Intelligent driving method and system of liquid crystal display screen

Abstract

The invention relates to the technical field of liquid crystal display, and discloses an intelligent driving method and system for a liquid crystal display screen, and the method comprises the following steps: S1, collecting environment, content and user behavior information, and constructing a multi-modal fusion feature vector; s2, extracting features by using a physical enhanced neural network, and generating visual importance mapping in combination with a liquid crystal response model; s3, dividing a display area based on visual importance and a content change rate, and calculating a refresh priority and a refresh rate; s4, determining an optimal driving parameter in combination with a multi-objective optimization model; s5, executing driving control and collecting response data; and S6, updating the liquid crystal health map and network parameters to realize closed-loop optimization. According to the method, a multi-modal perception and fusion feature vector construction technology is adopted, environment information, display content information and user behavior data can be comprehensively collected, and high-dimensional feature vectors are generated by reasonably fusing the information.

Description

Technical Field

[0001] The present invention relates to the field of liquid crystal display technology, and in particular to an intelligent driving method and system for a liquid crystal display screen. Background Art

[0002] With the widespread adoption of electronic devices, LCD screens have become an integral part of daily life. Whether in mobile phones, televisions, computers, or other smart devices, LCD screens play a crucial role in user interaction. However, with the diversification of application requirements, traditional LCD screens often struggle to meet the demands of diverse environments. In particular, under the influence of factors such as lighting, temperature, viewing angle, and user behavior, LCD display performance often falls short of expectations. Therefore, improving the adaptability and display quality of LCD screens in these changing environments has become a pressing technical challenge.

[0003] Existing LCD display driver systems primarily rely on static display drive modes, typically adjusting display parameters based on the brightness and contrast of the displayed content to ensure optimal display quality under certain conditions. For example, conventional adaptive brightness adjustment can adjust screen brightness under varying lighting conditions, ensuring users can clearly see the screen content both indoors and outdoors. Furthermore, LCD display response optimization technology typically adjusts the display panel's voltage, refresh rate, and other parameters to achieve a certain degree of display quality improvement and meet users' basic visual needs.

[0004] Although the existing technology can provide reasonable display effects in some basic scenarios, its performance is obviously insufficient under the dynamic changes of the environment and user behavior. First, the driving method of the traditional LCD screen ignores the impact of environmental information and user behavior on the display effect, and cannot achieve personalized adjustment of the display effect. In the existing technology, most display systems fail to dynamically adjust the display quality according to information such as the user's gaze area, viewing angle or dwell time, resulting in inaccurate display effects or waste of resources; secondly, the existing response optimization method usually does not fully combine the physical response characteristics of the liquid crystal, such as the impact of temperature and voltage changes on the liquid crystal pixels. Therefore, under different usage conditions, the response delay and brightness change of the display screen will lead to a poor visual experience; thirdly, the refresh rate control in the existing technology is mostly statically allocated, and the refresh rate cannot be dynamically adjusted according to the changes in the displayed content, resulting in power consumption waste; finally, the existing feedback mechanism is often unable to perform adaptive optimization based on real-time response data. After long-term use, the display effect will gradually decline. To this end, those skilled in the art propose an intelligent driving method and system for LCD screens to solve the above problems. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the present invention provides an intelligent driving method and system for a liquid crystal display screen, which solves the problem in the existing technology that the liquid crystal display screen cannot dynamically optimize the display effect according to environmental changes and user behavior.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent driving method for a liquid crystal display screen, comprising the following steps:

[0007] S1, collect environmental information, display content information and user behavior information through sensors, and construct a multimodal fusion feature vector;

[0008] S2. Based on the multimodal fused feature vector, a physical information enhanced neural network is used to extract spatial features and temporal features, and a visual importance map is generated in combination with a liquid crystal physical response model;

[0009] S3. Divide the display area according to the visual importance mapping and the content change rate, and calculate the refresh priority and refresh rate of each area;

[0010] S4, determining optimal driving parameters according to the refresh priority, computing resource status, and a multi-objective optimization model of display quality, power consumption, and response speed;

[0011] S5. Execute the optimal driving parameters to control the pixel driving circuit of the liquid crystal display to output a driving signal and collect liquid crystal response data;

[0012] S6. Update the liquid crystal health status map and neural network parameters according to the response data to achieve closed-loop optimization.

[0013] Preferably, the step S1 includes:

[0014] Collecting environmental information, including ambient light intensity, display panel temperature, and device attitude angle collected by an ambient light sensor, a temperature sensor, and an attitude sensor, respectively;

[0015] Acquire display content information, where the display content information is an image pixel matrix of a current display frame;

[0016] Collecting user behavior information, including user gaze area, viewing angle, and dwell time;

[0017] Performing feature extraction and normalization processing on the environmental information, display content information, and user behavior information;

[0018] The normalized features are combined to construct a multimodal fusion feature vector, which serves as the input data for intelligent driving.

