A double-sided liquid crystal display television system
By using information acquisition, processing, model training and adaptive control modules in the double-sided LCD display TV system, the display control coefficient is calculated, and the adaptive control of brightness, contrast, color temperature and refresh rate is achieved, and the double-sided screen is solved. The problem of not being able to independently regulate in different environments improves display quality and user experience, and reduces energy consumption.
Patent Information
- Application Number
- CN202411322431.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-23
AI Technical Summary
When facing different environments, double-sided LCD display TVs cannot independently control the display parameters of both screens, resulting in one side that may be too bright or too dark, affecting the overall display effect and user experience.
Using a system including information collection module, information processing module, model training module and adaptive control module, real-time data is obtained through sensors, preprocessing and feature extraction, establishing a time series model and machine learning model, calculating and displaying control coefficients, and realizing self-important Adapt to adjust brightness, contrast, color temperature and refresh rate.
It realizes real-time response to environmental changes, precise control of display parameters, improve display quality and user experience, reduce energy consumption, and ensure optimization and synchronization of display effects.
Smart Images

Figure CN118840982B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of liquid crystal television display, in particular to a double-sided liquid crystal display television system. Background Art
[0002] Double-sided LCD TV systems are widely used in public places, commercial advertisements and exhibitions due to their unique double-sided display characteristics. However, in actual applications, this system often faces a key problem, namely the synchronization of the two screens. In order to ensure the synchronization of the screen, traditional double-sided LCD TVs usually make uniform adjustments to the brightness, contrast, color temperature and other parameters of the two screens, which means that no matter how different the environments of the two screens are, the display parameters are the same. Although this method ensures the synchronization of the screen, it also brings significant disadvantages.
[0003] Since the two screens of a double-sided LCD TV are often in different environments, for example, one side may be exposed to direct sunlight while the other side is in the shadow, resulting in different requirements for display parameters. In the traditional display process, the display parameters are regulated synchronously, and the control adjustment results of the TV act on the two LCD screens, which cannot be independently regulated according to their respective environmental requirements. This limitation causes one screen to be too bright or too dark, while the other side fails to make full use of the ambient light, affecting the overall display effect and user experience. The inability to independently control the display parameters of the two screens makes it impossible for double-sided LCD TVs to respond flexibly to complex and changing environments and it is difficult to provide the best visual experience. Summary of the invention
[0004] In view of the deficiencies of the prior art, the present invention provides a double-sided liquid crystal display television system, which solves the problems mentioned in the background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a double-sided liquid crystal display television system, including an information acquisition module, an information processing module, a model training module and an adaptive control module;
[0006] The information acquisition module obtains the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data group;
[0007] The information processing module performs preprocessing and feature extraction on the real-time information collection data group, the preprocessing includes normalization processing and difference processing, and the feature extraction includes moving average extraction, moving standard deviation extraction and time lag feature extraction to form a processed information collection data group;
[0008] The model training module establishes a time series model and a machine learning model for training and analysis on the processed information collection data group to obtain: display control coefficient Tkxs;
[0009] The adaptive control module matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme, and specifically controls the current liquid crystal display according to the content of the current display surface control evaluation scheme.
[0010] Further, the information acquisition module includes an acquisition unit;
[0011] The acquisition unit acquires the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data set;
[0012] Among them, several sensors include ambient light sensors and ambient color temperature sensors; environmental information includes ambient light intensity information and ambient color temperature information; operating parameter information includes screen brightness information, screen contrast information, screen color temperature information and screen refresh rate information.
[0013] Further, the information processing module includes a processing unit and an extraction unit;
[0014] The processing unit performs preprocessing on the real-time information collection data group, including normalization processing to adjust the data to the same scale to eliminate the dimensional differences between different parameters and differential processing to perform differential processing on the normalized time series data to make the data more stable;
[0015] The extraction unit performs feature extraction on the pre-processed real-time information collection data group, including moving average extraction, moving standard deviation extraction and time lag feature extraction, to form a processed information collection data group.
[0016] Furthermore, the normalization process is performed by The formula is used for processing, where represents the processed data value, x represents the original data value, Represents the average value of the data set. Represents the standard deviation of the data in the data set; the difference processing is done by The formula is used for processing, where is the data value at time t, represents the data value at time t-1, Indicates time and time The difference data value.
