An intelligent control system for liquid crystal display screen

Abstract

The present invention relates to an intelligent control system for a liquid crystal display screen, comprising a plurality of modules to optimize display effects and enhance user experience. In the system, a visual health management module adjusts the brightness, contrast and color temperature of the liquid crystal display screen based on the detected ambient light intensity and color temperature, and adjusts the refresh rate of the display screen based on the noise level to reduce visual fatigue; a user behavior recognition module detects the user's eye movement, posture, usage time and user identity through sensors to generate user behavior data; an adaptive content optimization module adjusts the display mode of the displayed content according to the user behavior data; and a personalized setting storage module stores user personalized settings and quickly loads the corresponding settings when switching between different users. The system realizes intelligent and personalized control of the liquid crystal display screen by comprehensively considering environmental factors and user behavior, thereby enhancing the user's visual comfort and user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid crystal display screens, and in particular to an intelligent control system of a liquid crystal display screen. Background Art

[0002] LCD screen control systems usually include basic brightness and contrast adjustment functions. Some advanced systems are also equipped with ambient light sensors to automatically adjust the screen brightness according to the surrounding lighting conditions.

[0003] However, the main problem with the existing technology is that it cannot dynamically adjust display parameters by comprehensively considering multiple factors such as ambient light, color temperature and noise level, and lacks the ability to detect and respond to user behavior in real time. In addition, the existing system cannot quickly load personalized settings when the user switches, resulting in a less than ideal user experience. These limitations make the existing LCD display control system insufficient in improving user visual comfort and user experience.

[0004] Therefore, there is an urgent need to develop a new intelligent control system for liquid crystal display screens. Summary of the invention

[0005] The present application provides an intelligent control system for a liquid crystal display screen to enhance the user's visual comfort and usage experience.

[0006] The present application provides an intelligent control system for a liquid crystal display screen, comprising:

[0007] Intelligent environment sensing module, used to detect the ambient light intensity, color temperature and surrounding noise level of the LCD screen in real time;

[0008] a visual health management module, connected to the intelligent environment perception module, for adjusting the brightness, contrast and color temperature of the LCD screen based on the detected ambient light intensity and color temperature, and adjusting the refresh rate of the LCD screen based on the detected ambient noise level to reduce visual fatigue;

[0009] User behavior recognition module, which uses sensors to detect the user's eye movements, posture, usage duration, and user identity, and generates user behavior data based on the detection results;

[0010] An adaptive content optimization module, connected to the user behavior recognition module, for adjusting the display mode of the content displayed on the liquid crystal display screen according to the user behavior data, wherein the display mode includes font size, interface layout and color contrast;

[0011] A personalized setting storage module is connected to the visual health management module and the adaptive content optimization module, and is used to store user personalized settings and quickly load corresponding settings when switching between different users; wherein the user personalized settings include the brightness, contrast, color temperature, refresh rate of the LCD screen set by the user, and the display mode of the content displayed on the LCD screen.

[0012] Furthermore, the visual health management module adjusts the refresh rate of the LCD screen based on the following formula 1:

[0013]

[0014] Among them, R max is the maximum refresh rate; S represents the current noise level; S0 is the baseline noise level; κ is the adjustment coefficient used to control the refresh rate with the noise level; λ is the adjustment coefficient used to control the refresh rate with the fatigue index; F is the user's fatigue index, calculated according to the following formula 2:

[0015]

[0016] Among them, B i represents the ith behavior parameter, which includes the user's eye movement frequency, posture change frequency and usage time; μ B is the average value of all behavioral parameters; n is the number of behavioral parameters; T is the user's cumulative usage time; T0 is the benchmark usage time; α is the adjustment coefficient, which is used to control the impact of usage time on fatigue index.

[0017] Furthermore, the visual health management module adjusts the brightness L, contrast C and color temperature T of the LCD screen based on the following formulas 3-5:

[0018]

[0019] Among them, L base represents the basic brightness value; E represents the ambient light intensity; T represents the detected color temperature; N represents the surrounding noise level; β and γ are adjustment coefficients; C base represents the basic contrast value; L represents the brightness calculated according to formula 3; δ and ∈ are adjustment coefficients; T base represents the basic color temperature value; C represents the contrast calculated according to Formula 4; ζ and η are adjustment coefficients.

[0020] Furthermore, the adaptive content optimization module uses a trained hybrid neural network model to obtain a display mode of the content displayed on the LCD screen based on the user behavior data; the hybrid neural network model includes a self-attention transformation layer, a graph convolutional network layer, a generative adversarial network optimization layer and a variational autoencoder layer; the self-attention transformation layer uses an attention mechanism to process the user behavior data matrix to obtain a first feature matrix processed by the attention mechanism; the graph convolutional network layer uses a graph convolutional neural network to process the first feature matrix to obtain a second feature matrix processed by the graph convolutional neural network; the generative adversarial network optimization layer includes a generator and a discriminator; the generator is used to generate a third feature matrix based on the second feature matrix, and the discriminator is used to adversarially train the generator; the variational autoencoder layer processes the third feature matrix through an encoder and a decoder to generate an optimized feature matrix, and the optimized feature matrix is ​​used to adjust the display content.

[0021] Furthermore, the self-attention transformation layer implements the attention mechanism through the following steps:

[0022] Perform linear transformation on the user behavior data matrix to obtain the query matrix, key matrix and value matrix;

[0023] Calculate the dot product of the query matrix and the key matrix and divide it by the scaling factor, and apply the softmax function to obtain the attention weight matrix;

[0024] The attention weight matrix is ​​multiplied by the value matrix to generate a first feature matrix processed by the attention mechanism.

[0025] Furthermore, the graph convolution network layer implements the graph convolution operation through the following steps:

[0026] The first feature matrix is ​​normalized using the adjacency matrix of the graph;

[0027] The processed matrix is ​​multiplied with the first feature matrix and combined with the trainable weight matrix, and processed by a nonlinear activation function to generate a second feature matrix processed by the graph convolutional neural network.

[0028] Furthermore, the variational autoencoder layer is implemented by the following steps:

[0029] The encoder is used to map the third feature matrix to the latent space to generate the mean vector and standard deviation vector of the latent variables;

[0030] The latent vector is obtained by reparameterization technique,

[0031] The decoder is used to map the latent vector back to the feature space, generating an optimized feature matrix for display content adjustment.

[0032] Furthermore, the user behavior recognition module includes an eye movement sensor for detecting the user's eye movement.

[0033] Furthermore, the personalized settings storage module is configured to synchronize user personalized settings with a cloud server via wireless communication.