[0019] Preferably, the step S2 includes:

[0020] Extract the spatial texture and edge features of the current frame image based on the convolutional neural network;

[0021] Perform temporal modeling of continuous frame sequences based on recurrent neural networks or Transformer architectures to extract dynamic change trends;

[0022] Construct physical auxiliary features based on liquid crystal response speed, twist angle, and voltage and temperature relationships;

[0023] Fuse spatial features, temporal features and physical auxiliary features;

[0024] The visual importance map is calculated through a fully connected mapping network to represent the contribution of each pixel to human eye perception.

[0025] Preferably, step S3 includes:

[0026] Divide the display area into fixed grids or adaptive regions based on clustering results;

[0027] Calculate the mean visual importance, historical frame change rate and regional display content complexity in each region respectively;

[0028] Determine the refresh priority of each area based on the weighted calculation model;

[0029] Assign corresponding refresh rates to each area based on refresh priority and system resource limitations;

[0030] Output the area refresh rate as one of the inputs for subsequent multi-objective optimization.

[0031] Preferably, step S4 includes:

[0032] Construct a multi-objective optimization function, with the optimization objectives including maximizing display quality, minimizing power consumption, and optimizing response speed;

[0033] The regional refresh rate, compensation voltage, and pixel state are taken as optimizable variables;

[0034] Set hardware load, voltage limit, and display stability as constraints;

[0035] Use weighted sum method or Pareto frontier method to search for optimal parameters;

[0036] Output the driving parameter set with the best comprehensive performance under the constraints.

[0037] Preferably, step S5 includes:

[0038] Mapping the optimal driving parameters into pixel-level driving signals;

[0039] The voltage change of the TFT liquid crystal pixel unit is controlled by the voltage driving module;

[0040] Synchronously collect liquid crystal response data during each driving cycle, including response delay, brightness change and pixel stability;

[0041] The response data is passed to the subsequent feedback module for status update.

[0042] Preferably, step S6 includes:

[0043] Based on the collected response data, the current health status of each pixel is modeled and corrected;

[0044] Updated LCD health status map to reflect pixel fatigue, electromigration and response deviation;

[0045] Retrain or fine-tune the neural network using the updated health map as additional input;

[0046] Adjust visual importance mapping parameters and physical modeling structure;

[0047] Realize closed-loop optimization and dynamic adaptation of the drive system.

[0048] Preferably, the physical information enhanced neural network adopts a lightweight and variable architecture, supporting the division of model depth and complexity by region, specifically including:

[0049] Using deep network calculations for areas of high visual importance;

[0050] Use pruned or quantized small networks for areas of low visual attention;

[0051] The model inference process uses parallel processing of heterogeneous computing units;

[0052] Improve the inference efficiency of repeated areas through caching mechanism.

[0053] Preferably, the physical response model is processed by combining historical voltage drive data and temperature response curves through the following processing steps:

[0054] Construct a pixel-level dynamic model based on liquid crystal torsional torque and applied electric field;

[0055] The actual response time of the fusion collection is calibrated and corrected;

[0056] The modeling results are used to assist the neural network in predicting nonlinear responses;

[0057] A mapping relationship is established between the response compensation result and the driving signal.

[0058] An intelligent driving system for a liquid crystal display screen, comprising:

[0059] Multimodal information acquisition module, used to obtain environmental information, display content information and user behavior information, and construct a fusion feature vector;

[0060] The neural network inference module is used to receive the fused feature vector and extract the spatial and temporal features, while fusing the physical modeling results to generate a visual importance map;

[0061] The area division and refresh control module is used to divide the display area according to the visual importance mapping and content change rate, and calculate the refresh priority and refresh rate;

[0062] A multi-objective optimization module, which is used to build an optimization function and determine the optimal driving parameters based on display quality, power consumption, and response speed;

[0063] A driving execution module, configured to generate a driving signal according to the optimal driving parameters to control the display behavior of the liquid crystal pixel unit;

[0064] The closed-loop feedback module is used to collect liquid crystal response data and update the liquid crystal health status map and neural network model parameters.

[0065] The present invention provides an intelligent driving method and system for a liquid crystal display screen, which has the following beneficial effects:

[0066] 1. This invention utilizes multimodal perception and fusion feature vector construction technology to comprehensively collect environmental information, display content information, and user behavior data, and rationally fuses this information to generate high-dimensional feature vectors. This technical solution enables the LCD display's driver to precisely adapt to different environments and usage scenarios. Compared to the traditional single input mode in existing technologies, it solves the problem of slow response to changes in the environment and user behavior, significantly improving the personalized and intelligent display effect.

[0067] 2. The present invention extracts spatial and temporal features through a physical information enhanced neural network (PE-NN) and generates a visual importance map (VIMM) in combination with a liquid crystal physical response model, thereby achieving a deep fusion of image content and the physical characteristics of liquid crystal display. This technical solution effectively improves the visual effect and response accuracy of liquid crystal display, solves the problem of neglecting physical response and dynamic display characteristics in the existing technology, and ensures a high degree of match between display content and device performance.