[0017] Furthermore, the moving average is extracted by The formula extracts the moving average from the differenced data to smooth the time series data and reduce the impact of short-term fluctuations. represents the moving average of time t, n represents the window size, i represents the i-th window, Represents the data value at time ti;
[0018] Moving standard deviation is extracted by The formula extracts the moving standard deviation from the differenced data to measure the volatility of the time series data and evaluate the volatility and abnormal changes of the data. represents the moving standard deviation at time t, and x represents the original data value.
[0019] Furthermore, the model training module includes a modeling unit and a fitting unit;
[0020] The modeling unit extracts the processed time series data from the processed information collection data group. , obtain the order p of the autoregressive term and the order q of the moving average term through the autocorrelation function and partial autocorrelation function, and then use the least squares method or maximum likelihood estimation method to estimate the parameter autoregressive coefficients , moving average coefficient Estimation and acquisition of the constant term c, establishment of time series models including ARIMA models, capturing the time dependence and trend of data, and obtaining time prediction factors ; Extract input data from the processed information acquisition data set And the corresponding label data, establish a machine learning model including LSTM model to build the model architecture, define LSTM unit, including input layer, hidden layer and output layer, and calculate the hidden state of each time step , as the predicted value, use the training data to train the model, optimize the model parameters, and obtain the predicted output factor ;
[0021] The fitting unit predicts the time factor of the training results of the ARIMA model and the LSTM model and the predicted output factor Perform fitting and obtain the display regulation coefficient Tkxs.
[0022] Furthermore, the time prediction factor Obtained through the following calculation formula:
[0023] ;
[0024] In the formula, represents the time prediction factor, c represents the constant term, Represents the time prediction factor of the value at time tp for time t The impact of Represents the time prediction factor of the value at time tp for time t , p represents the order of the autoregressive term, specifically, p past time points are used for autoregression; Represents the error at time t-1 versus the data value at time t The impact of The time prediction factor representing the error at time tq for time t The impact of represents the error term at time t, It represents the error at time tq, q represents the order of the moving average term, and means that the errors of q past time points are used for moving average.
[0025] Furthermore, the display control coefficient Tkxs is obtained by the following calculation formula:
[0026] ;
[0027] Where, Tkxs represents the display regulation coefficient, represents a constant term, represents the time prediction factor, represents the predicted output factor, and Represent the time prediction factors and the predicted output factor The regression coefficient of .
[0028] Furthermore, the adaptive control module includes a matching unit and an execution unit;
[0029] The matching unit matches the preset relevant information with the required comparison value, including matching the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme;
[0030] The execution unit performs specific regulation on the current liquid crystal display according to the current display surface regulation evaluation scheme, including regulation of screen brightness regulation, screen contrast regulation, screen color temperature regulation and screen refresh rate regulation.
[0031] Furthermore, the current display surface control evaluation scheme is obtained by the following matching method:
[0032] Display control coefficient Tkxs<display evaluation control threshold T, obtain the non-control evaluation scheme of the liquid crystal display screen of the current display surface;
[0033] Display control coefficient Tkxs≥display evaluation control threshold T, obtain the LCD screen control evaluation plan of the current display surface, the control includes screen brightness control, screen contrast control, screen color temperature control and screen refresh rate control.
[0034] The present invention provides a double-sided liquid crystal display television system, which has the following beneficial effects:
[0035] (1) When the system is running, the information acquisition module obtains real-time information acquisition data groups, and the information processing module performs preprocessing and feature extraction to form a processed information acquisition data group. The model training module establishes a time series model and a machine learning model to train and analyze the processed information acquisition data group to obtain: display control coefficient Tkxs, and the adaptive control module matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation plan, and performs specific control on the current LCD display according to the plan content, so as to obtain and process the operation and environmental information of the display screen in real time, respond quickly to environmental changes, and improve display quality and user experience through intelligent control and precise control of display parameters, and avoid unnecessary energy consumption and reduce energy consumption.
[0036] (2) Accurately capture the time dependency of data and trend time prediction factors by building time series models and machine learning models and the predicted output factor The results of the two models are fitted through the fitting unit to calculate the display control coefficient Tkxs, thereby accurately reflecting the current display state.