[0034] The beneficial effects of the present application mainly include: (1) The system detects the ambient light intensity, color temperature and surrounding noise level in real time through the intelligent environment perception module, and can automatically adjust the brightness, contrast and color temperature of the display screen according to environmental changes to ensure that the display effect is always optimal. (2) The visual health management module uses the detected environmental data to dynamically adjust the brightness, contrast, color temperature and refresh rate of the display screen to reduce user visual fatigue, especially in the case of long-term use, effectively protecting the user's eyesight. (3) The user behavior recognition module generates user behavior data by detecting the user's eye movement, posture, usage time and identity through sensors, and can accurately understand the user's usage habits and needs, so as to make intelligent adjustments. (4) The adaptive content optimization module adjusts the display mode of the displayed content according to the user behavior data, including font size, interface layout and color contrast, so that the display effect is more in line with the user's personalized needs and improves the user experience. (5) The personalized setting storage module stores the user's personalized settings and the optimization parameters automatically generated by the system, and quickly loads the corresponding settings when switching between different users, ensuring that each user can enjoy the best personalized display effect when using it. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of an intelligent control system of a liquid crystal display screen provided in the first embodiment of the present application. DETAILED DESCRIPTION

[0036] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0037] The first embodiment of the present application provides an intelligent control system for a liquid crystal display screen. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 The first embodiment of the present application provides an intelligent control system for a liquid crystal display screen for detailed description.

[0038] The intelligent control system includes an intelligent environment perception module 101 , a visual health management module 102 , a user behavior recognition module 103 , an adaptive content optimization module 104 and a personalized setting storage module 105 .

[0039] The intelligent environment sensing module 101 is used to detect the ambient light intensity, color temperature and surrounding noise level of the liquid crystal display screen in real time.

[0040] The intelligent environment perception module 101 is an important part of the intelligent control system, responsible for real-time detection and analysis of multiple parameters of the environment around the LCD screen to provide accurate environmental data for use by other modules of the system. The intelligent environment perception module 101 mainly includes a light sensor, a color temperature sensor, a noise sensor, and a data processing unit.

[0041] First, the light sensor is installed in the border area of ​​the LCD screen, which can detect the ambient light intensity around the screen in real time. The sensor uses a highly sensitive photodiode or photoresistor, which can accurately sense changes in light and convert the detected light signal into an electrical signal through a built-in amplifier circuit. After the electrical signal is processed by the analog-to-digital converter (ADC), a digital signal is generated, which represents the current ambient light intensity.

[0042] Secondly, the color temperature sensor is used to detect the color temperature of the ambient light. The sensor usually uses a three-channel (red, green, and blue) optical filter and a corresponding photodetector to measure the intensity of the red, green, and blue components of the ambient light. By comparing the proportional relationship of these three components, the data processing unit can calculate the color temperature of the current ambient light. This data is particularly important for adjusting the color temperature parameters of the LCD screen to ensure the accuracy and consistency of the display effect under different ambient light conditions.

[0043] In addition, a noise sensor is installed near the display screen to detect the noise level of the surrounding environment. The sensor uses a highly sensitive microphone that can capture sound waves in the environment and convert the sound wave signals into electrical signals. Similar to the light sensor, the output signal of the noise sensor is also processed by an amplifier circuit and an analog-to-digital converter to generate a digital signal representing the noise intensity. By monitoring the ambient noise level in real time, the system can appropriately adjust the refresh rate of the LCD screen in a high-noise environment to reduce visual fatigue.

[0044] All the data from these sensors are transmitted to the data processing unit, which integrates a variety of signal processing algorithms and data fusion technologies to comprehensively analyze data such as light intensity, color temperature, and noise level. The data processing unit converts these environmental parameters into actionable control signals through a preset algorithm model and transmits them to the visual health management module 102 to dynamically adjust the display parameters of the LCD screen.

[0045] Through the precise detection and data processing of the intelligent environment sensing module 101, the LCD display intelligent control system can adapt to the changes in the external environment in real time, thereby providing a more comfortable and healthy viewing experience.

[0046] The visual health management module 102 is connected to the intelligent environment perception module and is used to adjust the brightness, contrast and color temperature of the LCD screen based on the detected ambient light intensity and color temperature, and adjust the refresh rate of the LCD screen based on the detected surrounding noise level to reduce visual fatigue.

[0047] The visual health management module 102 is a key component in the intelligent control system, which is designed to dynamically adjust various display parameters of the LCD screen based on the environmental data provided by the intelligent environmental perception module 101 to reduce the user's visual fatigue and improve viewing comfort.

[0048] The module first receives real-time environmental data from the intelligent environmental perception module 101, which includes ambient light intensity, color temperature, and surrounding noise level. The visual health management module 102 has multiple processing units built in, each of which is responsible for adjusting specific display parameters. Specifically, the module contains a brightness adjustment unit, a contrast adjustment unit, and a color temperature adjustment unit, which work together to make fine adjustments based on environmental data.

[0049] The brightness adjustment unit is mainly responsible for adjusting the brightness of the LCD screen according to the ambient light intensity. The brightness adjustment unit has an automatic adjustment algorithm. When the ambient light intensity is strong, the brightness adjustment unit will increase the brightness of the display screen to ensure that the displayed content is still clearly visible under strong light; when the ambient light intensity is weak, the brightness adjustment unit will reduce the brightness of the display screen to reduce eye fatigue and discomfort.

[0050] The contrast adjustment unit adjusts the contrast of the LCD screen according to the ambient light intensity and color temperature. The core of this unit is a contrast optimization algorithm, which can analyze the intensity of ambient light and color temperature changes in real time, thereby dynamically adjusting the contrast of the display screen to ensure that the displayed content is clearly layered and the colors are true.

[0051] The color temperature adjustment unit adjusts the color temperature of the LCD screen according to the detected ambient color temperature. The color temperature adjustment unit has a built-in color temperature adjustment algorithm, which can automatically adjust the color temperature of the display screen according to the color temperature change of the ambient light to adapt to different ambient lighting conditions. For example, when the color temperature of the ambient light is high (bluish), the color temperature adjustment unit will lower the color temperature of the display screen to make the screen display softer and reduce the stimulation of blue light to the eyes; conversely, when the color temperature of the ambient light is low (yellowish), the color temperature adjustment unit will increase the color temperature of the display screen to enhance the clarity and visual comfort of the display effect.

[0052] In addition, the visual health management module 102 also includes a refresh rate adjustment unit, which dynamically adjusts the refresh rate of the LCD screen based on the ambient noise level detected by the intelligent environment perception module 101. When the noise level is high, the refresh rate adjustment unit will appropriately reduce the refresh rate of the display screen to reduce the flickering of the display screen, thereby alleviating visual fatigue caused by long-term viewing; while in a quiet environment with a low noise level, the refresh rate adjustment unit can increase the refresh rate to improve the smoothness and clarity of dynamic images.

[0053] In summary, the visual health management module 102 automatically adjusts the brightness, contrast, color temperature and refresh rate of the LCD screen by comprehensively analyzing the ambient light intensity, color temperature and noise level, ensuring that users can enjoy the best viewing experience in different environments and effectively reducing visual fatigue caused by prolonged screen use.