[0068] 3. The present invention introduces a region division and refresh rate optimization mechanism. Based on visual importance mapping, content change rate and liquid crystal response characteristics, it rationally allocates the refresh rate of the display area, optimizes display performance and energy efficiency. This solution enables the LCD screen to still operate efficiently under high load and complex scenarios, solves the problems of inaccurate refresh rate control and unreasonable resource allocation in the existing technology, and improves the smoothness and energy efficiency of the display effect.

[0069] 4. The present invention uses a closed-loop feedback mechanism to update the liquid crystal health status map based on real-time response data and fine-tune the neural network parameters to achieve dynamic adaptation of the drive system. This technical solution enables the LCD screen to maintain a stable display effect during long-term operation and adjust the display parameters in a timely manner, avoiding the problems of display degradation or response lag in the existing technology, and significantly improving the long-term stability and reliability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 Schematic diagram of the method flow of the present invention;

[0071] Figure 2 Schematic diagram of the system architecture of the present invention. DETAILED DESCRIPTION

[0072] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments 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 making creative efforts are within the scope of protection of the present invention.

[0073] Please see the attached Figure 1 , an embodiment of the present invention provides an intelligent driving method for a liquid crystal display screen, comprising the following steps:

[0074] S1, collect environmental information, display content information and user behavior information through sensors, and construct a multimodal fusion feature vector;

[0075] Specifically, in this embodiment, step S1 involves collecting information about the LCD screen's external environment, display content, and user behavior through various sensors to construct a multimodal fusion feature vector. This feature vector serves as input for intelligent driving, and subsequent decision-making and optimization steps are based on this data. The collection of this information helps achieve personalized display driving based on the environment and user behavior.

[0076] Generally speaking, environmental information, display content information, and user behavior information are intertwined and correlated. Therefore, accurately acquiring and processing this information and integrating it appropriately are crucial for subsequent display optimization. In this step, by integrating this information, the LCD screen is provided with more intelligent and dynamic display adjustment capabilities.

[0077] Environmental information collection: In one possible implementation, environmental information collection typically includes data from the following sensors:

[0078] Ambient light sensor: Used to collect the light intensity in the current environment and help adjust the brightness and contrast of the display to cope with different ambient light conditions.

[0079] Temperature sensor: used to monitor the temperature changes of the display panel. The performance of the LCD screen will vary under different temperature conditions, so the input of the temperature sensor is particularly important for optimizing the display effect and energy saving.

[0080] Attitude sensor: This sensor is used to obtain the device's attitude angles, including pitch, roll, and yaw. This information can be used to adjust the orientation of the screen display, especially on devices with auto-rotating screens, to provide more accurate visual effects.

[0081] Display content information acquisition: Display content information primarily refers to the image pixel matrix of the current display frame. Each frame is formed by the arrangement of the pixel array, with varying brightness and color distribution. To enable intelligent driving, it is necessary to acquire data from each frame in real time and extract the characteristic information of each pixel using image processing algorithms.

[0082] In some embodiments, the acquired display content information may also include data such as content complexity, display change rate, etc. These data will help determine which areas require a higher refresh rate or a stronger drive signal.

[0083] User behavior information collection: The collection of user behavior information includes the following aspects:

[0084] User Gaze Area: Eye tracking technology is used to determine the screen area where the user is currently looking, which is crucial for dynamic optimization of the display area. This information helps the system identify the user's focus and provide better display quality in high-attention areas.

[0085] Viewing angle: The relative angle between the user and the screen significantly affects visual perception. By understanding the user's viewing angle, the system can adjust the screen display to ensure the best viewing experience.

[0086] Dwell time: The time a user spends in a certain area can also reflect their attention to that area. Longer dwell time means that the area may require higher display quality and therefore needs to be prioritized.

[0087] Data processing and normalization: After collecting the various types of data mentioned above, feature extraction and normalization are required. Specifically, multi-dimensional data such as ambient light, temperature, posture, and gaze area can be normalized to bring data of different scales into the same dimension.

[0088] For example, the outputs of ambient light intensity and temperature sensors may have different dimensions, so they need to be mapped to a unified scale based on a preset standard range for subsequent processing and analysis.

[0089] In one embodiment, the normalized features may be processed as follows:

[0090]

[0091] Where: X represents the original data; μ is the mean of the data; σ is the standard deviation; The data are normalized.

[0092] Through the above method, the dimensional differences between different sensor data can be eliminated and converted into a unified feature vector, which is convenient for subsequent fusion and analysis.

[0093] Multimodal fusion feature vector construction: After normalization, all data from environmental information, display content information, and user behavior information are combined to form a multimodal fusion feature vector. This vector incorporates information from different data sources and can provide a more comprehensive input for subsequent neural network inference.

[0094] Specifically, this information can be arranged according to specific rules to form a high-dimensional feature vector. For example, environmental information can occupy the front part of the feature vector, display content information can occupy the middle part, and user behavior information can occupy the back part. In this case, the fused feature vector can be expressed as:

[0095] F=[F env ,F content ,F user ];

[0096] Among them: F env is the environmental information vector; F content is the display content information vector; F user is the user behavior information vector.

