluate and optimize prediction models and partitioning strategies.

Images