[0037] (3) Intelligent control of the LCD screen is achieved through the adaptive control module, ensuring the optimization and synchronization of the display effect. The adaptive control module includes a matching unit and an execution unit. The matching unit matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the control evaluation plan of the current display surface. The execution unit specifically controls the brightness, contrast, color temperature and refresh rate of the screen according to the evaluation plan. This adaptive control method can not only respond to changes in the environment and operating conditions in real time, but also accurately adjust the display parameters, significantly improving the display quality and user experience of the LCD display, extending the service life of the equipment, and reducing energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 The present invention is a schematic diagram of a double-sided liquid crystal display television system block diagram process. DETAILED DESCRIPTION
[0039] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0040] Double-sided LCD TV systems are widely used in public places, commercial advertisements and exhibitions due to their unique double-sided display characteristics. However, in actual applications, this system often faces a key problem, namely the synchronization of the two screens. In order to ensure the synchronization of the screen, traditional double-sided LCD TVs usually make uniform adjustments to the brightness, contrast, color temperature and other parameters of the two screens, which means that no matter how different the environments of the two screens are, the display parameters are the same. Although this method ensures the synchronization of the screen, it also brings significant disadvantages.
[0041] Since the two screens of a double-sided LCD TV are often in different environments, for example, one side may be exposed to direct sunlight while the other side is in the shadow, resulting in different requirements for display parameters. In the traditional display process, the display parameters are regulated synchronously, and the control adjustment results of the TV act on the two LCD screens, which cannot be independently regulated according to their respective environmental requirements. This limitation causes one screen to be too bright or too dark, while the other side fails to make full use of the ambient light, affecting the overall display effect and user experience. The inability to independently control the display parameters of the two screens makes it impossible for double-sided LCD TVs to respond flexibly to complex and changing environments and it is difficult to provide the best visual experience.
[0042] Example 1
[0043] The present invention provides a double-sided liquid crystal display television system, see Figure 1 , including information collection module, information processing module, model training module and adaptive control module;
[0044] The information acquisition module obtains the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data group;
[0045] The information processing module performs preprocessing and feature extraction on the real-time information collection data group, the preprocessing includes normalization processing and difference processing, and the feature extraction includes moving average extraction, moving standard deviation extraction and time lag feature extraction to form a processed information collection data group;
[0046] The model training module establishes a time series model and a machine learning model for training and analysis on the processed information collection data group to obtain: display control coefficient Tkxs;
[0047] The adaptive control module matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme, and specifically controls the current liquid crystal display according to the content of the current display surface control evaluation scheme.
[0048] In this embodiment, a real-time information collection data group is acquired through the information acquisition module, and preprocessing and feature extraction are performed through the information processing module to form a processed information collection data group. The model training module is used to establish a time series model and a machine learning model for training and analysis on the processed information collection data group to obtain: a display control coefficient Tkxs, and the adaptive control module matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation plan, and the current LCD display is specifically controlled according to the content of the plan, so as to achieve real-time acquisition and processing of the operation and environmental information of the display screen, rapid response to environmental changes, and intelligent and precise control of display parameters to improve display quality and user experience, avoid unnecessary energy consumption, and reduce energy consumption.
[0049] Example 2
[0050] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the information acquisition module includes an acquisition unit;
[0051] The acquisition unit acquires the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data set;
[0052] Among them, the several sensors include an ambient light sensor and an ambient color temperature sensor; the environmental information includes ambient light intensity information and ambient color temperature information; the operating parameter information includes screen brightness information, screen contrast information, screen color temperature information and screen refresh rate information;
[0053] Among them, the screen brightness information and ambient light intensity information: the ambient light sensor detects the light intensity in the environment, converts the detected light intensity value into an electrical signal, and the electrical signal is transmitted to the information acquisition module through a data interface or wirelessly;
[0054] Ambient color temperature sensor and screen color temperature information: Through the deployment of several ambient color temperature sensors, the detected color temperature value is converted into an electrical signal, which is then transmitted to the information acquisition module through a data interface or wirelessly;
[0055] The information processing module includes a processing unit and an extraction unit;
[0056] The processing unit performs preprocessing on the real-time information collection data group, including normalization processing to adjust the data to the same scale to eliminate the dimensional differences between different parameters and differential processing to perform differential processing on the normalized time series data to make the data more stable;
[0057] Normalization: Different parameters may have different dimensions and ranges. Normalization adjusts the data to the same scale to facilitate comparison and comprehensive processing. In machine learning models, normalized data helps the gradient descent algorithm converge faster, improves training speed, and prevents certain features from dominating the model training process due to excessive dimensions, thereby improving the stability and performance of the model.