[0054] Furthermore, the visual health management module adjusts the refresh rate of the LCD screen based on the following formula 1:

[0055]

[0056] Among them, R max is the maximum refresh rate; S represents the current noise level; S0 is the baseline noise level; k is the adjustment coefficient used to control the refresh rate with the noise level; λ is the adjustment coefficient used to control the refresh rate with the fatigue index; F is the user's fatigue index, calculated according to the following formula 2:

[0057]

[0058] Among them, B i represents the ith behavior parameter, which includes the user's eye movement frequency, posture change frequency and usage time; μ B is the average value of all behavioral parameters; n is the number of behavioral parameters; T is the user's cumulative usage time; T0 is the benchmark usage time; α is the adjustment coefficient, which is used to control the impact of usage time on fatigue index.

[0059] The visual health management module adjusts the refresh rate of the LCD display based on Formula 1:

[0060]

[0061] In formula 1, the detailed description of each parameter is as follows:

[0062] R max : Maximum refresh rate. This is the highest refresh rate that an LCD screen can achieve, usually determined by the display's hardware specifications.

[0063] S: Current noise level. This value is detected in real time by the noise sensor in the intelligent environment perception module and indicates the noise intensity in the current environment.

[0064] S0: Baseline noise level. This is a pre-set noise level for comparison with the current noise level. This value can be set based on the user's environmental noise habits or standard noise level.

[0065] κ: A factor that controls the refresh rate as the noise level changes. This factor determines how sensitive the noise level is to the refresh rate adjustment. A larger κ value makes the refresh rate more sensitive to noise level changes.

[0066] λ: A coefficient used to control the refresh rate as the fatigue index changes. This coefficient determines the sensitivity of the user's fatigue index to the refresh rate adjustment. A larger λ value will make the refresh rate more sensitive to changes in the fatigue index.

[0067] F: User fatigue index, calculated by formula 2:

[0068]

[0069] In formula 2, the detailed description of each parameter is as follows:

[0070] B i : The i-th behavior parameter. The behavior parameters include the user’s eye movement frequency, posture change frequency, and usage time. These data are obtained by real-time monitoring by sensors in the user behavior recognition module.

[0071] User's eye movement frequency: The eye movement sensor detects the number of times the user's eye movements occur.

[0072] Posture change frequency: The frequency of changes in the user's body or head posture detected by the posture sensor.

[0073] Usage time: The total time the user uses the LCD screen is recorded through the time detection sensor.

[0074] μ B: The average value of all behavioral parameters. It is calculated by adding up the values ​​of all behavioral parameters and then dividing by the number of behavioral parameters n.

[0075]

[0076] n: Number of behavioral parameters. This is the total number of behavioral parameters used to calculate the fatigue index.

[0077] T: User's cumulative usage time. This value indicates the total usage time since the user started using the LCD screen, which can be accumulated and calculated by the time detection sensor.

[0078] T0: Baseline usage time. This is a preset time constant used to standardize the accumulated usage time.

[0079] α: Adjustment coefficient, used to control the impact of usage time on fatigue index. A larger α value will increase the weight of usage time in the fatigue index calculation, making the usage time have a greater impact on the fatigue index.

[0080] Through the above formula and parameters, the visual health management module can dynamically adjust the refresh rate of the LCD screen according to the current noise level and the user's fatigue index. The first part of formula 1 The second part (1-λ·F) in Formula 1 represents the effect of the user fatigue index on the refresh rate. When the user fatigue index is high, the refresh rate will also be reduced to further reduce visual fatigue.

[0081] Furthermore, the visual health management module adjusts the brightness L, contrast C and color temperature T of the LCD screen based on the following formulas 3-5:

[0082]

[0083]

[0084] Among them, L base represents the basic brightness value; E represents the ambient light intensity; T represents the detected color temperature; N represents the surrounding noise level; β and γ are adjustment coefficients; C base represents the basic contrast value; L represents the brightness calculated according to formula 3; δ and ∈ are adjustment coefficients; T base represents the basic color temperature value; C represents the contrast calculated according to Formula 4; ζ and η are adjustment coefficients.

[0085] For each of these formulas, the detailed description is as follows:

[0086] 1. Formula (3): Calculation of brightness L

[0087] L base : Basic brightness value. This is the brightness of the LCD screen under standard conditions, usually preset by the device manufacturer.

[0088] E: Ambient light intensity. It is obtained through real-time detection by the light sensor in the intelligent environment perception module, indicating the light intensity in the current environment.

[0089] T: Detected color temperature. Detected by the color temperature sensor in the intelligent environment perception module, indicating the color temperature of the current ambient light.

[0090] β: Adjustment coefficient, used to control the effect of color temperature on brightness adjustment. This coefficient determines the sensitivity of color temperature changes to brightness adjustment.

[0091] γ: Adjustment coefficient, used to control the effect of noise level on brightness adjustment. This coefficient determines the sensitivity of noise level changes to brightness adjustment.

[0092] N: Surrounding noise level. It is obtained through real-time detection by the noise sensor in the intelligent environment perception module, indicating the noise intensity in the current environment.

[0093] The first part of formula (3)

[0094]

[0095] It reflects the influence of ambient light intensity E and color temperature T on brightness L. The higher the light intensity, the greater the brightness adjustment; the higher the color temperature, the smaller the influence of light on brightness.

[0096] The second part of formula (3)

[0097]

[0098] It reflects the influence of noise level N on brightness L. The higher the noise level, the smaller the amplitude of brightness adjustment to reduce the interference of noise on user vision.

[0099] 2. Formula (4): Calculation of contrast C

[0100] C base : Basic contrast value. This is the contrast of the LCD screen under standard conditions, usually preset by the device manufacturer.

[0101] E: Ambient light intensity. Same as in formula (3), obtained through light sensor detection.

[0102] L: The brightness calculated according to formula (3). It indicates the currently adjusted brightness of the LCD screen.

[0103] δ: Adjustment coefficient, used to control the effect of brightness on contrast adjustment. This coefficient determines the sensitivity of brightness changes to contrast adjustment.

[0104] ∈: Adjustment coefficient, used to control the effect of noise level on contrast adjustment. This coefficient determines the sensitivity of contrast adjustment to changes in noise level.

[0105] N: ambient noise level. Same as in formula (3), obtained by detecting with a noise sensor.

[0106] The first part of formula (4)

[0107]

[0108] It reflects the influence of ambient light intensity E and brightness L on contrast C. The higher the light intensity, the greater the contrast adjustment; the higher the brightness, the smaller the influence of light on contrast.

[0109] The second part of formula (4)

[0110]

[0111] It reflects the effect of noise level N on contrast C. The higher the noise level, the smaller the contrast adjustment amplitude is, so as to reduce the interference of noise on user vision.