[0097] The ultimate goal of this step is to provide a comprehensive input feature vector for subsequent steps. In this embodiment, the various types of information collected by the sensors are processed to form a multimodal feature vector, which serves as the input to the neural network and is passed to the subsequent neural network inference module for further spatial and temporal feature extraction. The quality and accuracy of this feature vector directly influences the accuracy of subsequent optimization algorithms and driving parameters, thereby improving the intelligent driving effect of the LCD display.

[0098] S2. Based on the multimodal fused feature vector, a physical information enhanced neural network is used to extract spatial features and temporal features, and a visual importance map is generated in combination with a liquid crystal physical response model;

[0099] Specifically, in this embodiment, step S2 introduces a physical information enhanced neural network module (PE-NN), combining deep neural networks with liquid crystal physical modeling to extract the spatial and temporal features of the image and the physical response characteristics of the liquid crystal, thereby generating a visual importance map (VIMM). This map will provide effective support for subsequent refresh rate control and driving strategies, further improving the intelligent driving effect of the LCD display.

[0100] First, based on the image frame data contained in the fused feature vector, a convolutional neural network (CNN) is used to extract the spatial structure features of the current frame.

[0101] Spatial feature extraction: Let the current frame image be Where w and h represent the width and height of the image respectively, and the channel dimension is RGB three channels.

[0102] In one possible implementation, the spatial feature extraction function can be expressed as:

[0103] F spatial =f CNN (C);

[0104] Where: f CNN (·) represents a deep feature extraction network consisting of multiple convolutional layers, activation layers, and pooling layers. Output features Where d represents the number of channels after extraction, and w′ and h′ are the spatial sizes of the feature map after downsampling.

[0105] Temporal feature extraction: In order to obtain the dynamic change trend of the displayed content in the time dimension, the system also introduces a recurrent neural network structure to model continuous frames.

[0106] In some embodiments, the time series modeling can adopt a standard RNN structure, or can adopt an LSTM or GRU network to enhance the ability to capture long-term dependencies.

[0107] Assume that the continuous input frame sequence is Where T is the sequence length, the temporal feature extraction expression is:

[0108]

[0109] in: Represents the high-level feature representation of each frame image in the time dimension; f RNN (·) represents the neural network function used to model temporal dependencies.

[0110] Modeling of liquid crystal physical response characteristics: Based on the above, in order to further improve the model's adaptability to the response characteristics of liquid crystal entities, the system introduces liquid crystal torsion dynamics modeling and response time modeling as physical auxiliary modules.

[0111] The response behavior of a liquid crystal pixel can usually be modeled by the following kinetic equation:

[0112]

[0113] Where: θ represents the twist angle of the liquid crystal molecules; E is the external electric field strength; φ is the pretilt angle; K1 and K2 are the elastic constant and electro-optic coupling constant of the liquid crystal material, respectively.

[0114] In addition, to model the effect of driving conditions (such as temperature, applied voltage, etc.) on the response time of the liquid crystal, the following empirical model is introduced:

[0115]

[0116] Where: τ represents the response time; T is the display panel temperature; V is the voltage value; p is the pixel position or state parameter; α, β, γ are fitting constants; and the function f(p) is used to describe the correction of spatial non-uniformity to the response.

[0117] The results of these physical models will be encoded as auxiliary features and fused with the features extracted by the neural network to enhance the accuracy and physical consistency of perception.

[0118] Feature fusion: In one possible implementation, spatial features, temporal features, and response time features can be jointly input into a fusion network to construct the following fusion function:

[0119] F phys =Φ(F spatial ,F temporal ,τ);

[0120] Where: Φ(·) represents the fusion module, which is usually a set of fully connected layers or cross-attention layers.

[0121] Visual Importance Mapping Generation: The final output physical enhancement features will be used to construct a Visual Importance Mapping (VIMM), which is used to measure the importance of different display areas to human eye perception:

[0122] VIMM(x,y)=σ(W·F phys (x,y)+b);

[0123] Where: σ(·) represents the Sigmoid or Softmax function; W is the trainable weight matrix; b is the bias term; Fphys (x,y) represents the fusion feature of the (x,y)th pixel.

[0124] Improved model inference efficiency: As an extension, in some embodiments, this module can employ a variable-structure neural network to improve model inference efficiency and computing resource utilization. In this architecture, regions of high visual importance utilize deep network pathways, while regions of low visual importance utilize pruned or quantized lightweight networks, achieving a balance between heterogeneous computing and energy consumption.

[0125] In addition, to speed up the inference process, the model can cooperate with the caching mechanism to reuse intermediate feature representations in repeated areas, avoid redundant calculations, and improve processing efficiency.

[0126] S3. Divide the display area according to the visual importance mapping and the content change rate, and calculate the refresh priority and refresh rate of each area;

[0127] Specifically, in this embodiment, step S3, based on the visual importance map (VIMM) obtained in step S2, further introduces a region division mechanism to spatially reconstruct the entire frame image according to importance level to support differentiated refresh rate control. This region division not only needs to consider the visual weight distribution at the pixel level, but also needs to be dynamically adjusted based on multiple factors such as liquid crystal response characteristics, the continuity of the user's gaze area, and the inter-frame change rate, thereby providing basic support for subsequent refresh rate resource allocation and drive parameter configuration.