[0058] Differential processing: Time series data often contain trend components. Differential processing can effectively eliminate these trends and make the data stable. The data characteristics of stable time series are easier to model and predict. Differential processing improves the stability of data and is suitable for the application of time series models. It can also reduce the autocorrelation in the data, so that the model can more accurately capture the dynamic characteristics of the data.
[0059] The extraction unit performs feature extraction on the pre-processed real-time information collection data group, including moving average extraction, moving standard deviation extraction and time lag feature extraction, to form a processed information collection data group.
[0060] Normalization is done by The formula is used for processing, where represents the processed data value, x represents the original data value, Represents the average value of the data set. Represents the standard deviation of the data in the data set; the difference processing is done by The formula is used for processing, where is the data value at time t, represents the data value at time t-1, Indicates time and time The difference data value.
[0061] Moving average extraction by The formula extracts the moving average from the differenced data to smooth the time series data and reduce the impact of short-term fluctuations. represents the moving average of time t, n represents the window size, i represents the i-th window, Represents the data value at time ti;
[0062] Moving standard deviation is extracted by The formula extracts the moving standard deviation from the differenced data to measure the volatility of the time series data and evaluate the volatility and abnormal changes of the data. represents the moving standard deviation at time t, Represents the data value at time ti.
[0063] In this embodiment, the collected real-time information collection data group is optimized through normalization and difference processing, including adjusting the data to the same scale for comparison and processing, the normalized data helps the gradient descent algorithm converge faster, improves the stability and performance of the model, makes the time series data more stable, facilitates modeling and prediction, improves the model's ability to capture the dynamic characteristics of the data, and the stable data facilitates subsequent feature extraction, thereby improving the accuracy of feature extraction.
[0064] Example 3
[0065] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the model training module includes a modeling unit and a fitting unit;
[0066] The modeling unit extracts the processed time series data from the processed information collection data group. , obtain the order p of the autoregressive term and the order q of the moving average term through the autocorrelation function and partial autocorrelation function, and then use the least squares method or maximum likelihood estimation method to estimate the parameter autoregressive coefficients , moving average coefficient Estimation and acquisition of the constant term c, establishment of time series models including ARIMA models, capturing the time dependence and trend of data, and obtaining time prediction factors ; Extract input data from the processed information acquisition data set And the corresponding label data, establish a machine learning model including LSTM model to build the model architecture, define LSTM unit, including input layer, hidden layer and output layer, and calculate the hidden state of each time step , as the predicted value, use the training data to train the model, optimize the model parameters, and obtain the predicted output factor ;
[0067] The prediction output factor The calculation is obtained by the following steps:
[0068] Step 1: Forget gate calculation:
[0069] feet = σ ( Wf × [ ht − 1,xt ] + brother ) ;
[0070] In the formula, represents the output of the forget gate, Represents the Sigmoid activation function, which specifically indicates how much past memory is retained. represents the data input at time t, represents the hidden state at time t-1, represents the weight matrix, represents the bias vector;
[0071] Step 2: Input gate calculation:
[0072] dog = σ ( Yes × [ ht − 1,xt ] + one ) and <m> C < / m> t = longing ( Toilet × [ ht − 1,xt ] × BC ) ;
[0073] In the formula, represents the output of the input gate, Represents the Sigmoid activation function, which specifically indicates how much new input information is received. represents the weight matrix, represents the hidden state at time t-1, represents the bias vector, represents the data input at time t; Indicates the state of the candidate memory unit, specifically indicates the state of the candidate memory unit, represents the tanh activation function, represents the weight matrix, represents the bias vector;
[0074] Step 3: Memory unit update calculation:
[0075] ;
[0076] In the formula, represents the state of the memory cell at the current time t, represents the state of the memory unit at the previous time step t−1, specifically the memory at the previous moment, represents the output of the forget gate, represents the output of the input gate, Represents the state of candidate memory cells;
[0077] Step 4: Output gate calculation:
[0078] fire = σ ( Wow × [ ht − 1,xt ] + this ) and ;
[0079] In the formula, Represents the output of the output gate, specifically how much current memory is output, represents the Sigmoid activation function, represents the hidden state at time t-1, represents the data input at time t, represents the bias vector, Represents the predicted output factor, specifically the hidden state of the current time step t, which is the output of the LSTM unit. represents the tanh activation function, Represents the state of the memory unit at the current time t;
[0080] The fitting unit predicts the time factor of the training results of the ARIMA model and the LSTM model and the predicted output factor Perform fitting and obtain the display regulation coefficient Tkxs.