[0112] 3. Formula (5): Calculation of color temperature T

[0113] T base : Basic color temperature value. This is the color temperature of the LCD screen under standard conditions, usually preset by the device manufacturer.

[0114] E: Ambient light intensity. Same as in formulas (3) and (4), obtained through light sensor detection.

[0115] C: Contrast calculated according to formula (4). Indicates the currently adjusted contrast of the LCD screen.

[0116] ζ: Adjustment coefficient, used to control the effect of contrast on color temperature adjustment. This coefficient determines the sensitivity of contrast changes to color temperature adjustment.

[0117] η: Adjustment coefficient, used to control the effect of noise level on color temperature adjustment. This coefficient determines the sensitivity of noise level changes to color temperature adjustment.

[0118] N: ambient noise level. Same as in formulas (3) and (4), obtained by detecting with a noise sensor.

[0119] The first part of formula (5)

[0120]

[0121] It reflects the influence of ambient light intensity E and contrast C on color temperature T. The higher the light intensity, the greater the adjustment of color temperature; the higher the contrast, the smaller the influence of light on color temperature.

[0122] The second part of formula (5)

[0123]

[0124] It reflects the effect of noise level N on color temperature T. The higher the noise level, the smaller the color temperature adjustment is, so as to reduce the interference of noise on user vision.

[0125] By implementing these steps, the visual health management module is able to dynamically adjust the brightness, contrast, and color temperature of the LCD display based on the current ambient light intensity, color temperature, and noise level, thereby optimizing the user's visual experience.

[0126] The user behavior recognition module 103 is used for sensors to detect the user's eye movements, postures, usage duration and user identity, and to generate user behavior data based on the detection results.

[0127] The user behavior recognition module 103 is an important part of the intelligent control system, which is designed to detect and analyze the user's behavior data in real time through multiple sensors to provide personalized display adjustment and optimization services. The module includes eye movement sensors, posture sensors and time detection sensors, and recognizes the user's behavior pattern and identity by integrating the data of these sensors.

[0128] First, the eye movement sensor is installed in the border area of ​​the LCD screen, which can capture the user's eye movement in real time. The sensor uses an infrared light source and a high-resolution camera to track the user's gaze point and eye movement trajectory by detecting the reflected light of the eyeball. By analyzing this data, the system can understand the user's gaze point distribution, gaze time, and frequent line of sight movements. This information is important for adjusting the layout of the screen display content and reducing user visual fatigue.

[0129] The posture sensor is used to detect the user's sitting posture and head posture. This sensor usually uses an accelerometer and a gyroscope to determine the user's current posture by measuring the user's body tilt angle and head movement. For example, when it is detected that the user has been sitting in a posture for a long time, the system can prompt the user to take a proper rest or adjust the sitting posture to prevent fatigue and discomfort. The data from the posture sensor can also be used to adjust the height and angle of the display to suit the user's optimal viewing angle.

[0130] The time detection sensor is used to record the user's usage time. By recording the time when the user starts using the LCD screen and the duration of continuous use, the system can monitor the user's eye usage time and issue a break reminder when the user uses the screen for a long time to prevent eye fatigue caused by excessive use of the screen. In addition, the time detection data can also help the system understand the user's usage habits, thereby providing different display modes and brightness adjustment strategies in different time periods.

[0131] The user behavior recognition module 103 not only obtains user behavior data through the above sensors, but also integrates a user identity recognition function. This function usually uses facial recognition technology to capture the user's facial image through a camera and compare it with the pre-stored user facial data to confirm the user's identity. The accuracy of identity recognition is guaranteed by the matching algorithm of facial feature points, which can quickly identify different users and switch to corresponding personalized settings.

[0132] Based on the data from the above sensors, the user behavior recognition module 103 generates detailed user behavior data, including eye movement trajectory, posture changes, usage time and user identity. These data are transmitted to the adaptive content optimization module 104, which is used to dynamically adjust the display content on the LCD screen to make it more in line with the user's personalized needs. For example, the system can adjust the font size and interface layout according to the user's eye movement data, adjust the display angle according to the posture data, and adjust the screen brightness and issue a rest reminder according to the usage time.

[0133] Through the comprehensive analysis and processing of the user behavior recognition module 103, the LCD display intelligent control system can achieve highly personalized display optimization and provide a more comfortable and healthy user experience.

[0134] Furthermore, the user behavior recognition module includes an eye movement sensor for detecting the user's eye movement.

[0135] The user behavior recognition module includes an eye movement sensor for detecting the user's eye movement. The eye movement sensor plays a vital role in the intelligent control system of the LCD display screen. It is designed to capture and analyze the user's eye movement data in real time, thereby optimizing the display content and user experience.

[0136] Eye tracking sensors are usually installed on the border of the LCD screen or other locations that do not affect the user's line of sight. These sensors detect and track the movement of the user's eyeballs through an infrared light source and a high-resolution camera. Specifically, the light emitted by the infrared light source shines on the user's eyes and is then reflected back to the camera by the eyeballs. After the camera captures the reflected light, it analyzes the position and movement trajectory of the reflected light spot through an image processing algorithm to determine the user's gaze point and eye movement pattern.

[0137] The data obtained by the eye movement sensor includes the user's gaze point, gaze time, eye movement direction and speed, etc. These data are transmitted to the user behavior recognition module for processing and analysis. By analyzing the user's gaze point and gaze time, the system can determine what content the user is interested in, and dynamically adjust the displayed content. For example, when the user stares at a certain area for a long time, the system can enlarge the font or increase the contrast in that area to improve reading comfort; when the user's line of sight moves frequently, the system can simplify the interface layout to make it easier for the user to find the required information.

[0138] In addition, the eye movement sensor can also help detect the user's fatigue state. When the user's eye movement slows down, the gaze point is unstable, or the user blinks frequently, the system can determine that the user may feel tired, and issue a rest reminder or automatically adjust the screen brightness and color temperature to reduce eye fatigue.

[0139] To ensure the accuracy and stability of the eye tracking sensor, the system uses a variety of calibration and compensation techniques. When used for the first time, the system will ask the user to calibrate, by guiding the user to look at a specific point on the screen to correct the measurement parameters of the eye tracking sensor. In addition, the system will continuously calibrate itself to compensate for measurement errors caused by changes in user position or changes in ambient light.

[0140] The eye movement sensor is used in combination with data from other sensors, such as gesture sensors and time detection sensors, to provide more comprehensive and accurate user behavior data. These comprehensive data are not only used to optimize display content, but also to provide support for other functional modules, such as adaptive content optimization module and visual health management module.

[0141] By implementing the user behavior recognition module and its eye movement sensor function, it is ensured that the liquid crystal display intelligent control system can detect the user's eye movement in real time and accurately, thereby providing the best user experience.

[0142] The adaptive content optimization module 104 is connected to the user behavior recognition module and is used to adjust the display mode of the content displayed on the liquid crystal display screen according to the user behavior data. The display mode includes font size, interface layout and color contrast.