[0128] This process takes over the output of step S2 and passes the result directly to the refresh rate control unit, which is the intermediate bridge link in the entire dynamic drive framework.

[0129] Normalization and classification of visual importance maps: First, based on the visual importance map VIMM(x,y) obtained in step S2, importance classification is performed on the entire frame. Generally, the visual importance values are normalized and multiple threshold intervals are set for classification, for example:

[0130]

[0131] in: represents the normalized importance value; v min With v max Respectively represent the minimum and maximum values of VIMM in the frame.

[0132] Region segmentation: In one possible implementation, based on the normalized visual importance The entire frame image is divided into high attention region (HAR), medium attention region (MAR) and low attention region (LAR).

[0133] The specific classification criteria are as follows:

[0134] like Then the pixel belongs to HAR;

[0135] like It belongs to MAR;

[0136] like It belongs to LAR.

[0137] Where: θ h and θ l It is an empirical threshold, which is generally between [0.6, 0.9] and [0.2, 0.4]. The specific value can be set according to the experiment or adjusted dynamically.

[0138] Region merging: To reduce the fragmentation of the segmented regions, an aggregated spatial fusion algorithm is used to enhance the connectivity of the initial region partitioning results. In some embodiments, a method based on connected component growth (CCG) is used to merge consecutive pixel blocks with the same importance level to generate several closed sub-regions.

[0139] Let each sub-region be denoted as R k , where k is the region number, then the final frame partition set can be recorded as:

[0140] S={R1,R2,…,R K};

[0141] Among them: K represents the total number of divided areas, Ω is the pixel set of the image frame, satisfying:

[0142] Regional change rate modeling and refresh priority: To further improve the spatial efficiency of refresh rate control, it is necessary to statistically model the frequency of inter-frame changes between regions. In one implementation, the regional change rate function ρ is introduced. k , defined as follows:

[0143]

[0144] Where: C t (x,y) represents the pixel value of the current frame, C t-1 (x,y) is the corresponding pixel value of the previous frame, |Rk | is the number of pixels in the region, ρ k Reflects the degree of dynamic change of image content in the area.

[0145] Regional refresh priority function: In order to optimize according to the visual importance, dynamic changes and liquid crystal response characteristics of the region, the liquid crystal response time model (see step S2) can also be embedded in the regional evaluation function to weight the refresh priority of different regions. Let the average response time corresponding to each region be Then refresh the priority function π k The definition is as follows:

[0146]

[0147] in: is the average visual importance value in the area; ω1, ω2, ω3 are weighting coefficients, satisfying ω1+ω2+ω3=1; each parameter can be determined according to the pre-strategy or optimization algorithm.

[0148] Refresh rate allocation and resource scheduling: refresh priority according to region π k An appropriate refresh rate is assigned to each region, and overall refresh rate scheduling is optimized based on system resource availability. In some embodiments, to improve real-time performance and stability, the region division strategy is executed every N frames, and interpolation is used to map and predict intermediate frames to reduce computational overhead. Furthermore, region information is cached in the driver chip and transmitted to the control unit in a compressed format, ensuring real-time refresh scheduling and balanced power consumption.

[0149] S4, determining optimal driving parameters according to the refresh priority, computing resource status, and a multi-objective optimization model of display quality, power consumption, and response speed;

[0150] Specifically, after completing the visual importance mapping and region segmentation process in step S3, the system has obtained the visual weight and dynamic change information for each region. Step S4 further determines the optimal driver parameters using a multi-objective optimization function, combining multiple optimization objectives such as display quality, power consumption, and response speed. This step provides efficient and balanced parameters for subsequent driver execution, ensuring the optimal balance between display quality, energy efficiency, and response speed.

[0151] In this embodiment, a multi-objective optimization function is first constructed, with the following objectives: maximizing display quality, minimizing power consumption, and optimizing response speed. To effectively optimize these objectives, the system uses the refresh rate, compensation voltage, and pixel state of each region as optimizable variables. Furthermore, the system also sets constraints such as hardware load, voltage limits, and display stability to ensure that the optimized drive parameters meet the actual hardware and performance requirements.

[0152] Multi-objective optimization model construction: In general, multi-objective optimization problems involve multiple competing objectives that need to be balanced through a reasonable weighting strategy. In this embodiment, the constructed optimization function includes the following main objectives:

[0153] Display quality: Display quality Q(x) can be measured by image quality metrics, usually based on parameters such as visual quality and color accuracy. The expression of the display quality objective function is:

[0154] Q(x)=Q VEM (x);

[0155] Where: Q VEM (x) represents Visual Effect Measurement, which is used to evaluate the display quality of the image.