[0081] The time predictor Obtained through the following calculation formula:
[0082] ;
[0083] In the formula, represents the time prediction factor, c represents the constant term, Represents the time prediction factor of the value at time tp for time t The impact of Represents the time prediction factor of the value at time tp for time t , p represents the order of the autoregressive term, specifically, p past time points are used for autoregression; Represents the error at time t-1 versus the data value at time t The impact of The time prediction factor representing the error at time tq for time t The impact of represents the error term at time t, It represents the error at time tq, q represents the order of the moving average term, and means that the errors of q past time points are used for moving average.
[0084] The display control coefficient Tkxs is obtained by the following calculation formula:
[0085] ;
[0086] Where, Tkxs represents the display regulation coefficient, represents a constant term, represents the time prediction factor, represents the predicted output factor, and Represent the time prediction factors and the predicted output factor The regression coefficient of .
[0087] In this embodiment, a time series model and a machine learning model are established to accurately capture the time dependency of the data and the trend time prediction factor. and the predicted output factor The results of the two models are fitted through the fitting unit to calculate the display control coefficient Tkxs, thereby accurately reflecting the current display state.
[0088] Example 4
[0089] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the adaptive control module includes a matching unit and an execution unit;
[0090] The matching unit matches the preset relevant information with the required comparison value, including matching the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme;
[0091] The execution unit performs specific regulation on the current liquid crystal display according to the current display surface regulation evaluation scheme, including regulation of screen brightness regulation, screen contrast regulation, screen color temperature regulation and screen refresh rate regulation.
[0092] The current display surface control evaluation scheme is obtained by the following matching method:
[0093] Display control coefficient Tkxs<display evaluation control threshold T, obtain the non-control evaluation scheme of the liquid crystal display screen of the current display surface;
[0094] Display control coefficient Tkxs≥display evaluation control threshold T, obtain the LCD screen control evaluation plan of the current display surface, the control includes screen brightness control, screen contrast control, screen color temperature control and screen refresh rate control, and after the control, match the preset display evaluation control threshold T with the display control coefficient Tkxs again until the display control coefficient Tkxs<display evaluation control threshold T.
[0095] In this example, the intelligent control of the LCD screen is achieved through the adaptive control module, ensuring the optimization and synchronization of the display effect. The adaptive control module includes a matching unit and an execution unit. The matching unit matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the control evaluation scheme of the current display surface. When the display control coefficient Tkxs < the display evaluation control threshold T, it is determined that no control is required; when the display control coefficient Tkxs ≥ the display evaluation control threshold T, an evaluation scheme that needs to be controlled is generated. The execution unit specifically controls the brightness, contrast, color temperature and refresh rate of the screen according to the evaluation scheme. This adaptive control method can not only respond to changes in the environment and operating conditions in real time, but also accurately adjust the display parameters, significantly improving the display quality and user experience of the LCD display, extending the service life of the equipment, and reducing energy consumption.
[0096] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention.