[0143] The adaptive content optimization module 104 is one of the core components of the intelligent control system, which aims to dynamically adjust the display mode of the displayed content according to the user behavior data generated by the user behavior recognition module 103, thereby improving the user's viewing experience and visual comfort. This module is mainly responsible for adaptively adjusting the font size, interface layout and color contrast.

[0144] First, the adaptive content optimization module 104 receives real-time user behavior data from the user behavior recognition module 103, which includes the user's eye movement trajectory, posture, usage time and identity information. By analyzing this data, the adaptive content optimization module can accurately determine the user's current needs and preferences and make corresponding adjustments.

[0145] In terms of font size adjustment, when the user behavior recognition module 103 detects that the user's eye movement track is concentrated in a certain area and stays there for a long time, the adaptive content optimization module 104 can infer that the user may be reading the content in this area. In order to improve reading comfort, the system can automatically enlarge the font size of the area to ensure that the user can read the text clearly. In addition, if the user's eye movement track jumps frequently, showing the behavior of browsing the entire screen, the system will maintain or appropriately reduce the font size to adapt to the user's browsing mode.

[0146] Adjustment of the interface layout is another important function of the adaptive content optimization module 104. The module determines the user's viewing angle and distance based on the user's posture data. For example, when the user tilts his head at a large angle, the system may need to rearrange the interface elements to ensure that the user can clearly see all important information at a non-normal viewing angle. Adjustment of the interface layout also includes quickly loading the user's preferred interface layout when the user switches identities to ensure a personalized experience between different users.

[0147] The adaptive adjustment of color contrast is based on user behavior data and ambient light intensity information. When the user's usage time increases and the ambient light intensity is low, the adaptive content optimization module 104 will reduce eye fatigue by adjusting the color contrast. For example, increasing the contrast between the background and the text makes the text more prominent, thereby reducing the pressure on the user to read in a low-light environment. On the contrary, when the ambient light intensity is high, the system will reduce the contrast to prevent the discomfort caused by strong contrast to the eyes.

[0148] The adaptive content optimization module 104 not only adjusts the above display mode, but also continuously optimizes the display effect according to the user's usage habits and preferences. This module has a built-in machine learning algorithm, which can gradually improve the accuracy and personalization of the adjustment strategy by continuously learning the user's behavior data. In this way, as the usage time increases, the system will understand the user's preferences more and more, thereby providing a more intimate and intelligent display optimization service.

[0149] In summary, the adaptive content optimization module 104 ensures that each user can enjoy the best visual experience and personalized service when using the LCD screen by adaptively adjusting the font size, interface layout and color contrast, combined with the detailed data provided by the user behavior recognition module 103.

[0150] Furthermore, the adaptive content optimization module uses a trained hybrid neural network model to obtain a display mode of the content displayed on the LCD screen based on the user behavior data; the hybrid neural network model includes a self-attention transformation layer, a graph convolutional network layer, a generative adversarial network optimization layer and a variational autoencoder layer; the self-attention transformation layer uses an attention mechanism to process the user behavior data matrix to obtain a first feature matrix processed by the attention mechanism; the graph convolutional network layer uses a graph convolutional neural network to process the first feature matrix to obtain a second feature matrix processed by the graph convolutional neural network; the generative adversarial network optimization layer includes a generator and a discriminator; the generator is used to generate a third feature matrix based on the second feature matrix, and the discriminator is used to adversarially train the generator; the variational autoencoder layer processes the third feature matrix through an encoder and a decoder to generate an optimized feature matrix, and the optimized feature matrix is ​​used to adjust the display content.

[0151] The adaptive content optimization module first processes the user behavior data matrix through the self-attention transformation layer. The user behavior data matrix is ​​a multidimensional matrix that contains information such as the user's eye movements, posture, usage time, and identity. The self-attention transformation layer processes this data through the attention mechanism. The specific operations include: first, linearly transform the user behavior data matrix to generate the query matrix, key matrix, and value matrix. Then, calculate the dot product of the query matrix and the key matrix and divide it by the scaling factor to obtain the attention weight matrix. Next, apply the softmax function to normalize the attention weight matrix, and finally multiply the weight matrix with the value matrix to generate the first feature matrix processed by the attention mechanism.

[0152] Next, the first feature matrix is ​​input to the graph convolutional network layer. The graph convolutional network layer uses a graph convolutional neural network to process the feature matrix. This process includes standardizing the first feature matrix using the adjacency matrix of the graph, then multiplying the standardized matrix with the first feature matrix, combining it with the trainable weight matrix, and finally processing it through a nonlinear activation function to generate a second feature matrix processed by the graph convolutional neural network. In this way, the graph structure information and complex relationships in the user behavior data can be fully captured.

[0153] The optimization layer of the generative adversarial network consists of two parts: the generator and the discriminator. The generator generates the third feature matrix based on the second feature matrix. The specific process is: the generator maps the second feature matrix to a higher-dimensional feature space through a multi-layer perceptron to generate the third feature matrix. The discriminator compares the third feature matrix generated by the generator with the real feature matrix, and continuously optimizes the generator through adversarial training to make the feature matrix it generates more realistic and representative. The discriminator uses the adversarial loss function to evaluate the performance of the generator and provide feedback to improve the generator.

[0154] Finally, the third feature matrix is ​​input to the variational autoencoder layer for processing. The variational autoencoder layer first maps the third feature matrix to the latent space through the encoder to generate the mean vector and standard deviation vector of the latent variables. Then, the latent vector is obtained through the reparameterization technique. This process ensures that the generated latent vector can effectively represent the high-dimensional information of the third feature matrix. The decoder maps the latent vector back to the feature space to generate an optimized feature matrix. The optimized feature matrix is ​​used to finally adjust the display mode of the displayed content, including font size, interface layout, and color contrast, to provide the best user viewing experience.

[0155] Furthermore, the self-attention transformation layer implements the attention mechanism through the following steps:

[0156] Perform linear transformation on the user behavior data matrix to obtain the query matrix, key matrix and value matrix;

[0157] Calculate the dot product of the query matrix and the key matrix and divide it by the scaling factor, and apply the softmax function to obtain the attention weight matrix;

[0158] The attention weight matrix is ​​multiplied by the value matrix to generate a first feature matrix processed by the attention mechanism.

[0159] The self-attention transformation layer implements the attention mechanism through a series of specific steps. First, the user behavior data matrix is ​​linearly transformed. This step maps the original data to the query matrix, key matrix, and value matrix. These matrices are obtained through linear transformation. The specific operation is to perform matrix multiplication operation on the input data matrix and the pre-defined weight matrix to generate the query matrix, key matrix, and value matrix. This process can be expressed as: the user behavior data matrix X is respectively multiplied by the weight matrix W Q , W K and W V Multiply them together to get the query matrix Q, key matrix K and value matrix V.