[0156] Power consumption: Power consumption P(x) is closely related to the display voltage and refresh rate of the LCD screen. In order to minimize power consumption, the objective function needs to consider the voltage compensation, refresh rate and area of each area. The expression of the power consumption objective function is:

[0157]

[0158] Where: V comp (x i ) is the area x i Compensation voltage, R(x i ) is the area x i The refresh rate, A(x i ) is the area x i display area.

[0159] The objective function reflects the electricity consumption of the region.

[0160] Response speed: Response speed S(x) is closely related to the response time of the liquid crystal pixel. To improve the response speed, it is necessary to minimize the response delay of each area. The expression of the response speed objective function is:

[0161] S(x)=max(τ(x));

[0162] Where: τ(x) is the response time of region x, which represents the time it takes for a pixel to switch from one state to another.

[0163] Constraints: When performing multi-objective optimization, you also need to consider various hardware and physical constraints, such as hardware load, voltage limits, and display stability. Typically, these constraints can be expressed as:

[0164] g j (x)≤0,hk (x) = 0;

[0165] Where: g j (x) represents the inequality constraint, h k (x) represents an equality constraint, which ensures that the optimized solution meets the hardware and physical constraints.

[0166] Weighted Sum and Pareto Frontier Methods: To balance multiple objectives, the weighted sum method or Pareto frontier method is generally used to search for optimal parameters. The weighted sum method combines all objective functions into a single objective function by assigning a weight coefficient to each objective. The optimization process can be solved using standard optimization algorithms. The objective function of the weighted sum method can be expressed as:

[0167] F(x)=w1·Q(x)+w2·P(x)+w3·S(x);

[0168] Wherein: w1, w2, w3 are weight coefficients of display quality, power consumption and response speed respectively, satisfying w1+w2+w3=1.

[0169] In another possible implementation, the Pareto frontier method is used to obtain a set of balanced optimal solutions by optimizing non-dominated solutions among multiple objectives.

[0170] Determination of optimal driving parameters: By solving the above multi-objective optimization function, the system will obtain a set of optimal driving parameters. These parameters include the refresh rate, compensation voltage, and pixel status of each area. The final optimization result can be expressed by the following formula:

[0171]

[0172] in: represents the solution space that satisfies all constraints, x * Indicates the final optimal solution. The optimized drive parameter set will be used for subsequent drive execution.

[0173] S5, executing the optimal driving parameters, controlling the pixel driving circuit of the liquid crystal display to output a driving signal, and collecting liquid crystal response data;

[0174] Specifically, after completing the multi-objective optimization and determining the optimal drive parameters in step S4, step S5 is primarily responsible for executing the obtained optimal drive parameters and outputting the corresponding drive signal by controlling the pixel drive circuit of the LCD screen. In addition, the system also synchronously collects the response data of the LCD screen during each drive cycle, including response delay, brightness change, and pixel stability. This response data will be used for subsequent status updates and system optimization. Step S5 is a key link in the drive signal execution and feedback loop, providing the necessary data support for the dynamic adjustment and optimization of the system.

[0175] In this embodiment, the optimal driving parameters obtained in step S4 are first mapped into specific pixel-level driving signals. These driving signals will be actually executed in the liquid crystal driving circuit to control the voltage change of each pixel to achieve the optimized display effect.

[0176] Generation and mapping of optimal drive signals: Based on the optimal drive parameters, the system maps these parameters into specific pixel-level drive signals. Specifically, the drive signal D(x,y) for each pixel can be expressed as follows:

[0177] D(x,y)=f driver (V comp (x,y),R(x,y),A(x,y));

[0178] Where: V comp (x,y) is the compensation voltage of the (x,y)th pixel, R(x,y) is the refresh rate corresponding to the pixel, and A(x,y) is the area displayed by the pixel.

[0179] This driving signal is transmitted to each pixel unit through the liquid crystal driving circuit to control its display state.

[0180] Liquid crystal response data acquisition: While the drive signal is being executed, the system also needs to synchronously collect the response data of the LCD screen. The response data includes the following key indicators:

[0181] Response delay: It refers to the time required from the input of the driving signal to the pixel reaching the target state.

[0182] Brightness changes: Monitor the brightness changes of each pixel to ensure display accuracy and consistency.

[0183] Pixel stability: reflects the stability of each pixel under long-term driving, preventing distortion or discoloration caused by long-term operation.

[0184] Response data R d (x,y,t) can be expressed by the following formula:

[0185] R d(x,y,t)=[ΔL(x,y,t),Δτ(x,y,t),σ(x,y,t)];

[0186] Where: ΔL(x,y,t) is the brightness change of pixel (x,y) at time t, Δτ(x,y,t) is the response delay, and σ(x,y,t) is the standard deviation of pixel stability.

[0187] Data Feedback and Status Update: The collected LCD response data is fed back to the subsequent feedback module to update the LCD health status map and the neural network parameters. This process further enhances the system's adaptive adjustment capabilities.

[0188] The health state map H(x,y) will be updated using the following formula:

[0189] H(x,y)=H previous (x,y)+α·(ΔL(x,y)+Δτ(x,y));

[0190] Where α is the correction coefficient, ΔL(x,y) and Δτ(x,y) represent the changes in brightness and response delay, respectively. By continuously updating the health status map, the system can make real-time corrections for LCD pixel fatigue, electromigration, and response deviation.