Claims
1. A double-sided liquid crystal display television system, characterized in that: It includes information collection module, information processing module, model training module and adaptive control module; The information acquisition module obtains the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data group; the information acquisition module includes a collection unit; the collection unit obtains the operating parameter information and environmental information of the LCD screen through a number of sensors and reading the machine operating parameters, forming a real-time information acquisition data group; Among them, the several sensors include an ambient light sensor and an ambient color temperature sensor; the environmental information includes ambient light intensity information and ambient color temperature information; the operating parameter information includes screen brightness information, screen contrast information, screen color temperature information and screen refresh rate information; The information processing module performs preprocessing and feature extraction on the real-time information collection data group, the preprocessing includes normalization processing and difference processing, and the feature extraction includes moving average extraction, moving standard deviation extraction and time lag feature extraction to form a processed information collection data group; The model training module establishes a time series model and a machine learning model for training and analysis on the processed information collection data group to obtain: display control coefficient Tkxs; The model training module includes a modeling unit and a fitting unit; The modeling unit extracts the processed time series data from the processed information collection data group. , obtain the order p of the autoregressive term and the order q of the moving average term through the autocorrelation function and partial autocorrelation function, and then use the least squares method or maximum likelihood estimation method to estimate the parameter autoregressive coefficients , moving average coefficient Estimation and acquisition of the constant term c, establishment of time series models including ARIMA models, capturing the time dependence and trend of data, and obtaining time prediction factors ; Extract input data from the processed information acquisition data set And the corresponding label data, establish a machine learning model including LSTM model to build the model architecture, define LSTM unit, including input layer, hidden layer and output layer, and calculate the hidden state of each time step , as the predicted value, use the training data to train the model, optimize the model parameters, and obtain the predicted output factor ; The fitting unit predicts the time factor of the training results of the ARIMA model and the LSTM model and the predicted output factor Perform fitting and obtain the display control coefficient Tkxs; The time predictor Obtained through the following calculation formula: ; In the formula, represents the time prediction factor, c represents the constant term, Represents the time prediction factor of the value at time tp for time t The impact of Represents the time prediction factor of the value at time tp for time t , p represents the order of the autoregressive term, specifically, p past time points are used for autoregression; Represents the error at time t-1 versus the data value at time t The impact of The time prediction factor representing the error at time tq for time t The impact of represents the error term at time t, represents the error at time tq, q represents the order of the moving average term, indicating that the errors of q past time points are used for moving average; The display control coefficient Tkxs is obtained by the following calculation formula: ; Where, Tkxs represents the display regulation coefficient, represents a constant term, represents the time prediction factor, represents the predicted output factor, and Represent the time prediction factors and the predicted output factor The regression coefficient of The adaptive control module matches the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme, and specifically controls the current liquid crystal display according to the content of the current display surface control evaluation scheme; The adaptive control module includes a matching unit and an execution unit; The matching unit matches the preset relevant information with the required comparison value, including matching the preset display evaluation control threshold T with the display control coefficient Tkxs to obtain the current display surface control evaluation scheme; The execution unit specifically controls the current liquid crystal display according to the current display surface control evaluation scheme, including controlling the brightness control of the screen, the contrast control of the screen, the color temperature control of the screen, and the refresh rate control of the screen; The current display surface control evaluation scheme is obtained by the following matching method: Display control coefficient Tkxs<display evaluation control threshold T, obtain the non-control evaluation scheme of the liquid crystal display screen of the current display surface; Display control coefficient Tkxs≥display evaluation control threshold T, obtain the LCD screen control evaluation plan of the current display surface, the control includes screen brightness control, screen contrast control, screen color temperature control and screen refresh rate control.
2. A double-sided liquid crystal display television system according to claim 1, characterized in that: The information processing module includes a processing unit and an extraction unit; The processing unit performs preprocessing on the real-time information collection data group, including normalization processing to adjust the data to the same scale to eliminate the dimensional differences between different parameters and differential processing to perform differential processing on the normalized time series data to make the data more stable; The extraction unit performs feature extraction on the pre-processed real-time information collection data group, including moving average extraction, moving standard deviation extraction and time lag feature extraction, to form a processed information collection data group.
3. A double-sided liquid crystal display television system according to claim 2, characterized in that: Normalization is done by The formula is used for processing, where represents the processed data value, x represents the original data value, Represents the average value of the data set. Represents the standard deviation of the data in the data set; the difference processing is done by The formula is used for processing, where is the data value at time t, represents the data value at time t-1, Indicates time and time The difference data value.
4. A double-sided liquid crystal display television system according to claim 3, characterized in that: Moving average extraction by The formula extracts the moving average from the differenced data to smooth the time series data and reduce the impact of short-term fluctuations. represents the moving average of time t, n represents the window size, i represents the i-th window, Represents the data value at time ti; Moving standard deviation is extracted by The formula extracts the moving standard deviation from the differenced data to measure the volatility of the time series data and evaluate the volatility and abnormal changes of the data. represents the moving standard deviation at time t.