[0160] Next, the dot product of the query matrix and the key matrix is ​​calculated. The dot product operation multiplies each vector of the query matrix Q and the key matrix K element by element and sums the results to generate an intermediate score matrix. In order to make the calculation more stable, the intermediate score matrix is ​​divided by a scaling factor, which is usually d k The square root of k is the dimension of the key matrix. The mathematical representation of this step is: the score matrix S is equal to

[0161] Then, the score matrix is ​​normalized by applying the softmax function. The softmax function converts the values ​​in the score matrix into a probability distribution, ensuring that the sum of all attention weights is equal to 1. The specific operation is to perform an exponential operation on each value in the score matrix and then normalize the result. This step ensures that each element in the attention weight matrix is ​​between 0 and 1, and the sum is 1, indicating the relevance of each key matrix vector to the query matrix vector.

[0162] Finally, the attention weight matrix is ​​multiplied by the value matrix to generate the first feature matrix after the attention mechanism. This step applies the weights in the attention weight matrix to the corresponding elements of the value matrix, and accumulates the weighted value matrix through matrix multiplication to generate the final feature matrix. Specifically, the attention weight matrix A is multiplied by the value matrix V to obtain the first feature matrix X'.

[0163] Through these detailed steps, the self-attention transformation layer can effectively process user behavior data, extract important features, and provide optimized input data for subsequent graph convolutional network layers.

[0164] Furthermore, the graph convolution network layer implements the graph convolution operation through the following steps:

[0165] The first feature matrix is ​​normalized using the adjacency matrix of the graph;

[0166] The processed matrix is ​​multiplied with the first feature matrix and combined with the trainable weight matrix, and processed by a nonlinear activation function to generate a second feature matrix processed by the graph convolutional neural network.

[0167] First, the first feature matrix is ​​normalized using the adjacency matrix of the graph. This step involves normalizing the adjacency matrix of the graph to ensure that the influence of each node is evenly distributed during the graph convolution process. Specifically, the adjacency matrix A is added to the identity matrix I to account for the influence of self-loops, and then the degree matrix D (whose diagonal elements are the degrees of each node) is calculated, and then the adjacency matrix is ​​normalized to obtain the standardized adjacency matrix The calculation method is

[0168]

[0169] This step ensures the numerical stability and feature smoothness of the graph convolution operation.

[0170] Next, multiply the standardized adjacency matrix by the first feature matrix. This step is to multiply the standardized adjacency matrix by Applied to the first feature matrix X', its purpose is to combine the features of each node with the features of its neighboring nodes to capture the relationship between nodes. Through matrix multiplication, the normalized adjacency matrix The node features are propagated and mixed to generate new node feature representations.

[0171] This processed matrix is ​​then combined with the trainable weight matrix. This is done by combining the matrix obtained in the previous step Multiply it with the weight matrix W, which is the parameter learned through the training process and is used to map the features to the new feature space. The mathematical expression of this operation is

[0172]

[0173] Among them, Z is the intermediate feature matrix, which contains the feature representation combined with the node neighbor information.

[0174] Finally, the intermediate feature matrix Z is subjected to nonlinear activation processing. Commonly used nonlinear activation functions include ReLU (Rectified Linear Unit), which introduces nonlinearity so that the model can capture more complex feature patterns. The specific operation is to apply the ReLU function to each element in the matrix Z to obtain the second feature matrix H after the graph convolutional neural network processing. The mathematical expression of this process is:

[0175]

[0176] Among them, σ represents a nonlinear activation function.

[0177] Through these detailed steps, the graph convolutional network layer can effectively combine the graph structure information with the node features to generate a second feature matrix containing rich relationship information, providing high-quality feature input for subsequent processing.

[0178] Furthermore, the variational autoencoder layer is implemented by the following steps:

[0179] The encoder is used to map the third feature matrix to the latent space to generate the mean vector and standard deviation vector of the latent variables;

[0180] The latent vector is obtained by reparameterization technique,

[0181] The decoder is used to map the latent vector back to the feature space, generating an optimized feature matrix for display content adjustment.

[0182] The variational autoencoder layer is implemented through a series of specific steps. First, the encoder is used to map the third feature matrix to the latent space, which involves compressing the input high-dimensional feature data into a low-dimensional latent representation. During this process, the encoder network processes the third feature matrix through a series of neural network layers, and finally generates two key outputs: the mean vector and standard deviation vector of the latent variables. Specifically, the encoder passes the input feature matrix H' through a series of linear transformations and nonlinear activation functions to generate a mean vector μ and a standard deviation vector σ, which together describe the distribution of the latent space.

[0183] Next, the latent vector is obtained by the reparameterization technique. The purpose of this step is to sample the latent variable while ensuring that the gradient can be transferred. The specific operation is to multiply the standard deviation vector σ with a noise vector ε sampled from a standard normal distribution, and then add the mean vector μ. The mathematical expression of this operation is:

[0184] z=μ+σ·ε

[0185] The reparameterization technique ensures the differentiability of the sampling process, making the entire network trainable via backpropagation.

[0186] Finally, the decoder is used to map the latent vector back to the feature space to generate an optimized feature matrix for display content adjustment. The decoder network converts the latent vector z back to a high-dimensional feature space through a series of reverse neural network layers. This step is to input the latent vector into the decoder, and after a series of linear transformations and nonlinear activation functions, gradually restore it to a high-dimensional feature representation similar to the input feature matrix, and finally output the optimized feature matrix V. This optimized feature matrix will be used to adjust the display content of the LCD screen, including parameters such as font size, interface layout, and color contrast, to provide the best user viewing experience.

[0187] Through the above detailed steps, the variational autoencoder layer can effectively reduce the dimensionality of complex high-dimensional input features, and generate a high-quality optimized feature matrix through the representation and reconstruction of the latent space.

[0188] The following is a detailed implementation code of a hybrid neural network model, which includes a self-attention transformer layer, a graph convolutional network layer, a generative adversarial network optimization layer, and a variational autoencoder layer.

[0189]

[0190]

[0191]

[0192]

[0193]

[0194] The code description is as follows:

[0195] 1. Self-attention transformation layer: First, the input user behavior data is linearly transformed to generate the query matrix, key matrix and value matrix. Then the dot product of the query matrix and the key matrix is ​​calculated and divided by the scaling factor, and the softmax function is applied to obtain the attention weight matrix. Finally, the attention weight matrix is ​​multiplied with the value matrix to generate the first feature matrix.

[0196] 2. Graph convolutional network layer: The first feature matrix is ​​standardized using the adjacency matrix of the graph, and then the standardized matrix is ​​multiplied by the first feature matrix and combined with the trainable weight matrix. After being processed by a nonlinear activation function, the second feature matrix is ​​generated.