[0191] Neural network parameter fine-tuning: As the health state map is updated, the system fine-tunes the model parameters in the neural network based on this data. Specifically, the feedback liquid crystal response data will be used to retrain or fine-tune the neural network to improve the network's prediction accuracy of the liquid crystal response characteristics.

[0192] The fine-tuning process can be expressed by the following formula:

[0193]

[0194] in: represents the loss function; R d (x, y) is the actual collected response data; is the response data predicted by the neural network; θ i is the neural network parameter; λ is the regularization coefficient; θ previous,i are the previous network parameters.

[0195] By minimizing the loss function, the neural network gradually improves its prediction accuracy.

[0196] S6. Update the liquid crystal health status map and neural network parameters according to the response data to achieve closed-loop optimization.

[0197] Specifically, in step S5, the system executes the optimal drive signal and collects response data from the LCD. To further optimize the LCD display and achieve closed-loop adjustments, step S6 primarily processes this response data, updates the LCD health status map, and fine-tunes the neural network parameters. In this way, the system can dynamically adapt and optimize the display based on real-time feedback, ensuring the long-term stability and efficiency of the LCD.

[0198] In this embodiment, the response data is used to build and update a liquid crystal health map. This map reflects the display's health over different life cycles, including pixel fatigue, electromigration issues, and response deviation. Updates to this map provide feedback for subsequent drive signal adjustments and optimizations.

[0199] Health state map modeling and updating: In step S5, the system has collected the response data R of each pixel d (x, y, t), these data are used to reflect the display stability and dynamic changes of the pixel. Based on these response data, the system will model and update the health status of each pixel. The health status map H(x, y) will be adjusted based on response delay, brightness change and pixel stability. The specific update formula is:

[0200] H(x,y)=H previous (x,y)+α·(ΔL(x,y)+Δτ(x,y));

[0201] Among them: H previous (x, y) is the health status of the pixel at the previous time point, ΔL(x, y) and Δτ(x, y) represent the currently acquired brightness change and response delay change, respectively, and α is the correction coefficient used to control the update step size.

[0202] In some embodiments, the health state map can further refine the health state estimate for each pixel by incorporating corrections to the physical model of the liquid crystal response. For example, to account for the effects of temperature variations and long-term electromigration of the LCD panel, the system can use these physical properties as additional input features to improve the accuracy of the health state map.

[0203] Neural Network Parameter Fine-tuning: Based on the health map update, the system will continue to fine-tune the neural network to improve its adaptability to the LCD's response characteristics. By using the health map as an additional input, the system can fine-tune the neural network parameters to better reflect the actual operating status of the LCD.

[0204] Fine-tuning of neural network parameters can be optimized using the following loss function:

[0205]

[0206] in: is the loss function, R d (x,y) is the actual response data collected, is the response data predicted by the neural network, θ i is the parameter of the neural network, λ is the regularization coefficient, θ previous,i By minimizing the loss function, the neural network can gradually adjust its parameters to make the response prediction more consistent with the actual liquid crystal response data.

[0207] Closed-Loop Optimization and Dynamic Adaptation: The system achieves closed-loop optimization by updating the health state map and fine-tuning the neural network parameters. Whenever new response data is collected, the system updates the health state map in real time and adjusts the drive parameters based on the updated map. This dynamic adaptability ensures that the LCD screen maintains optimal performance under varying environments and usage conditions.

[0208] In some embodiments, the system can further optimize the structure of the neural network based on long-term feedback data. For example, a deeper network structure can be used to process areas of high visual importance, or pruning and quantization techniques can be used to reduce the computational burden of areas of low visual attention.

[0209] The intelligent driving system of a liquid crystal display described below and the intelligent driving method of a liquid crystal display described above can refer to each other.

[0210] Please see the attached Figure 2 , an intelligent driving system for a liquid crystal display, comprising:

[0211] Multimodal information acquisition module, used to obtain environmental information, display content information and user behavior information, and construct a fusion feature vector;

[0212] The neural network inference module is used to receive the fused feature vector and extract the spatial and temporal features, while fusing the physical modeling results to generate a visual importance map;

[0213] The area division and refresh control module is used to divide the display area according to the visual importance mapping and content change rate, and calculate the refresh priority and refresh rate;

[0214] A multi-objective optimization module, which is used to build an optimization function and determine the optimal driving parameters based on display quality, power consumption, and response speed;

[0215] A driving execution module, configured to generate a driving signal according to the optimal driving parameters to control the display behavior of the liquid crystal pixel unit;

[0216] Closed-loop feedback module, used to collect LCD response data and update LCD health status map and neural network model parameters

[0217] The system of this embodiment can be used to execute the above method embodiments, and its principles and technical effects are similar, so they will not be repeated here.