[0197] 3. Generative adversarial network optimization layer: The generator generates the third feature matrix based on the second feature matrix. The discriminator is used to train the generator adversarially. The generator is continuously optimized by calculating the adversarial loss, making the feature matrix it generates more realistic and representative.

[0198] 4. Variational autoencoder layer: The encoder maps the third feature matrix to the latent space, generating the mean vector and standard deviation vector of the latent variables. The latent vector is obtained through the reparameterization technique, and then the decoder maps the latent vector back to the feature space to generate an optimized feature matrix for display content adjustment.

[0199] In the intelligent control system of LCD screens, the training of neural network models is a multi-stage, iterative process involving a series of steps from data collection to model deployment. This process ensures that the model can effectively learn and adapt to the dynamic changes of user behavior and environment to achieve optimized display content adjustment. The following are the specific steps of model training:

[0200] Step 1: Data collection and preprocessing

[0201] First, the system needs to collect user behavior data. The collected data is preprocessed, including data cleaning, denoising, and normalization, to ensure data quality and consistency, laying a solid foundation for subsequent model training.

[0202] Step 2: Model training

[0203] 1. Self-attention transformation layer training: In this stage, the system uses an adaptive weighting technique based on the attention mechanism to process user behavior data. First, the user behavior data matrix is ​​linearly transformed to generate a query matrix, a key matrix, and a value matrix. Then, the dot product of the query matrix and the key matrix is ​​calculated and divided by the scaling factor, and the softmax function is applied to obtain the attention weight matrix. Finally, the attention weight matrix is ​​multiplied with the value matrix to generate the first feature matrix after the attention mechanism.

[0204] 2. Graph convolutional network layer training: In this stage, the system uses the graph convolutional neural network to process the first feature matrix. The specific operation is to use the adjacency matrix of the graph to standardize the first feature matrix, then multiply the standardized matrix with the first feature matrix, and combine it with the trainable weight matrix, and process it through a nonlinear activation function to generate the second feature matrix after the graph convolutional neural network processing.

[0205] 3. Generative adversarial network optimization layer training: This stage includes the training of the generator and the discriminator. The generator generates the third feature matrix based on the second feature matrix. The discriminator compares the third feature matrix generated by the generator with the true feature matrix, evaluates the performance of the generator through the adversarial loss function, and provides feedback to optimize the generator. During the training process, the generator and the discriminator are continuously optimized through adversarial training, and the generated feature matrix is ​​more realistic and representative.

[0206] 4. Variational autoencoder layer training: In this stage, the system uses the encoder to map the third feature matrix to the latent space to generate the mean vector and standard deviation vector of the latent variables. Then, the latent vector is obtained through the reparameterization technique, and the decoder is used to map the latent vector back to the feature space to generate an optimized feature matrix. This optimized feature matrix is ​​used to adjust the display mode of the displayed content, including parameters such as font size, interface layout, and color contrast.

[0207] Step 3: Model evaluation and testing

[0208] After training is completed, the model needs to undergo rigorous evaluation and testing. Use cross-validation and various performance indicators (such as accuracy, response time, etc.) to verify the effectiveness and reliability of the model and ensure that its performance in actual applications meets system requirements.

[0209] Step 4: Model deployment and implementation

[0210] The models that have been evaluated and tested will be deployed in the intelligent control system of the LCD screen to process user behavior data and environmental data in real time, dynamically adjust the display content according to the results predicted by the model, and optimize the user's viewing experience.

[0211] Step 5: Continuous monitoring and iterative updates

[0212] After deployment, the system will continuously monitor the performance and actual application effects of the model, collect new operating data for iterative updates and optimization of the model, ensure that the model can adapt to environmental changes and changes in user behavior, and continue to provide efficient and accurate display content adjustment solutions.

[0213] Through these detailed steps, the neural network model in the LCD display intelligent control system can effectively learn and adapt to dynamic changes, thereby optimizing the displayed content and providing users with the best viewing experience.

[0214] The personalized setting storage module 105 is connected to the visual health management module and the adaptive content optimization module, and is used to store user personalized settings and quickly load corresponding settings when switching between different users; wherein the user personalized settings include the brightness, contrast, color temperature, refresh rate of the LCD screen set by the user, and the display mode of the content displayed on the LCD screen.

[0215] The personalized settings storage module 105 is a key component in the intelligent control system. It is designed to store and manage the personalized settings of users and quickly switch these settings between different users to provide a tailored display experience. This module connects the visual health management module 102 and the adaptive content optimization module 104 to ensure the best display effect in different environments and user behaviors.

[0216] The personalized setting storage module 105 uses a non-volatile memory, such as a flash memory or an EEPROM, to ensure that the user's personalized settings are not lost after the device is powered off or restarted. Whenever the user makes settings adjustments, such as modifying the screen brightness, contrast, color temperature, refresh rate, or adjusting the font size, interface layout, and color contrast of the displayed content, these settings will be recorded in real time and stored in the personalized setting storage module.

[0217] This module not only stores the settings manually entered by the user, but also records the optimization parameters automatically generated by the system based on the user behavior data. Through the data such as eye movement, posture, usage time and user identity collected by the user behavior recognition module 103, the adaptive content optimization module 104 and the visual health management module 102 will make corresponding adjustments according to the preset algorithm, and these dynamic adjustment parameters will also be stored in the personalized settings storage module. This design ensures that the system can quickly call and apply all settings suitable for the current user when the user identity is switched.

[0218] In order to achieve fast switching and efficient management, a user configuration database is designed inside the personalized setting storage module 105. The database uses indexing technology to quickly retrieve the corresponding configuration files according to the user identity. When the user identity is successfully identified, the system will immediately access the user configuration database in the personalized setting storage module 105 and load the personalized settings corresponding to the user. These settings include the user's preferred LCD screen brightness, contrast, color temperature, refresh rate, and font size, interface layout and color contrast of the displayed content. In this way, whether it is the optimization parameters automatically generated by the system or the settings manually adjusted by the user, they can be quickly applied when the user switches, ensuring that the display is always in the best state.

[0219] The personalized settings storage module 105 also includes an intelligent synchronization mechanism. When a user accesses the same LCD screen through different devices, the module can synchronize data with the cloud server to ensure that the user's personalized settings on different devices are consistent. This design is particularly suitable for multi-user environments, such as homes, offices or public places, ensuring that each user can enjoy a consistent personalized display experience on any device.

[0220] In addition, the personalized settings storage module 105 also has data backup and recovery functions. The system will regularly back up user settings to the cloud or local storage device to prevent data loss. When the device fails or the user changes the device, the previous personalized settings can be quickly restored to the new device through the recovery function, which greatly facilitates the user's use.

[0221] Through the above detailed design and functional description, the personalized setting storage module 105 can not only efficiently store and manage the user's personalized settings, but also quickly switch these settings between different users, thereby providing a personalized and comfortable display experience.

[0222] Furthermore, the personalized settings storage module is configured to synchronize user personalized settings with a cloud server via wireless communication.

[0223] The personalized settings storage module is configured to synchronize the user personalized settings with the cloud server via wireless communication. Specifically, the storage module can synchronize the personalized parameters set by the user on the LCD screen, such as brightness, contrast, color temperature, refresh rate, and information such as font size, interface layout and color contrast of the displayed content, with the remote cloud server in real time or regularly.

[0224] The wireless communication function enables the personalized settings storage module to establish a connection with the cloud server via Wi-Fi, Bluetooth, cellular network or other wireless communication technologies. This connection ensures that the user's personalized settings can be securely transmitted to the cloud and stored on the cloud server. In this way, when the user changes the device or uses it on different devices, the system can download the latest personalized settings from the cloud server and automatically apply them to the new device, ensuring that the user can enjoy a consistent personalized experience on various devices.

[0225] To achieve this synchronization function, the personalized settings storage module integrates a wireless communication module and a data processing unit. The wireless communication module is responsible for establishing a connection with the cloud server, while the data processing unit is responsible for collecting, packaging, and encrypting the user's personalized settings data. Whenever the user changes the settings on the LCD screen, the data processing unit records these changes in real time and sends the updated settings to the cloud server through the wireless communication module. After the cloud server receives the data, it will store and back it up to prevent data loss.

[0226] In addition, the personalized settings storage module also has conflict detection and synchronization strategies. When users make settings changes on multiple devices, the system can detect conflicts and resolve them according to the preset synchronization strategy. For example, the system can select the latest settings for synchronization, or prompt the user to select the settings version to be retained.

[0227] Through this design, the personalized settings storage module ensures that the user's settings are consistent across different devices and usage environments. Users do not need to manually reconfigure the settings of each device, which greatly improves the convenience of use. At the same time, cloud storage also provides security and durability for user settings. The data backup mechanism on the cloud server ensures that even if the device is lost or damaged, the user's personalized settings will not be lost and can be restored at any time.

[0228] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. An intelligent control system for a liquid crystal display screen, characterized in that: include: Intelligent environment sensing module, used to detect the ambient light intensity, color temperature and surrounding noise level of the LCD screen in real time; a visual health management module, connected to the intelligent environment perception module, for adjusting the brightness, contrast and color temperature of the LCD screen based on the detected ambient light intensity and color temperature, and adjusting the refresh rate of the LCD screen based on the detected ambient noise level to reduce visual fatigue; User behavior recognition module, which uses sensors to detect the user's eye movements, posture, usage duration, and user identity, and generates user behavior data based on the detection results; An adaptive content optimization module, connected to the user behavior recognition module, for adjusting the display mode of the content displayed on the liquid crystal display screen according to the user behavior data, wherein the display mode includes font size, interface layout and color contrast; A personalized setting storage module, connected to the visual health management module and the adaptive content optimization module, for storing user personalized settings and quickly loading corresponding settings when switching between different users; wherein the user personalized settings include the brightness, contrast, color temperature, refresh rate of the LCD screen set by the user, and the display mode of the content displayed on the LCD screen; The visual health management module adjusts the refresh rate of the LCD screen based on the following formula 1: Among them, R max is the maximum refresh rate; S represents the current noise level; S0 is the baseline noise level; κ is the adjustment coefficient used to control the refresh rate with the noise level; λ is the adjustment coefficient used to control the refresh rate with the fatigue index; F is the user's fatigue index, calculated according to the following formula 2: Among them, B i represents the ith behavior parameter, which includes the user's eye movement frequency, posture change frequency and usage time; μ B is the average value of all behavioral parameters; n is the number of behavioral parameters; T is the user's cumulative usage time; T0 is the benchmark usage time; α is the adjustment coefficient, which is used to control the impact of usage time on fatigue index.

2. The intelligent control system of the liquid crystal display screen according to claim 1, characterized in that: The visual health management module adjusts the brightness L, contrast C and color temperature T of the LCD screen based on the following formulas 3-5: Among them, L base represents the basic brightness value; E represents the ambient light intensity; T represents the detected color temperature; N represents the surrounding noise level; β and γ are adjustment coefficients; C base represents the basic contrast value; L represents the brightness calculated according to formula 3; δ and ∈ are adjustment coefficients; T base represents the basic color temperature value; C represents the contrast calculated according to Formula 4; ζ and η are adjustment coefficients.

3. The intelligent control system of a liquid crystal display screen according to claim 1, characterized in that: The adaptive content optimization module uses a trained hybrid neural network model to obtain the display mode of the content displayed on the liquid crystal display screen based on the user behavior data; the hybrid neural network model includes a self-attention transformation layer, a graph convolutional network layer, a generative adversarial network optimization layer and a variational autoencoder layer; the self-attention transformation layer uses an attention mechanism to process the user behavior data matrix to obtain a first feature matrix processed by the attention mechanism; the graph convolutional network layer uses a graph convolutional neural network to process the first feature matrix to obtain a second feature matrix processed by the graph convolutional neural network; the generative adversarial network optimization layer includes a generator and a discriminator; the generator is used to generate a third feature matrix based on the second feature matrix, and the discriminator is used to adversarially train the generator; the variational autoencoder layer processes the third feature matrix through an encoder and a decoder to generate an optimized feature matrix, and the optimized feature matrix is ​​used to adjust the display content.

4. The intelligent control system of the liquid crystal display screen according to claim 3, characterized in that: The self-attention transformation layer implements the attention mechanism through the following steps: Perform linear transformation on the user behavior data matrix to obtain the query matrix, key matrix and value matrix; Calculate the dot product of the query matrix and the key matrix and divide it by the scaling factor, and apply the softmax function to obtain the attention weight matrix; The attention weight matrix is ​​multiplied by the value matrix to generate a first feature matrix processed by the attention mechanism.

5. The intelligent control system of the liquid crystal display screen according to claim 3, characterized in that: The graph convolution network layer implements the graph convolution operation through the following steps: The first feature matrix is ​​normalized using the adjacency matrix of the graph; The processed matrix is ​​multiplied with the first feature matrix and combined with the trainable weight matrix, and processed by a nonlinear activation function to generate a second feature matrix processed by the graph convolutional neural network.

6. The intelligent control system of a liquid crystal display screen according to claim 3, characterized in that: The variational autoencoder layer is implemented by the following steps: The encoder is used to map the third feature matrix to the latent space to generate the mean vector and standard deviation vector of the latent variables; The latent vector is obtained by reparameterization technique, The decoder is used to map the latent vector back to the feature space, generating an optimized feature matrix for display content adjustment.

7. The intelligent control system of a liquid crystal display screen according to claim 1, characterized in that: The user behavior recognition module includes an eye movement sensor for detecting the user's eye movement.

8. The intelligent control system of a liquid crystal display screen according to claim 1, characterized in that: The personalized settings storage module is configured to synchronize user personalized settings with a cloud server via wireless communication.

Images