[0218] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent driving method for a liquid crystal display screen, characterized in that: The following steps are involved: S1, collect environmental information, display content information and user behavior information through sensors, and construct a multimodal fusion feature vector; S2. Based on the multimodal fused feature vector, a physical information enhanced neural network is used to extract spatial features and temporal features, and a visual importance map is generated in combination with a liquid crystal physical response model; S3. Divide the display area according to the visual importance mapping and the content change rate, and calculate the refresh priority and refresh rate of each area; S4, determining optimal driving parameters according to the refresh priority, computing resource status, and a multi-objective optimization model of display quality, power consumption, and response speed; S5. Execute the optimal driving parameters to control the pixel driving circuit of the liquid crystal display to output a driving signal and collect liquid crystal response data; S6. Update the liquid crystal health status map and neural network parameters according to the response data to achieve closed-loop optimization.

2. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The step S1 comprises: Collecting environmental information, including ambient light intensity, display panel temperature, and device attitude angle collected by an ambient light sensor, a temperature sensor, and an attitude sensor, respectively; Acquire display content information, where the display content information is an image pixel matrix of a current display frame; Collecting user behavior information, including user gaze area, viewing angle, and dwell time; Performing feature extraction and normalization processing on the environmental information, display content information, and user behavior information; The normalized features are combined to construct a multimodal fusion feature vector, which serves as the input data for intelligent driving.

3. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The step S2 comprises: Extract the spatial texture and edge features of the current frame image based on the convolutional neural network; Perform temporal modeling of continuous frame sequences based on recurrent neural networks or Transformer architectures to extract dynamic change trends; Construct physical auxiliary features based on liquid crystal response speed, twist angle, and voltage and temperature relationships; Fuse spatial features, temporal features and physical auxiliary features; The visual importance map is calculated through a fully connected mapping network to represent the contribution of each pixel to human eye perception.

4. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The step S3 comprises: Divide the display area into fixed grids or adaptive regions based on clustering results; Calculate the mean visual importance, historical frame change rate and regional display content complexity in each region respectively; Determine the refresh priority of each area based on the weighted calculation model; Assign corresponding refresh rates to each area based on refresh priority and system resource limitations; Output the area refresh rate as one of the inputs for subsequent multi-objective optimization.

5. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The step S4 comprises: Construct a multi-objective optimization function, with the optimization objectives including maximizing display quality, minimizing power consumption, and optimizing response speed; The regional refresh rate, compensation voltage, and pixel state are taken as optimizable variables; Set hardware load, voltage limit, and display stability as constraints; Use weighted sum method or Pareto frontier method to search for optimal parameters; Output the driving parameter set with the best comprehensive performance under the constraints.

6. The intelligent driving method for a liquid crystal display screen according to claim 1, characterized in that: The step S5 comprises: Mapping the optimal driving parameters into pixel-level driving signals; The voltage change of the TFT liquid crystal pixel unit is controlled by the voltage driving module; Synchronously collect liquid crystal response data during each driving cycle, including response delay, brightness change and pixel stability; The response data is passed to the subsequent feedback module for status update.

7. The intelligent driving method for a liquid crystal display according to claim 1, characterized in that: The step S6 comprises: Based on the collected response data, the current health status of each pixel is modeled and corrected; Updated LCD health status map to reflect pixel fatigue, electromigration and response deviation; Retrain or fine-tune the neural network using the updated health map as additional input; Adjust visual importance mapping parameters and physical modeling structure; Realize closed-loop optimization and dynamic adaptation of the drive system.

8. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The physical information augmented neural network adopts a lightweight and variable architecture, supporting the division of model depth and complexity by region, including: Using deep network calculations for areas of high visual importance; Use pruned or quantized small networks for areas of low visual attention; The model inference process uses parallel processing of heterogeneous computing units; Improve the inference efficiency of repeated areas through caching mechanism.

9. The intelligent driving method of a liquid crystal display according to claim 1, characterized in that: The physical response model is processed by combining historical voltage drive data with temperature response curves through the following processing steps: Construct a pixel-level dynamic model based on liquid crystal torsional torque and applied electric field; The actual response time of the fusion collection is calibrated and corrected; The modeling results are used to assist the neural network in predicting nonlinear responses; A mapping relationship is established between the response compensation result and the driving signal.

10. An intelligent driving system for a liquid crystal display screen, applied to the intelligent driving method for a liquid crystal display screen according to any one of claims 1 to 9, characterized in that: include: Multimodal information acquisition module, used to obtain environmental information, display content information and user behavior information, and construct a fusion feature vector; The neural network inference module is used to receive the fused feature vector and extract the spatial and temporal features, while fusing the physical modeling results to generate a visual importance map; The area division and refresh control module is used to divide the display area according to the visual importance mapping and content change rate, and calculate the refresh priority and refresh rate; A multi-objective optimization module, which is used to build an optimization function and determine the optimal driving parameters based on display quality, power consumption, and response speed; A driving execution module, configured to generate a driving signal according to the optimal driving parameters to control the display behavior of the liquid crystal pixel unit; The closed-loop feedback module is used to collect liquid crystal response data and update the liquid crystal health status map and neural network model parameters.

Images